From 36faf3ea447c18c942aafdba1080cb40136907f0 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Sun, 22 Nov 2020 17:49:36 -0500
Subject: [PATCH 01/34] wip: new recipe syntax

---
 pangeo_forge/recipe/recipe.py | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)
 create mode 100644 pangeo_forge/recipe/recipe.py

diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe/recipe.py
new file mode 100644
index 00000000..266b0506
--- /dev/null
+++ b/pangeo_forge/recipe/recipe.py
@@ -0,0 +1,28 @@
+"""
+A Pangeo Forge Recipe
+"""
+
+import xarray as xr
+import fsspec
+
+
+class Recipe:
+    datasets = []
+    keys = []
+
+    def filenames_for_chunk(**chunk_keys):
+        raise NotImplementedError
+
+    def open_file(self, fname):
+        # todo: caching
+        return fsspec.open(fname)
+
+    def open_chunk(self, filenames):
+        files_to_open = [self.open_file(f) for f in filenames]
+        ds_chunk = xr.open_mfdataset(filenames, **self.open_chunk_kwargs)
+        return ds_chunk
+
+    def store_chunk(self, store_target, **chunk_keys):
+        filenames = self.filenames_for_chunk(**chunk_keys)
+        ds_chunk = self.open_chunk(filenames)
+        ds_chunk.to_zarr(store_target, mode='a')

From 7ee78f22c5108fc10009b9bcf8f7bb668b041a4d Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Sat, 28 Nov 2020 21:32:01 -0500
Subject: [PATCH 02/34] messy wip

---
 pangeo_forge/__init__.py           |  10 +-
 pangeo_forge/pipelines/__init__.py |   5 -
 pangeo_forge/recipe/__init__.py    |   0
 pangeo_forge/recipe/pipeline.py    |   0
 pangeo_forge/recipe/recipe.py      | 154 +++++++++++++++++++++++++++--
 pangeo_forge/recipe/target.py      |  22 +++++
 pangeo_forge/utils.py              |  18 ----
 requirements.txt                   |   1 -
 tests/__init__.py                  |   0
 tests/fixtures.py                  |  88 +++++++++++++++++
 tests/test_recipe.py               |  48 +++++++++
 tests/test_utils.py                |  17 +++-
 12 files changed, 322 insertions(+), 41 deletions(-)
 create mode 100644 pangeo_forge/recipe/__init__.py
 create mode 100644 pangeo_forge/recipe/pipeline.py
 create mode 100644 pangeo_forge/recipe/target.py
 create mode 100644 tests/__init__.py
 create mode 100644 tests/fixtures.py
 create mode 100644 tests/test_recipe.py

diff --git a/pangeo_forge/__init__.py b/pangeo_forge/__init__.py
index a9528c6e..722da05d 100644
--- a/pangeo_forge/__init__.py
+++ b/pangeo_forge/__init__.py
@@ -1,6 +1,6 @@
 from pkg_resources import DistributionNotFound, get_distribution
 
-from pangeo_forge.pipelines import AbstractPipeline
+#from pangeo_forge.pipelines import AbstractPipeline
 
 try:
     __version__ = get_distribution(__name__).version
@@ -10,7 +10,7 @@
 
 del get_distribution, DistributionNotFound
 
-
-__all__ = [
-    "AbstractPipeline",
-]
+# 
+# __all__ = [
+#     "AbstractPipeline",
+# ]
diff --git a/pangeo_forge/pipelines/__init__.py b/pangeo_forge/pipelines/__init__.py
index 122c0873..e69de29b 100644
--- a/pangeo_forge/pipelines/__init__.py
+++ b/pangeo_forge/pipelines/__init__.py
@@ -1,5 +0,0 @@
-from .base import AbstractPipeline
-
-__all__ = [
-    "AbstractPipeline",
-]
diff --git a/pangeo_forge/recipe/__init__.py b/pangeo_forge/recipe/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/pangeo_forge/recipe/pipeline.py b/pangeo_forge/recipe/pipeline.py
new file mode 100644
index 00000000..e69de29b
diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe/recipe.py
index 266b0506..3cc86180 100644
--- a/pangeo_forge/recipe/recipe.py
+++ b/pangeo_forge/recipe/recipe.py
@@ -4,25 +4,161 @@
 
 import xarray as xr
 import fsspec
+from ..utils import chunked_iterable
+from .target import Target
 
+### How to manually execute a recipe: ###
+#
+#   r = PangeoForgeTarget()
+#   r = MyRecipe(**opts) # 1
+#   r.set_target(tmp_dir) # 2
+#   # manual execution of recipe
+#   r.prepare() # 3
+#   for input_key in r.iter_inputs():
+#       r.cache_input(input_key) # 4
+#   for chunk_key in r.iter_chunks():
+#       r.store_chunk(chunk_key) # 5
+#   r.finalize() # 6
+#
+###
 
-class Recipe:
-    datasets = []
-    keys = []
 
-    def filenames_for_chunk(**chunk_keys):
-        raise NotImplementedError
+# 1) Initialize the Recipe object
+# 2) Point the Recipe at its Target
+# 3) Initialize the recipe.
+#    Check if the target exists; if not, create it.
+#
+
+
+#@dataclass
+class DatasetRecipe():
+    target: Target
+
+
+class FSSpecInputOpenerMixin:
 
     def open_file(self, fname):
         # todo: caching
         return fsspec.open(fname)
 
-    def open_chunk(self, filenames):
+
+class XarrayChunkOpenerMixin:
+
+    def open_chunk(self, filenames, load=True):
         files_to_open = [self.open_file(f) for f in filenames]
         ds_chunk = xr.open_mfdataset(filenames, **self.open_chunk_kwargs)
+        if load:
+            ds_chunk.load()
         return ds_chunk
 
-    def store_chunk(self, store_target, **chunk_keys):
-        filenames = self.filenames_for_chunk(**chunk_keys)
+
+class ZarrWriterMixin:
+
+
+    def store_chunk(self, chunk_key):
+        filenames = self.filenames_for_chunk(chunk_key)
         ds_chunk = self.open_chunk(filenames)
-        ds_chunk.to_zarr(store_target, mode='a')
+        target_mapper = self.target.get_mapper()
+        write_region = self.get_write_region(chunk_key)
+        ds_chunk.to_zarr(target_mapper, region=write_region, **target_kwargs)
+
+
+    def open_target(self):
+        target_mapper = self.target.get_mapper()
+        return xr.open_zarr(target_mapper)
+
+
+
+
+
+class FileSequenceRecipe(DatasetRecipe):
+
+
+    def __init__(self, file_urls, sequence_dim, files_per_chunk=1, nitems_per_file=1, chunksize_within_file=None):
+        self.file_urls = file_urls
+
+        if chunksize_within_file:
+            assert file_per_chunk is None
+            assert nitems_per_file >= chunksize_within_file
+
+        self.files_per_chunk = files_per_chunk
+        self.nitems_per_file = nitems_per_file
+        self.sequence_dim = sequence_dim
+
+        # mapping between chunks and file names
+        self._chunks_files = {k: v for k, v in
+                              enumerate(chunked_iterable(file_urls, files_per_chunk))}
+
+
+    def filenames_for_chunk(self, chunk_key):
+        return self._chunks_files[chunk_key]
+
+
+    def nitems_for_chunk(self, chunk_key):
+        return self.nitems_per_file * len(self.filenames_for_chunk(chunk_key))
+
+
+    def region_for_chunk(self, chunk_key):
+        # return a dict suitable to pass to xr.to_zarr(region=...)
+        # specifies where in the overall array to put this chunk's data
+        stride = self.nitems_per_file * self.files_per_chunk
+        start = chunk_key * stride
+        return {
+            self.sequence_dim:
+                slice(start, start + self.nitems_for_chunk(chunk_key))
+        }
+
+
+    def sequence_dim(self):
+        # tells the total size of dataset along the sequence dimension
+        return {
+            self.sequence_dim:
+                sum([self.nitems_for_chunk(k) for k in self.iter_chunks()])
+        }
+
+
+    def sequence_chunks(self):
+        # chunking
+        return {self.sequence_dim: self.files_per_chunk * self.nitems_per_file}
+
+
+    def iter_chunks(self):
+        for k in self._chunks_files:
+            yield k
+
+    def prepare(self):
+
+        target_store = self.get_store_target()
+
+        try:
+            ds = self.open_target(target_store)
+
+        except IOError:
+            first_chunk_key = next(self.iter_chunks())
+            ds = self.open_chunk(first_chunk_key).chunk()
+            ds.to_zarr(path, compute=False, consolidated=False)
+
+        encoding = {v: ds[v].encoding for v in ds}
+
+        # now resize the sequence dim at the zarr level
+        sequence_axes = {v: ds[v].get_axis_num(self.sequence_dim)
+                         for v in ds
+                         if self.sequence_dim in ds[v].dims}
+        N = self.sequence_dim()
+
+        zgroup = zarr.open_group(target_store)
+
+        for v, axis in sequence_axes.items():
+            arr = zgroup[v]
+            shape = list(arr.shape)
+            shape[axis] = N
+            arr.resize(shape)
+
+
+class StandardSequentialRecipe(
+        FileSequenceRecipe,
+        FSSpecInputOpenerMixin,
+        XarrayChunkOpenerMixin,
+        ZarrWriterMixin,
+    ):
+    pass
diff --git a/pangeo_forge/recipe/target.py b/pangeo_forge/recipe/target.py
new file mode 100644
index 00000000..b7274d94
--- /dev/null
+++ b/pangeo_forge/recipe/target.py
@@ -0,0 +1,22 @@
+from dataclasses import dataclass
+import fsspec
+
+
+@dataclass
+class Target:
+    """Representation of a storage target for Pangeo Forge.
+    Attributes
+    ----------
+    url : FileSystemSpec.AbtractFileSystem
+        The filesystem we are writing to. Should be instantiated outside this
+        class.
+    path : str
+        The path where the target data will be saved.
+    """
+
+    fs: fsspec.AbstractFileSystem
+    path: str
+
+    def get_mapper(self):
+        # don't want to use this because we want to use a fancier Zarr FSStore
+        return self.fs.get_mapper(self.path)
diff --git a/pangeo_forge/utils.py b/pangeo_forge/utils.py
index f4b76625..c49b3014 100644
--- a/pangeo_forge/utils.py
+++ b/pangeo_forge/utils.py
@@ -1,8 +1,4 @@
 import itertools
-from typing import Any, List, Tuple
-
-from prefect import task
-
 
 # https://alexwlchan.net/2018/12/iterating-in-fixed-size-chunks/
 def chunked_iterable(iterable, size):
@@ -12,17 +8,3 @@ def chunked_iterable(iterable, size):
         if not chunk:
             break
         yield chunk
-
-
-@task
-def chunk(sources: List[Any], size: int) -> List[Tuple[Any, ...]]:
-    """
-    Prefect task to chunk a list of sources into batches.
-
-    Examples
-    --------
-    >>> import pangeo_forge.utils
-    >>> pangeo_forge.utils.chunk.run([1, 2, 3, 4, 5], size=2)
-    [(1, 2), (3, 4), (5,)]
-    """
-    return list(chunked_iterable(sources, size))
diff --git a/requirements.txt b/requirements.txt
index c6a2743f..dbf84c5f 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,7 +2,6 @@ setuptools
 click
 dask
 distributed
-prefect
 xarray
 zarr
 fsspec[http]
diff --git a/tests/__init__.py b/tests/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/fixtures.py b/tests/fixtures.py
new file mode 100644
index 00000000..0e54d899
--- /dev/null
+++ b/tests/fixtures.py
@@ -0,0 +1,88 @@
+import subprocess
+import time
+
+import fsspec
+import numpy as np
+import pandas as pd
+import pytest
+import xarray as xr
+
+
+
+# where to run the http server
+_PORT = "8080"
+_ADDRESS = "127.0.0.1"
+
+
+@pytest.fixture(scope="session")
+def daily_xarray_dataset():
+    """Return a synthetic random xarray dataset."""
+    np.random.seed(1)
+    nt, ny, nx = 10, 18, 36
+    time = pd.date_range(start="2010-01-01", periods=nt, freq="D")
+    lon = (np.arange(nx) + 0.5) * 360 / nx
+    lon_attrs = {"units": "degrees_east", "long_name": "longitude"}
+    lat = (np.arange(ny) + 0.5) * 180 / ny
+    lat_attrs = {"units": "degrees_north", "long_name": "latitude"}
+    foo = np.random.rand(nt, ny, nx)
+    foo_attrs = {"long_name": "Fantastic Foo"}
+    bar = np.random.rand(nt, ny, nx)
+    bar_attrs = {"long_name": "Beautiful Bar"}
+    dims = ("time", "lat", "lon")
+    ds = xr.Dataset(
+        {"foo": (dims, foo, foo_attrs), "bar": (dims, bar, bar_attrs)},
+        coords={
+            "time": ("time", time),
+            "lat": ("lat", lat, lat_attrs),
+            "lon": ("lon", lon, lon_attrs),
+        },
+        attrs={"conventions": "CF 1.6"},
+    )
+    return ds
+
+
+@pytest.fixture(scope="session", params=["D", "2D"])
+def netcdf_local_paths(daily_xarray_dataset, tmpdir_factory, request):
+    """Return a list of paths pointing to netcdf files."""
+    tmp_path = tmpdir_factory.mktemp("netcdf_data")
+    gb = daily_xarray_dataset.resample(time=request.param)
+    _, datasets = zip(*gb)
+    fnames = [f"{n:03d}.nc" for n in range(len(datasets))]
+    paths = [tmp_path.join(fname) for fname in fnames]
+    print(len(paths))
+    xr.save_mfdataset(datasets, [str(path) for path in paths])
+    return paths
+
+
+@pytest.fixture(scope="session")
+def netcdf_http_server(netcdf_local_paths):
+    first_path = netcdf_local_paths[0]
+    # assume that all files are in the same directory
+    basedir = first_path.dirpath()
+    print(basedir)
+    fnames = [path.basename for path in netcdf_local_paths]
+
+    # this feels very hacky
+    command_list = ["python", "-m", "http.server", _PORT, "--bind", _ADDRESS]
+    p = subprocess.Popen(command_list, cwd=basedir)
+    url = f"http://{_ADDRESS}:{_PORT}"
+    time.sleep(0.1)  # let the server start up
+    yield url, fnames
+    p.kill()
+
+
+# tests that our fixtures work
+
+
+def test_fixture_local_files(daily_xarray_dataset, netcdf_local_paths):
+    paths = [str(path) for path in netcdf_local_paths]
+    ds = xr.open_mfdataset(paths, combine="nested", concat_dim="time")
+    assert ds.identical(daily_xarray_dataset)
+
+
+def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
+    url, paths = netcdf_http_server
+    urls = ["/".join([url, str(path)]) for path in paths]
+    open_files = [fsspec.open(url).open() for url in urls]
+    ds = xr.open_mfdataset(open_files, combine="nested", concat_dim="time")
+    assert ds.identical(daily_xarray_dataset)
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
new file mode 100644
index 00000000..7ca003a6
--- /dev/null
+++ b/tests/test_recipe.py
@@ -0,0 +1,48 @@
+import pytest
+
+from pangeo_forge.recipe import recipe
+
+from .fixtures import daily_xarray_dataset, netcdf_local_paths
+
+dummy_fnames = ["a.nc", "b.nc", "c.nc"]
+@pytest.mark.parametrize(
+    "file_urls, files_per_chunk, expected_keys, expected_filenames",
+    [
+        (dummy_fnames, 1, [0, 1, 2], [("a.nc",), ("b.nc",), ("c.nc",)]),
+        (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",)])
+    ]
+)
+def test_file_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames):
+
+    r = recipe.FileSequenceRecipe(
+        file_urls=file_urls,
+        sequence_dim="time",
+        files_per_chunk=files_per_chunk
+    )
+
+    chunk_keys = list(r.iter_chunks())
+    assert chunk_keys == expected_keys
+
+    for k, expected in zip(r.iter_chunks(), expected_filenames):
+        fnames = r.filenames_for_chunk(k)
+        assert fnames == expected
+
+
+def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_dir):
+
+    r = StandardSequentialRecipe(
+        file_urls=netcdf_local_paths,
+        sequence_dim='time',
+    )
+
+    r.set_target(tmp_dir)
+    # manual execution of recipe
+    r.prepare()
+    for input_key in r.iter_inputs():
+        r.cache_input(input_key)
+    for chunk_key in r.iter_chunks():
+        r.store_chunk(chunk_key)
+    r.finalize()
+
+    ds_target = xr.open_dataset(tmp_dir)
+    assert ds_target.identical(daily_xarray_dataset)
diff --git a/tests/test_utils.py b/tests/test_utils.py
index 1b4b8d74..03e6be7f 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -1,6 +1,17 @@
+import pytest
+
 import pangeo_forge.utils
 
 
-def test_chunk():
-    result = pangeo_forge.utils.chunk.run([1, 2, 3], 2)
-    assert result == [(1, 2), (3,)]
+@pytest.mark.parametrize(
+    "iterable, size, expected",
+    [
+        ([1, 2, 3], 1, [(1,), (2,), (3,)]),
+        ([1, 2, 3], 2, [(1, 2), (3,)]),
+        ([1, 2, 3], 3, [(1, 2, 3,)]),
+        ([1, 2, 3], 4, [(1, 2, 3,)])
+    ]
+)
+def test_chunked_iterable(iterable, size, expected):
+    actual = list(pangeo_forge.utils.chunked_iterable(iterable, size))
+    assert actual == expected

From f980e8f8cff0f01b72ad4db1cd0ae36214b02c22 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Sat, 28 Nov 2020 22:08:05 -0500
Subject: [PATCH 03/34] made target fixture

---
 pangeo_forge/recipe/recipe.py | 34 +++++++++++++++++++---------------
 tests/fixtures.py             | 19 ++++++++++++++++++-
 tests/test_recipe.py          |  3 ++-
 3 files changed, 39 insertions(+), 17 deletions(-)

diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe/recipe.py
index 3cc86180..38c67d95 100644
--- a/pangeo_forge/recipe/recipe.py
+++ b/pangeo_forge/recipe/recipe.py
@@ -2,11 +2,16 @@
 A Pangeo Forge Recipe
 """
 
+from dataclasses import dataclass
+
 import xarray as xr
 import fsspec
 from ..utils import chunked_iterable
 from .target import Target
 
+from typing import Optional, Iterable
+
+
 ### How to manually execute a recipe: ###
 #
 #   r = PangeoForgeTarget()
@@ -27,10 +32,15 @@
 # 2) Point the Recipe at its Target
 # 3) Initialize the recipe.
 #    Check if the target exists; if not, create it.
-#
+# 4) cache the inputs to proximate storage (OPTIONAL)
+#    Some recipes won't need this (e.g. cloud to cloud)
+#    If so, iter_inputs is just an empty iterator
+# 5) Load each chunk from the inputs and store it in the target
+#    Might be coming from the cache or might be read directly.
+# 6)
 
 
-#@dataclass
+@dataclass
 class DatasetRecipe():
     target: Target
 
@@ -71,23 +81,17 @@ def open_target(self):
 
 
 
+@dataclass
 class FileSequenceRecipe(DatasetRecipe):
+    file_urls: Iterable[str]
+    sequence_dim: str
+    files_per_chunk: int = 1
+    nitems_per_file: int = 1
 
 
-    def __init__(self, file_urls, sequence_dim, files_per_chunk=1, nitems_per_file=1, chunksize_within_file=None):
-        self.file_urls = file_urls
-
-        if chunksize_within_file:
-            assert file_per_chunk is None
-            assert nitems_per_file >= chunksize_within_file
-
-        self.files_per_chunk = files_per_chunk
-        self.nitems_per_file = nitems_per_file
-        self.sequence_dim = sequence_dim
-
-        # mapping between chunks and file names
+    def __post_init__(self):
         self._chunks_files = {k: v for k, v in
-                              enumerate(chunked_iterable(file_urls, files_per_chunk))}
+                              enumerate(chunked_iterable(self.file_urls, self.files_per_chunk))}
 
 
     def filenames_for_chunk(self, chunk_key):
diff --git a/tests/fixtures.py b/tests/fixtures.py
index 0e54d899..7f0f1d92 100644
--- a/tests/fixtures.py
+++ b/tests/fixtures.py
@@ -7,7 +7,7 @@
 import pytest
 import xarray as xr
 
-
+from pangeo_forge.recipe.target import Target
 
 # where to run the http server
 _PORT = "8080"
@@ -71,6 +71,14 @@ def netcdf_http_server(netcdf_local_paths):
     p.kill()
 
 
+@pytest.fixture()
+def tmp_target(tmpdir_factory):
+    import fsspec
+    fs = fsspec.get_filesystem_class("file")()
+    path = str(tmpdir_factory.mktemp("target"))
+    return Target(fs, path)
+
+
 # tests that our fixtures work
 
 
@@ -86,3 +94,12 @@ def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
     open_files = [fsspec.open(url).open() for url in urls]
     ds = xr.open_mfdataset(open_files, combine="nested", concat_dim="time")
     assert ds.identical(daily_xarray_dataset)
+
+
+def test_target(tmp_target):
+    mapper = tmp_target.get_mapper()
+    mybytes = b'bar'
+    mapper['foo'] = b'bar'
+    with open(tmp_target.path + '/foo') as f:
+        res = f.read()
+    assert res == 'bar'
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 7ca003a6..b797d73f 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -17,7 +17,8 @@ def test_file_sequence_recipe(file_urls, files_per_chunk, expected_keys, expecte
     r = recipe.FileSequenceRecipe(
         file_urls=file_urls,
         sequence_dim="time",
-        files_per_chunk=files_per_chunk
+        files_per_chunk=files_per_chunk,
+        target=
     )
 
     chunk_keys = list(r.iter_chunks())

From 0f311415f0802121ec2a85c66c015bcfcc7df99f Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Sat, 28 Nov 2020 22:09:25 -0500
Subject: [PATCH 04/34] made target fixture

---
 tests/test_recipe.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index b797d73f..11a0f41d 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -2,7 +2,7 @@
 
 from pangeo_forge.recipe import recipe
 
-from .fixtures import daily_xarray_dataset, netcdf_local_paths
+from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_target
 
 dummy_fnames = ["a.nc", "b.nc", "c.nc"]
 @pytest.mark.parametrize(
@@ -12,13 +12,13 @@
         (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",)])
     ]
 )
-def test_file_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames):
+def test_file_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames, tmp_target):
 
     r = recipe.FileSequenceRecipe(
         file_urls=file_urls,
         sequence_dim="time",
         files_per_chunk=files_per_chunk,
-        target=
+        target=tmp_target
     )
 
     chunk_keys = list(r.iter_chunks())
@@ -29,7 +29,7 @@ def test_file_sequence_recipe(file_urls, files_per_chunk, expected_keys, expecte
         assert fnames == expected
 
 
-def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_dir):
+def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target):
 
     r = StandardSequentialRecipe(
         file_urls=netcdf_local_paths,

From 15e240d2d19967534510ae9a3448965954f86b8f Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Sun, 29 Nov 2020 22:24:59 -0500
Subject: [PATCH 05/34] spaghetti at this point

---
 pangeo_forge/recipe/recipe.py  | 120 ++++++++++++++++++++++++++-------
 pangeo_forge/recipe/storage.py |  59 ++++++++++++++++
 pangeo_forge/recipe/target.py  |  22 ------
 3 files changed, 154 insertions(+), 47 deletions(-)
 create mode 100644 pangeo_forge/recipe/storage.py
 delete mode 100644 pangeo_forge/recipe/target.py

diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe/recipe.py
index 38c67d95..142fe849 100644
--- a/pangeo_forge/recipe/recipe.py
+++ b/pangeo_forge/recipe/recipe.py
@@ -9,14 +9,13 @@
 from ..utils import chunked_iterable
 from .target import Target
 
-from typing import Optional, Iterable
+from typing import Optional, Iterable, Callable
 
 
 ### How to manually execute a recipe: ###
 #
-#   r = PangeoForgeTarget()
-#   r = MyRecipe(**opts) # 1
-#   r.set_target(tmp_dir) # 2
+#   t = PangeoForgeTarget()
+#   r = MyRecipe(target=t, **opts) # 1
 #   # manual execution of recipe
 #   r.prepare() # 3
 #   for input_key in r.iter_inputs():
@@ -44,39 +43,110 @@
 class DatasetRecipe():
     target: Target
 
+    def prepare(self):
+        pass
+
+    def iter_inputs(self):
+        return []
+
+    def cache_input(self, input_key):
+        raise NotImplementedError
+
+    def iter_chunks(self):
+        raise NotImplementedError
+
+    def store_chunk(self, chunk_key):
+        raise NotImplementedError
+
+    def finalize(self):
+        raise NotImplementedError
+
+
+@dataclass
+class FSSpecFileOpenerMixin:
+    input_open_kwargs: dict = {}
+
+    @contextmanager
+    def input_opener(self, fname):
+        with fsspec.open(fname, **input_open_kwargs) as f:
+            yield f
+
 
-class FSSpecInputOpenerMixin:
+@dataclass
+class InputCachingMixin(FSSpecFileOpenerMixin):
+    input_cache: InputCache
+    require_cache: bool -> False
+
+    # returns a function that takes one input, the input_key
+    # this allows us to parallelize these operations
+    @property
+    def cache_input(self):
 
-    def open_file(self, fname):
-        # todo: caching
-        return fsspec.open(fname)
+        def cache_func(fname: str) -> None:
+            with super().input_opener(fname, mode="rb") as source:
+                with self.input_cache.open(fname, mode="wb") as target:
+                    target.write(source.read())
 
+        return cache_func
 
-class XarrayChunkOpenerMixin:
+    @contextmanager
+    def input_opener(self, fname):
+        if self.input_cache.exists(fname):
+            with self.input_chache.open(fname, mode='rb') as f:
+                yield f
+        elif self.require_cache:
+            raise IOError("Input can only be opened from cache. Call .cache_input first.")
+        else:
+            # This will bypass the cache. May be slow.
+            with super().input_opener(fname, mode="rb") as f:
+                yield f
 
-    def open_chunk(self, filenames, load=True):
-        files_to_open = [self.open_file(f) for f in filenames]
-        ds_chunk = xr.open_mfdataset(filenames, **self.open_chunk_kwargs)
-        if load:
-            ds_chunk.load()
-        return ds_chunk
+
+
+@dataclass
+class XarrayInputOpener:
+    xarray_open_kwargs: Any
+
+    def open_input(self, fname):
+        with self.input_opener(fname)  as f:
+            ds = xr.open_dataset(f, **self.xarray_open_kwargs)
+        return ds
+
+
+@dataclass
+class XarrayConcatChunkOpener(XarrayInputOpener):
+    xarray_concat_kwargs: Any
+
+    def open_chunk(self, chunk_key):
+        inputs = self.inputs_for_chunk(chunk_key)
+        dsets = [self.open_input(i) for i in inputs]
+        combined = xr.concat(dsets, **xarray_concat_kwargs)
+        # TODO: maybe do some chunking here?
+        return combined
 
 
 class ZarrWriterMixin:
 
+    @property
+    def store_chunk(self) -> Callable:
+
+        def _store_chunk(chunk_key):
+            ds_chunk = self.open_chunk(chunk_key)
+            target_mapper = self.target.get_mapper()
+            write_region = self.get_write_region(chunk_key)
+            ds_chunk.to_zarr(target_mapper, region=write_region)
+
+        return _store_chunk
 
-    def store_chunk(self, chunk_key):
-        filenames = self.filenames_for_chunk(chunk_key)
-        ds_chunk = self.open_chunk(filenames)
-        target_mapper = self.target.get_mapper()
-        write_region = self.get_write_region(chunk_key)
-        ds_chunk.to_zarr(target_mapper, region=write_region, **target_kwargs)
 
+    @property
+    def open_target(self) -> Callable:
 
-    def open_target(self):
-        target_mapper = self.target.get_mapper()
-        return xr.open_zarr(target_mapper)
+        def _open_target():
+            target_mapper = self.target.get_mapper()
+            return xr.open_zarr(target_mapper)
 
+        return _open_target
 
 
 
@@ -94,7 +164,7 @@ def __post_init__(self):
                               enumerate(chunked_iterable(self.file_urls, self.files_per_chunk))}
 
 
-    def filenames_for_chunk(self, chunk_key):
+    def inputs_for_chunk(self, chunk_key):
         return self._chunks_files[chunk_key]
 
 
diff --git a/pangeo_forge/recipe/storage.py b/pangeo_forge/recipe/storage.py
new file mode 100644
index 00000000..474597a9
--- /dev/null
+++ b/pangeo_forge/recipe/storage.py
@@ -0,0 +1,59 @@
+from dataclasses import dataclass
+from contextlib import closing, contextmanager
+import fsspec
+
+
+@dataclass
+class Target:
+    """Representation of a storage target for Pangeo Forge.
+    Attributes
+    ----------
+    fs : FileSystemSpec.AbtractFileSystem
+        The filesystem we are writing to. Should be instantiated outside this
+        class.
+    path : str
+        The path where the target data will be saved.
+    """
+
+    fs: fsspec.AbstractFileSystem
+    path: str
+
+    def get_mapper(self):
+        # don't want to use this because we want to use a fancier Zarr FSStore
+        return self.fs.get_mapper(self.path)
+
+
+def _hash_path(path: str) -> str:
+    return str(hash(path))
+
+
+@dataclass
+class InputCache:
+    """Representation of an intermediate storage location where remote files
+    Can be cached locally.
+
+    Attributes
+    ----------
+    fs : FileSystemSpec.AbtractFileSystem
+        The filesystem we are writing to. Should be instantiated outside this
+        class.
+    prefix : str
+        A path prepended to all paths.
+    """
+
+    fs: fsspec.AbstractFileSystem
+    prefix: str = ""
+
+    def _full_path(self, path):
+        return os.path.join(self.prefix, _hash_path(path))
+
+    def exists(self, path):
+        return self.fs.exists(self._full_path(path))
+
+    def rm(self, path):
+        self.fs.rm(self._full_path(path))
+
+    @contextmanager
+    def open(self, path):
+        with self.fs.open(self._full_path(path)) as f:
+            yield f
diff --git a/pangeo_forge/recipe/target.py b/pangeo_forge/recipe/target.py
deleted file mode 100644
index b7274d94..00000000
--- a/pangeo_forge/recipe/target.py
+++ /dev/null
@@ -1,22 +0,0 @@
-from dataclasses import dataclass
-import fsspec
-
-
-@dataclass
-class Target:
-    """Representation of a storage target for Pangeo Forge.
-    Attributes
-    ----------
-    url : FileSystemSpec.AbtractFileSystem
-        The filesystem we are writing to. Should be instantiated outside this
-        class.
-    path : str
-        The path where the target data will be saved.
-    """
-
-    fs: fsspec.AbstractFileSystem
-    path: str
-
-    def get_mapper(self):
-        # don't want to use this because we want to use a fancier Zarr FSStore
-        return self.fs.get_mapper(self.path)

From 18a895a65e1037da863e4ecbaf7b7c5ea6370b04 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 15:34:05 -0500
Subject: [PATCH 06/34] working storage classes

---
 pangeo_forge/recipe/storage.py |  9 +++++++--
 tests/fixtures.py              | 20 +++++++++++++++++++-
 2 files changed, 26 insertions(+), 3 deletions(-)

diff --git a/pangeo_forge/recipe/storage.py b/pangeo_forge/recipe/storage.py
index 474597a9..04e949a0 100644
--- a/pangeo_forge/recipe/storage.py
+++ b/pangeo_forge/recipe/storage.py
@@ -1,6 +1,7 @@
 from dataclasses import dataclass
 from contextlib import closing, contextmanager
 import fsspec
+import os
 
 
 @dataclass
@@ -54,6 +55,10 @@ def rm(self, path):
         self.fs.rm(self._full_path(path))
 
     @contextmanager
-    def open(self, path):
-        with self.fs.open(self._full_path(path)) as f:
+    def open(self, path, **kwargs):
+        with self.fs.open(self._full_path(path), **kwargs) as f:
             yield f
+
+    def __post_init__(self):
+        if not self.fs.isdir(self.prefix):
+            self.fs.mkdir(self.prefix)
diff --git a/tests/fixtures.py b/tests/fixtures.py
index 7f0f1d92..28b2ead5 100644
--- a/tests/fixtures.py
+++ b/tests/fixtures.py
@@ -7,7 +7,7 @@
 import pytest
 import xarray as xr
 
-from pangeo_forge.recipe.target import Target
+from pangeo_forge.recipe.storage import Target, InputCache
 
 # where to run the http server
 _PORT = "8080"
@@ -79,6 +79,13 @@ def tmp_target(tmpdir_factory):
     return Target(fs, path)
 
 
+@pytest.fixture()
+def tmp_cache(tmpdir_factory):
+    path = str(tmpdir_factory.mktemp("cache"))
+    fs = fsspec.get_filesystem_class("file")()
+    cache = InputCache(fs, prefix='cache')
+    return cache
+
 # tests that our fixtures work
 
 
@@ -103,3 +110,14 @@ def test_target(tmp_target):
     with open(tmp_target.path + '/foo') as f:
         res = f.read()
     assert res == 'bar'
+
+
+def test_cache(tmp_cache):
+    assert not tmp_cache.exists('foo')
+    with tmp_cache.open('foo', mode='w') as f:
+        f.write('bar')
+    assert tmp_cache.exists('foo')
+    with tmp_cache.open('foo', mode='r') as f:
+        assert f.read() == 'bar'
+    tmp_cache.rm('foo')
+    assert not tmp_cache.exists('foo')

From 301206b4bc01565f7f2b1dc5e02d7477c4cc2840 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 16:39:13 -0500
Subject: [PATCH 07/34] recipe working pretty well

---
 pangeo_forge/recipe/recipe.py | 255 +++++++++++++++++++++++++---------
 1 file changed, 186 insertions(+), 69 deletions(-)

diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe/recipe.py
index 142fe849..25702df1 100644
--- a/pangeo_forge/recipe/recipe.py
+++ b/pangeo_forge/recipe/recipe.py
@@ -2,15 +2,21 @@
 A Pangeo Forge Recipe
 """
 
-from dataclasses import dataclass
+import logging
+from dataclasses import dataclass, field
+from contextlib import contextmanager
+from typing import Optional, Iterable, Callable, Any
 
+import numpy as np
 import xarray as xr
 import fsspec
-from ..utils import chunked_iterable
-from .target import Target
+import zarr
 
-from typing import Optional, Iterable, Callable
+from ..utils import chunked_iterable
+from .storage import Target, InputCache
 
+#logger = logging.getLogger(__name__)
+logger = logging.getLogger("recipe")
 
 ### How to manually execute a recipe: ###
 #
@@ -42,48 +48,76 @@
 @dataclass
 class DatasetRecipe():
     target: Target
+    chunk_preprocess_funcs: Iterable[Callable]
 
+    @property
     def prepare(self):
-        pass
+        def _prepare():
+            pass
+        return _prepare
 
     def iter_inputs(self):
         return []
 
-    def cache_input(self, input_key):
-        raise NotImplementedError
+    # need to figure out what's going on with these methods and inheritance
+    # @property
+    # def cache_input(self):
+    #     def _cache_input(input_key):
+    #         raise NotImplementedError
+    #     return _cache_input
+
+    # this only gets run when iterating, not preparing!
+    def preprocess_chunk(self, ds):
+        for f in self.chunk_preprocess_funcs:
+             ds = f(ds)
+        return ds
 
     def iter_chunks(self):
         raise NotImplementedError
 
-    def store_chunk(self, chunk_key):
-        raise NotImplementedError
+    # @property
+    # def store_chunk(self):
+    #     def _store_chunk(chunk_key):
+    #         raise NotImplementedError
+    #     return _store_chunk
 
-    def finalize(self):
-        raise NotImplementedError
+    # @property
+    # def finalize(self):
+    #
+    #     def _finalize():
+    #         pass
+    #     return _finalize
 
+# Notes about dataclasses:
+# - https://www.python.org/dev/peps/pep-0557/#inheritance
+# - https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
+# This means that, for now, I can't get default arguments to work.
 
 @dataclass
 class FSSpecFileOpenerMixin:
-    input_open_kwargs: dict = {}
+    #input_open_kwargs: dict #= field(default_factory=dict)
 
     @contextmanager
-    def input_opener(self, fname):
-        with fsspec.open(fname, **input_open_kwargs) as f:
+    def input_opener(self, fname, **kwargs):
+        logger.info(f"Opening input '{fname}'")
+        with fsspec.open(fname, **kwargs) as f:
             yield f
 
 
 @dataclass
 class InputCachingMixin(FSSpecFileOpenerMixin):
+    require_cache: bool #= False
     input_cache: InputCache
-    require_cache: bool -> False
 
     # returns a function that takes one input, the input_key
     # this allows us to parallelize these operations
     @property
     def cache_input(self):
 
+        opener = super().input_opener
         def cache_func(fname: str) -> None:
-            with super().input_opener(fname, mode="rb") as source:
+            logger.info(f"Caching input '{fname}'")
+            with opener(fname, mode="rb") as source:
                 with self.input_cache.open(fname, mode="wb") as target:
                     target.write(source.read())
 
@@ -92,11 +126,14 @@ def cache_func(fname: str) -> None:
     @contextmanager
     def input_opener(self, fname):
         if self.input_cache.exists(fname):
-            with self.input_chache.open(fname, mode='rb') as f:
+            logger.info(f"Input '{fname}' found in cache")
+            with self.input_cache.open(fname, mode='rb') as f:
                 yield f
         elif self.require_cache:
+            # this creates an error on prepare because nothing is cached
             raise IOError("Input can only be opened from cache. Call .cache_input first.")
         else:
+            logger.info(f"Input '{fname}' not found in cache. Opening directly.")
             # This will bypass the cache. May be slow.
             with super().input_opener(fname, mode="rb") as f:
                 yield f
@@ -105,26 +142,39 @@ def input_opener(self, fname):
 
 @dataclass
 class XarrayInputOpener:
-    xarray_open_kwargs: Any
+    xarray_open_kwargs: dict
 
     def open_input(self, fname):
-        with self.input_opener(fname)  as f:
-            ds = xr.open_dataset(f, **self.xarray_open_kwargs)
+        with self.input_opener(fname) as f:
+            logger.info(f"Opening input with Xarray '{fname}'")
+            ds = xr.open_dataset(f, **self.xarray_open_kwargs).load()
+            # do we always want to remove encoding? I think so.
+        ds = _fix_scalar_attr_encoding(ds)
+        logger.debug(f"{ds}")
         return ds
 
 
 @dataclass
 class XarrayConcatChunkOpener(XarrayInputOpener):
-    xarray_concat_kwargs: Any
+    xarray_concat_kwargs: dict
 
     def open_chunk(self, chunk_key):
+        logger.info(f"Concatenating inputs for chunk '{chunk_key}'")
         inputs = self.inputs_for_chunk(chunk_key)
         dsets = [self.open_input(i) for i in inputs]
-        combined = xr.concat(dsets, **xarray_concat_kwargs)
+        # CONCAT DELETES ENCODING!!!
+        ds = xr.concat(dsets, self.sequence_dim, **self.xarray_concat_kwargs)
+        logger.debug(f"{ds}")
+
+        # do we really want to just delete all encoding?
+        #for v in ds.variables:
+        #    ds[v].encoding = {}
+
         # TODO: maybe do some chunking here?
-        return combined
+        return ds
 
 
+@dataclass
 class ZarrWriterMixin:
 
     @property
@@ -132,50 +182,99 @@ def store_chunk(self) -> Callable:
 
         def _store_chunk(chunk_key):
             ds_chunk = self.open_chunk(chunk_key)
+            ds_chunk = self.preprocess_chunk(ds_chunk)
             target_mapper = self.target.get_mapper()
-            write_region = self.get_write_region(chunk_key)
+            write_region = self.region_for_chunk(chunk_key)
+            logger.info(f"Storing chunk '{chunk_key}' to Zarr region {write_region}")
             ds_chunk.to_zarr(target_mapper, region=write_region)
 
         return _store_chunk
 
 
-    @property
-    def open_target(self) -> Callable:
+    def open_target(self):
+        target_mapper = self.target.get_mapper()
+        return xr.open_zarr(target_mapper)
 
-        def _open_target():
-            target_mapper = self.target.get_mapper()
-            return xr.open_zarr(target_mapper)
 
-        return _open_target
+    def initialize_target(self, ds, **expand_dims):
+        logger.info(f"Creating a new dataset in target")
+        target_mapper = self.target.get_mapper()
+        ds.to_zarr(target_mapper, mode='w', compute=False)
+
+
+    def expand_target_dim(self, dim, dimsize):
+        target_mapper = self.target.get_mapper()
+        zgroup = zarr.open_group(target_mapper)
+
+        ds = self.open_target()
+        sequence_axes = {v: ds[v].get_axis_num(dim)
+                         for v in ds.variables
+                         if dim in ds[v].dims}
+
+        for v, axis in sequence_axes.items():
+            arr = zgroup[v]
+            shape = list(arr.shape)
+            shape[axis] = dimsize
+            arr.resize(shape)
 
 
 
 
 @dataclass
-class FileSequenceRecipe(DatasetRecipe):
-    file_urls: Iterable[str]
+class ZarrConsolidatorMixin():
+    consolidate_zarr: bool #= True
+
+    @property
+    def finalize(self):
+
+        def _finalize():
+            if self.consolidate_zarr:
+                logger.info(f"Consolidating Zarr metadata")
+                target_mapper = self.target.get_mapper()
+                zarr.consolidate_metadata(target_mapper)
+
+        return _finalize
+
+
+@dataclass
+class SequenceRecipe(DatasetRecipe):
+    input_urls: Iterable[str]
     sequence_dim: str
-    files_per_chunk: int = 1
-    nitems_per_file: int = 1
+    inputs_per_chunk: int = 1
+    nitems_per_input: int = 1
 
 
     def __post_init__(self):
-        self._chunks_files = {k: v for k, v in
-                              enumerate(chunked_iterable(self.file_urls, self.files_per_chunk))}
+        self._chunks_inputs = {k: v for k, v in
+                              enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))}
+
+        def drop_vars(ds):
+            # writing a region means that all the variables MUST have sequence_dim
+            to_drop = [v for v in ds.variables
+                       if self.sequence_dim not in ds[v].dims]
+            return ds.drop(to_drop)
+
+        self.chunk_preprocess_funcs.append(drop_vars)
 
 
     def inputs_for_chunk(self, chunk_key):
-        return self._chunks_files[chunk_key]
+        return self._chunks_inputs[chunk_key]
+
+
+    def iter_inputs(self):
+        for chunk_key in self.iter_chunks():
+            for input in self.inputs_for_chunk(chunk_key):
+                yield input
 
 
     def nitems_for_chunk(self, chunk_key):
-        return self.nitems_per_file * len(self.filenames_for_chunk(chunk_key))
+        return self.nitems_per_input * len(self.inputs_for_chunk(chunk_key))
 
 
     def region_for_chunk(self, chunk_key):
         # return a dict suitable to pass to xr.to_zarr(region=...)
         # specifies where in the overall array to put this chunk's data
-        stride = self.nitems_per_file * self.files_per_chunk
+        stride = self.nitems_per_input * self.inputs_per_chunk
         start = chunk_key * stride
         return {
             self.sequence_dim:
@@ -183,56 +282,74 @@ def region_for_chunk(self, chunk_key):
         }
 
 
-    def sequence_dim(self):
+    def sequence_len(self):
         # tells the total size of dataset along the sequence dimension
-        return {
-            self.sequence_dim:
-                sum([self.nitems_for_chunk(k) for k in self.iter_chunks()])
-        }
+        return sum([self.nitems_for_chunk(k) for k in self.iter_chunks()])
 
 
     def sequence_chunks(self):
         # chunking
-        return {self.sequence_dim: self.files_per_chunk * self.nitems_per_file}
+        return {self.sequence_dim: self.inputs_per_chunk * self.nitems_per_input}
 
 
     def iter_chunks(self):
-        for k in self._chunks_files:
+        for k in self._chunks_inputs:
             yield k
 
+    @property
     def prepare(self):
 
-        target_store = self.get_store_target()
-
-        try:
-            ds = self.open_target(target_store)
+        def _prepare():
 
-        except IOError:
-            first_chunk_key = next(self.iter_chunks())
-            ds = self.open_chunk(first_chunk_key).chunk()
-            ds.to_zarr(path, compute=False, consolidated=False)
+            target_store = self.target.get_mapper()
 
-        encoding = {v: ds[v].encoding for v in ds}
+            try:
+                ds = self.open_target()
+                logger.info(f"Found an existing dataset in target")
+                logger.debug(f"{ds}")
+            except (IOError, zarr.errors.GroupNotFoundError):
+                first_chunk_key = next(self.iter_chunks())
+                ds = self.open_chunk(first_chunk_key).chunk()
 
-        # now resize the sequence dim at the zarr level
-        sequence_axes = {v: ds[v].get_axis_num(self.sequence_dim)
-                         for v in ds
-                         if self.sequence_dim in ds[v].dims}
-        N = self.sequence_dim()
+                # make sure the concat dim has a valid fill_value to avoid
+                # overruns when writing chunk
+                #ds[self.sequence_dim].encoding = {'_FillValue': 0}
+                # actually not necessary if we use decode_times=False
+                self.initialize_target(ds)
 
-        zgroup = zarr.open_group(target_store)
+            self.expand_target_dim(self.sequence_dim, self.sequence_len())
 
-        for v, axis in sequence_axes.items():
-            arr = zgroup[v]
-            shape = list(arr.shape)
-            shape[axis] = N
-            arr.resize(shape)
+        return _prepare
 
 
+@dataclass
 class StandardSequentialRecipe(
-        FileSequenceRecipe,
-        FSSpecInputOpenerMixin,
-        XarrayChunkOpenerMixin,
+        SequenceRecipe,
+        InputCachingMixin,
+        XarrayConcatChunkOpener,
         ZarrWriterMixin,
+        ZarrConsolidatorMixin
     ):
     pass
+
+
+# helper utilities
+
+# only needed because of
+# https://github.com/pydata/xarray/issues/4631
+def _fix_scalar_attr_encoding(ds):
+
+    def _fixed_attrs(d):
+        fixed = {}
+        for k, v in d.items():
+            if isinstance(v, np.ndarray) and len(v) == 1:
+                fixed[k] = v[0]
+        return fixed
+
+    ds = ds.copy()
+    ds.attrs.update(_fixed_attrs(ds.attrs))
+    ds.encoding.update(_fixed_attrs(ds.encoding))
+    for v in ds.variables:
+        ds[v].attrs.update(_fixed_attrs(ds[v].attrs))
+        ds[v].encoding.update(_fixed_attrs(ds[v].encoding))
+    return ds

From a11fdc35923f1f7815cfe40ee9b1342db765447c Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 21:21:32 -0500
Subject: [PATCH 08/34] recipe tests pass

---
 pangeo_forge/recipe/recipe.py |  6 ++---
 tests/fixtures.py             |  2 ++
 tests/test_recipe.py          | 44 +++++++++++++++++++++++------------
 3 files changed, 34 insertions(+), 18 deletions(-)

diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe/recipe.py
index 25702df1..2c828500 100644
--- a/pangeo_forge/recipe/recipe.py
+++ b/pangeo_forge/recipe/recipe.py
@@ -175,7 +175,7 @@ def open_chunk(self, chunk_key):
 
 
 @dataclass
-class ZarrWriterMixin:
+class ZarrXarrayWriterMixin:
 
     @property
     def store_chunk(self) -> Callable:
@@ -313,7 +313,7 @@ def _prepare():
 
                 # make sure the concat dim has a valid fill_value to avoid
                 # overruns when writing chunk
-                #ds[self.sequence_dim].encoding = {'_FillValue': 0}
+                ds[self.sequence_dim].encoding = {'_FillValue': -1}
                 # actually not necessary if we use decode_times=False
                 self.initialize_target(ds)
 
@@ -327,7 +327,7 @@ class StandardSequentialRecipe(
         SequenceRecipe,
         InputCachingMixin,
         XarrayConcatChunkOpener,
-        ZarrWriterMixin,
+        ZarrXarrayWriterMixin,
         ZarrConsolidatorMixin
     ):
     pass
diff --git a/tests/fixtures.py b/tests/fixtures.py
index 28b2ead5..4ac47d9f 100644
--- a/tests/fixtures.py
+++ b/tests/fixtures.py
@@ -45,6 +45,8 @@ def daily_xarray_dataset():
 def netcdf_local_paths(daily_xarray_dataset, tmpdir_factory, request):
     """Return a list of paths pointing to netcdf files."""
     tmp_path = tmpdir_factory.mktemp("netcdf_data")
+    items_per_file = {"D": 1, "2D": 2}
+    daily_xarray_dataset.attrs['items_per_file'] = items_per_file[request.param]
     gb = daily_xarray_dataset.resample(time=request.param)
     _, datasets = zip(*gb)
     fnames = [f"{n:03d}.nc" for n in range(len(datasets))]
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 11a0f41d..7b266473 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -1,8 +1,10 @@
 import pytest
+import xarray as xr
+import zarr
 
 from pangeo_forge.recipe import recipe
 
-from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_target
+from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache
 
 dummy_fnames = ["a.nc", "b.nc", "c.nc"]
 @pytest.mark.parametrize(
@@ -12,32 +14,43 @@
         (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",)])
     ]
 )
-def test_file_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames, tmp_target):
+def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames, tmp_target):
 
-    r = recipe.FileSequenceRecipe(
-        file_urls=file_urls,
+    r = recipe.SequenceRecipe(
+        input_urls=file_urls,
         sequence_dim="time",
-        files_per_chunk=files_per_chunk,
-        target=tmp_target
+        inputs_per_chunk=files_per_chunk,
+        target=tmp_target,
+        chunk_preprocess_funcs=[],
     )
 
+    assert r.sequence_len() == len(file_urls)
+
     chunk_keys = list(r.iter_chunks())
     assert chunk_keys == expected_keys
 
     for k, expected in zip(r.iter_chunks(), expected_filenames):
-        fnames = r.filenames_for_chunk(k)
+        fnames = r.inputs_for_chunk(k)
         assert fnames == expected
 
 
-def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target):
+def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
 
-    r = StandardSequentialRecipe(
-        file_urls=netcdf_local_paths,
-        sequence_dim='time',
+    r = recipe.StandardSequentialRecipe(
+        consolidate_zarr=True,
+        xarray_open_kwargs={},
+        xarray_concat_kwargs={},
+        require_cache=False,
+        input_cache=tmp_cache,
+        target=tmp_target,
+        chunk_preprocess_funcs=[],
+        input_urls=netcdf_local_paths,
+        sequence_dim="time",
+        inputs_per_chunk=1,
+        nitems_per_input=daily_xarray_dataset.attrs['items_per_file']
     )
 
-    r.set_target(tmp_dir)
-    # manual execution of recipe
+    # this is the cannonical way to manually execute a recipe
     r.prepare()
     for input_key in r.iter_inputs():
         r.cache_input(input_key)
@@ -45,5 +58,6 @@ def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target):
         r.store_chunk(chunk_key)
     r.finalize()
 
-    ds_target = xr.open_dataset(tmp_dir)
-    assert ds_target.identical(daily_xarray_dataset)
+    ds_target = xr.open_zarr(tmp_target.get_mapper(), consolidated=True).load()
+    ds_expected = daily_xarray_dataset.compute()
+    assert ds_target.identical(ds_expected)

From b1cc65b06b3356e41749300cdc35a833fcf0b640 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 21:24:07 -0500
Subject: [PATCH 09/34] prune old stuff

---
 pangeo_forge/pipelines/README.md           |   8 -
 pangeo_forge/pipelines/__init__.py         |   0
 pangeo_forge/pipelines/base.py             | 169 ---------------------
 pangeo_forge/pipelines/http_xarray_zarr.py |  35 -----
 pangeo_forge/pipelines/job.yaml            |  28 ----
 pangeo_forge/pipelines/worker_pod.yaml     |  41 -----
 pangeo_forge/tasks/README.md               |   8 -
 pangeo_forge/tasks/__init__.py             |   0
 pangeo_forge/tasks/http.py                 |  36 -----
 pangeo_forge/tasks/xarray.py               |  73 ---------
 pangeo_forge/tasks/zarr.py                 |  23 ---
 11 files changed, 421 deletions(-)
 delete mode 100644 pangeo_forge/pipelines/README.md
 delete mode 100644 pangeo_forge/pipelines/__init__.py
 delete mode 100644 pangeo_forge/pipelines/base.py
 delete mode 100644 pangeo_forge/pipelines/http_xarray_zarr.py
 delete mode 100644 pangeo_forge/pipelines/job.yaml
 delete mode 100644 pangeo_forge/pipelines/worker_pod.yaml
 delete mode 100644 pangeo_forge/tasks/README.md
 delete mode 100644 pangeo_forge/tasks/__init__.py
 delete mode 100644 pangeo_forge/tasks/http.py
 delete mode 100644 pangeo_forge/tasks/xarray.py
 delete mode 100644 pangeo_forge/tasks/zarr.py

diff --git a/pangeo_forge/pipelines/README.md b/pangeo_forge/pipelines/README.md
deleted file mode 100644
index a1738027..00000000
--- a/pangeo_forge/pipelines/README.md
+++ /dev/null
@@ -1,8 +0,0 @@
-# `pangeo_forge.pipelines`
-
-A collection of Pangeo-forge _Pipelines_.
-
-## Details
-
-- An abstract Pipeline class is defined in `base.py`.
-- A general file naming convention might be `{source}_{container}_{target}.py`. For example, a pipeline that downloads netCDF files from a remote http server and publishes a Zarr store would be `http_xarray_zarr.py`. The main class in each module would follow a similar convention, e.g. `HttpXarrayZarr`.
diff --git a/pangeo_forge/pipelines/__init__.py b/pangeo_forge/pipelines/__init__.py
deleted file mode 100644
index e69de29b..00000000
diff --git a/pangeo_forge/pipelines/base.py b/pangeo_forge/pipelines/base.py
deleted file mode 100644
index ebf5d3ea..00000000
--- a/pangeo_forge/pipelines/base.py
+++ /dev/null
@@ -1,169 +0,0 @@
-"""
-Base class for most pipelines.
-
-Design
-------
-
-To the extent possible, we want recipe maintainers to focus on the code
-needed to express the data transformations. In particular, we don't
-want them to worry about things like the execution environment and how
-their code is loaded into it.
-
-The Flow
-========
-
-We use prefect_ to express ETL pipelines. It provides us a few things:
-
-1. A Highlevel API for writing data transformations.
-2. A robust scheduling and orchestration system for executing pipelines.
-
-Every pangeo-forge recipe *must* have a ``prefect.Flow`` instance at the
-top-level of their pipeline module.
-
-
-.. code-block:: python
-
-   class Pipeline(pangeo_forge.AbstractPipeline):
-       @property
-       def flow(self) -> Flow:
-           with Flow(
-               self.name,
-               environment=self.environment,
-               storage=self.storage
-           ) as flow:
-              # Your pipeline goes here.
-              ...
-
-           return flow
-
-   pipeline = Pipeline()
-   flow = pipeline.flow
-
-The ``storage`` keyword controls how your source code is loaded into
-pangeo-forge, and the ``environment`` keyword controls where it's run.
-"""
-import textwrap
-from abc import ABC, abstractmethod
-from pathlib import Path
-from typing import List
-
-from prefect import Flow
-from prefect.environments import DaskKubernetesEnvironment, Environment
-from prefect.environments.storage import Storage
-from prefect.environments.storage.github import GitHub
-
-HERE = Path(__file__).parent.absolute()
-
-
-def finalize_flow(cls):
-    # Nothing to see here, move along.
-    # OK, fine... this is a small hack to hide Prefect's orchestration stuff
-    # (environments, storage) from the recipe writer. Prefect currently
-    # requires that the module stored on GitHub have
-    #   1. A `Flow` instance at the top-level of the module.
-    #   2. A properly set `storage` and `environment`.
-    # Which would require this at the bottom of every python module
-    # >>> pipeline = MyPipeline()
-    # >>> flow = pipeline.flow
-    # >>> flow.storage = pipeline.storage
-    # >>> flow.environment = pipeline.environment
-    # Which is ugly. To avoid that, we just override how `.` works on
-    # AbstractPipeline. If we're getting `.flow`, this little decorator
-    # takes over and attaches the storage and environment.
-    orig_getattribute = cls.__getattribute__
-
-    def new_getattribute(self, name):
-        result = orig_getattribute(self, name)
-        if name == "flow":
-            result.storage = self.storage
-            result.environment = self.environment
-        return result
-
-    cls.__getattribute__ = new_getattribute
-    return cls
-
-
-@finalize_flow
-class AbstractPipeline(ABC):
-    name = "AbstractPipeline"
-    path = "recipe/pipeline.py"
-
-    @property
-    @abstractmethod
-    def repo(self):
-        """The GitHub repository containing the pipeline definition."""
-
-    @property
-    @abstractmethod
-    def sources(self) -> List[str]:
-        """A list of source URLs containing the original data."""
-        pass
-
-    @property
-    @abstractmethod
-    def targets(self) -> List[str]:
-        """A list of target URLs where the transformed data is written."""
-        pass
-
-    @abstractmethod
-    def flow(self) -> Flow:
-        """The """
-        pass
-
-    @property
-    def environment(self) -> Environment:
-        """
-        The pipeline runtime environment.
-
-        Returns
-        -------
-        prefect.environments.Environment
-            An instance of a Prefect Environment. By default
-            a :class:`prefect.environments.DaskKubernetesEnvironment`
-            is used.
-        """
-        scheduler_spec_file = str(HERE / "job.yaml")
-        worker_spec_file = str(HERE / "worker_pod.yaml")
-
-        environment = DaskKubernetesEnvironment(
-            min_workers=1,
-            max_workers=30,
-            scheduler_spec_file=scheduler_spec_file,
-            worker_spec_file=worker_spec_file,
-            metadata=dict(image="pangeoforge/default-image"),
-        )
-        return environment
-
-    @property
-    def storage(self) -> Storage:
-        """
-        The pipeline storage.
-
-        Returns
-        -------
-        prefect.environments.storage.Storage
-            By default a :class:`prefect.environments.storage.github.GitHub`
-            environment is used with ``self.repo`` as the repository
-            and ``self.path`` as the path.
-        """
-        return GitHub(self.repo, path=self.path)
-
-    def _generate_run(self, source):
-        name = type(self).__name__
-        with open(source, encoding="utf-8") as f:
-            source = f.read()
-
-        return textwrap.dedent(
-            """\
-            # file: run.py
-            # generated by pangeo-forge.
-
-            {source}
-
-            # ----------------------------------------------------------------
-            pipe = {name}()
-            flow = pipe.flow
-            flow.storage = pipe.storage
-            flow.environment = pipe.environment
-            """
-        ).format(source=source, name=name)
diff --git a/pangeo_forge/pipelines/http_xarray_zarr.py b/pangeo_forge/pipelines/http_xarray_zarr.py
deleted file mode 100644
index 4b72703a..00000000
--- a/pangeo_forge/pipelines/http_xarray_zarr.py
+++ /dev/null
@@ -1,35 +0,0 @@
-from prefect import Flow
-
-from ..tasks.http import download
-from ..tasks.xarray import combine_and_write
-from ..tasks.zarr import consolidate_metadata
-from ..utils import chunked_iterable
-
-
-class HttpXarrayZarrMixin:
-    @property
-    def flow(self):
-
-        if len(self.targets) == 1:
-            target = self.targets[0]
-        else:
-            raise ValueError("Zarr target requires self.targets be a length one list")
-
-        with Flow(self.name) as _flow:
-
-            cached_sources = [download(k, self.cache_location) for k in self.sources]
-
-            write_tasks = []
-            for source_group in chunked_iterable(cached_sources, self.files_per_chunk):
-                write_task = combine_and_write(
-                    source_group, target, self.append_dim, self.concat_dim,
-                )
-                write_tasks.append(write_task)
-            cm = consolidate_metadata(target, write_tasks)
-
-        # create dependencies in imperative mode
-        for n in range(1, len(write_tasks)):
-            write_tasks[n].set_upstream(write_tasks[n - 1], flow=_flow)
-        cm.set_upstream(write_tasks[-1], flow=_flow)
-
-        return _flow
diff --git a/pangeo_forge/pipelines/job.yaml b/pangeo_forge/pipelines/job.yaml
deleted file mode 100644
index 5fa08ca8..00000000
--- a/pangeo_forge/pipelines/job.yaml
+++ /dev/null
@@ -1,28 +0,0 @@
-apiVersion: batch/v1
-kind: Job
-metadata:
-  name: prefect-dask-job
-  labels:
-    app: prefect-dask-job
-spec:
-  template:
-    metadata:
-      labels:
-        app: prefect-dask-job
-    spec:
-      serviceAccountName: pangeo-forge
-      serviceAccount: pangeo-forge
-      containers:
-        - name: flow
-          imagePullPolicy: Always
-          command: ["/bin/sh", "-c"]
-          args:
-            [
-              'python -c "import prefect; prefect.environments.execution.load_and_run_flow()"',
-            ]
-          resources:
-            requests:
-              cpu: "100m"
-            limits:
-              cpu: "100m"
-      restartPolicy: Never
diff --git a/pangeo_forge/pipelines/worker_pod.yaml b/pangeo_forge/pipelines/worker_pod.yaml
deleted file mode 100644
index d7a16440..00000000
--- a/pangeo_forge/pipelines/worker_pod.yaml
+++ /dev/null
@@ -1,41 +0,0 @@
-kind: Pod
-metadata:
-  labels:
-    app: prefect-dask-worker
-spec:
-  restartPolicy: Never
-  serviceAccountName: pangeo-forge
-  serviceAccount: pangeo-forge
-  containers:
-  - image: pangeoforge/noaa-oisst-avhrr
-    imagePullPolicy: Always
-    args: [dask-worker, --no-bokeh, --death-timeout, '60']
-    name: dask-worker
-    env:
-      - name: PREFECT__CLOUD__GRAPHQL
-        value: PREFECT__CLOUD__GRAPHQL
-      - name: PREFECT__CLOUD__AUTH_TOKEN
-        value: PREFECT__CLOUD__AUTH_TOKEN
-      - name: PREFECT__CONTEXT__FLOW_RUN_ID
-        value: PREFECT__CONTEXT__FLOW_RUN_ID
-      - name: PREFECT__CLOUD__USE_LOCAL_SECRETS
-        value: "false"
-      - name: PREFECT__ENGINE__FLOW_RUNNER__DEFAULT_CLASS
-        value: "prefect.engine.cloud.CloudFlowRunner"
-      - name: PREFECT__ENGINE__TASK_RUNNER__DEFAULT_CLASS
-        value: "prefect.engine.cloud.CloudTaskRunner"
-      - name: PREFECT__ENGINE__EXECUTOR__DEFAULT_CLASS
-        value: "prefect.engine.executors.DaskExecutor"
-      - name: PREFECT__LOGGING__LOG_TO_CLOUD
-        value: "true"
-      - name: PREFECT__LOGGING__LEVEL
-        value: "DEBUG"
-      - name: PREFECT__DEBUG
-        value: "true"
-      - name: PREFECT__LOGGING__EXTRA_LOGGERS
-        value: PREFECT__LOGGING__EXTRA_LOGGERS
-    resources:
-      requests:
-        cpu: "500m"
-      limits:
-        cpu: "500m"
diff --git a/pangeo_forge/tasks/README.md b/pangeo_forge/tasks/README.md
deleted file mode 100644
index 8a77c5e1..00000000
--- a/pangeo_forge/tasks/README.md
+++ /dev/null
@@ -1,8 +0,0 @@
-# `pangeo_forge.tasks`
-
-A collection of Prefect Tasks for Pangeo-Forge.
-
-## Details
-
-- This is simply a collection of Prefect Tasks. Much like https://docs.prefect.io/api/latest/#task-library or https://github.com/PrefectHQ/prefect/tree/master/src/prefect/tasks.
-- This should be considered low-level public API.
diff --git a/pangeo_forge/tasks/__init__.py b/pangeo_forge/tasks/__init__.py
deleted file mode 100644
index e69de29b..00000000
diff --git a/pangeo_forge/tasks/http.py b/pangeo_forge/tasks/http.py
deleted file mode 100644
index 32078d0c..00000000
--- a/pangeo_forge/tasks/http.py
+++ /dev/null
@@ -1,36 +0,0 @@
-import os
-
-import fsspec
-from prefect import task
-
-
-@task
-def download(source_url, cache_location):
-    """
-    Download a remote file to a cache.
-
-    Parameters
-    ----------
-    source_url : str
-        Path or url to the source file.
-    cache_location : str
-        Path or url to the target location for the source file.
-
-    Returns
-    -------
-    target_url : str
-        Path or url in the form of `{cache_location}/hash({source_url})`.
-    """
-    target_url = os.path.join(cache_location, str(hash(source_url)))
-
-    # there is probably a better way to do caching!
-    try:
-        fsspec.open(target_url).open()
-        return target_url
-    except FileNotFoundError:
-        pass
-
-    with fsspec.open(source_url, mode="rb") as source:
-        with fsspec.open(target_url, mode="wb") as target:
-            target.write(source.read())
-    return target_url
diff --git a/pangeo_forge/tasks/xarray.py b/pangeo_forge/tasks/xarray.py
deleted file mode 100644
index 20f65f2c..00000000
--- a/pangeo_forge/tasks/xarray.py
+++ /dev/null
@@ -1,73 +0,0 @@
-from typing import List
-
-import fsspec
-import xarray as xr
-from prefect import task
-
-
-@task
-def combine_and_write(
-    sources: List[str], target: str, append_dim: str, concat_dim: str
-) -> List[str]:
-    """
-    Write a batch of intermediate files to a combined zarr store.
-
-    Parameters
-    ----------
-    sources : List[str]
-        A list of URLs pointing to the intermediate files.
-    target : str
-        The URL for the target combined store.
-    append_dim : str
-        Name of the dimension of which datasets should be appended during write.
-    concat_dim : str
-        The dimension to concatenate along.
-
-    Returns
-    -------
-    target : str
-        The URL of the written combined Zarr store (same as target).
-
-    Examples
-    --------
-    >>> import pangeo_forge.tasks.xarray
-    >>> import fsspec
-    >>> import xarray as xr
-    >>> from prefect import Flow
-
-    >>> # Load sample data into `sources`.
-    >>> ds = xr.tutorial.open_dataset('rasm').load()
-    >>> fs = fsspec.get_filesystem_class("memory")()
-    >>> dsets = ds.isel(time=slice(18)), ds.isel(time=slice(18, None))
-    >>> for i, dset in enumerate(dsets):
-    ...     as_bytes = dset.to_netcdf()
-    ...     with fs.open(f"cache/{i}.nc", "wb") as f:
-    ...         f.write(as_bytes)
-
-    >>> sources = [f"memory://{dset}" for dset in fs.ls("cache")]
-    >>> with Flow("my-flow") as flow:
-    ...    result = pangeo_forge.tasks.xarray.combine_and_write(
-    ...        sources, "memory://target.zarr", concat_dim="time"
-    ...    )
-    >>> result
-    <Task: combine_and_write>
-
-    We can run that outside of a flow context with ``.run()``
-    >>> pangeo_forge.tasks.xarray.combine_and_write.run(
-    ...     sources, "memory://target.zarr", concat_dim="time"
-    ... )
-    'memory://target.zarr'
-    """
-    double_open_files = [fsspec.open(url).open() for url in sources]
-    ds = xr.open_mfdataset(double_open_files, combine="nested", concat_dim=concat_dim)
-    # by definition, this should be a contiguous chunk
-    ds = ds.chunk({append_dim: len(sources)})
-    mapper = fsspec.get_mapper(target)
-
-    if not len(mapper):
-        # The first write, .
-        kwargs = dict(mode="w")
-    else:
-        kwargs = dict(mode="a", append_dim=append_dim)
-    ds.to_zarr(mapper, **kwargs)
-    return target
diff --git a/pangeo_forge/tasks/zarr.py b/pangeo_forge/tasks/zarr.py
deleted file mode 100644
index 9fd60149..00000000
--- a/pangeo_forge/tasks/zarr.py
+++ /dev/null
@@ -1,23 +0,0 @@
-from typing import List, Optional
-
-import fsspec
-import zarr
-from prefect import task
-
-
-@task
-def consolidate_metadata(target, writes: Optional[List[str]] = None) -> None:
-    """
-    Consolidate the metadata the Zarr group at `target`.
-
-    Parameters
-    ----------
-    target : str
-        The URL for the (combined) Zarr group.
-    writes : list of strings, optional
-        The URLs the combined stores were written to. This is only a
-        parameter to introduce a dependency in the pipeline execution graph.
-        The actual value isn't used.
-    """
-    mapper = fsspec.get_mapper(target)
-    zarr.consolidate_metadata(mapper)

From 35e0c9f59d6e7aa64aaff5e4830c1705255d9fc8 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 21:34:11 -0500
Subject: [PATCH 10/34] big cleanup

---
 pangeo_forge/__init__.py             |   5 -
 pangeo_forge/{recipe => }/recipe.py  |   4 +-
 pangeo_forge/recipe/__init__.py      |   0
 pangeo_forge/recipe/pipeline.py      |   0
 pangeo_forge/{recipe => }/storage.py |   0
 tests/fixtures.py                    |   2 +-
 tests/tasks/__init__.py              |   0
 tests/tasks/test_xarray.py           |  36 ------
 tests/test_pipeline.py               | 161 ---------------------------
 tests/test_recipe.py                 |   2 +-
 10 files changed, 4 insertions(+), 206 deletions(-)
 rename pangeo_forge/{recipe => }/recipe.py (99%)
 delete mode 100644 pangeo_forge/recipe/__init__.py
 delete mode 100644 pangeo_forge/recipe/pipeline.py
 rename pangeo_forge/{recipe => }/storage.py (100%)
 delete mode 100644 tests/tasks/__init__.py
 delete mode 100644 tests/tasks/test_xarray.py
 delete mode 100644 tests/test_pipeline.py

diff --git a/pangeo_forge/__init__.py b/pangeo_forge/__init__.py
index 722da05d..30bf1a56 100644
--- a/pangeo_forge/__init__.py
+++ b/pangeo_forge/__init__.py
@@ -9,8 +9,3 @@
     pass
 
 del get_distribution, DistributionNotFound
-
-# 
-# __all__ = [
-#     "AbstractPipeline",
-# ]
diff --git a/pangeo_forge/recipe/recipe.py b/pangeo_forge/recipe.py
similarity index 99%
rename from pangeo_forge/recipe/recipe.py
rename to pangeo_forge/recipe.py
index 2c828500..88a13b1b 100644
--- a/pangeo_forge/recipe/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -12,7 +12,7 @@
 import fsspec
 import zarr
 
-from ..utils import chunked_iterable
+from .utils import chunked_iterable
 from .storage import Target, InputCache
 
 #logger = logging.getLogger(__name__)
@@ -252,7 +252,7 @@ def drop_vars(ds):
             # writing a region means that all the variables MUST have sequence_dim
             to_drop = [v for v in ds.variables
                        if self.sequence_dim not in ds[v].dims]
-            return ds.drop(to_drop)
+            return ds.drop_vars(to_drop)
 
         self.chunk_preprocess_funcs.append(drop_vars)
 
diff --git a/pangeo_forge/recipe/__init__.py b/pangeo_forge/recipe/__init__.py
deleted file mode 100644
index e69de29b..00000000
diff --git a/pangeo_forge/recipe/pipeline.py b/pangeo_forge/recipe/pipeline.py
deleted file mode 100644
index e69de29b..00000000
diff --git a/pangeo_forge/recipe/storage.py b/pangeo_forge/storage.py
similarity index 100%
rename from pangeo_forge/recipe/storage.py
rename to pangeo_forge/storage.py
diff --git a/tests/fixtures.py b/tests/fixtures.py
index 4ac47d9f..1ee9427e 100644
--- a/tests/fixtures.py
+++ b/tests/fixtures.py
@@ -7,7 +7,7 @@
 import pytest
 import xarray as xr
 
-from pangeo_forge.recipe.storage import Target, InputCache
+from pangeo_forge.storage import Target, InputCache
 
 # where to run the http server
 _PORT = "8080"
diff --git a/tests/tasks/__init__.py b/tests/tasks/__init__.py
deleted file mode 100644
index e69de29b..00000000
diff --git a/tests/tasks/test_xarray.py b/tests/tasks/test_xarray.py
deleted file mode 100644
index ae22014a..00000000
--- a/tests/tasks/test_xarray.py
+++ /dev/null
@@ -1,36 +0,0 @@
-import fsspec
-import xarray as xr
-from prefect import Flow, Task
-
-import pangeo_forge.tasks.xarray
-
-
-def test_combine_and_write():
-    ds = xr.tutorial.open_dataset("rasm").load()
-    dsets = ds.isel(time=slice(18)), ds.isel(time=slice(18, None))
-    fs = fsspec.get_filesystem_class("memory")()
-
-    for i, dset in enumerate(dsets):
-        as_bytes = dset.to_netcdf()
-
-        with fs.open(f"cache/{i}.nc", "wb") as f:
-            f.write(as_bytes)
-
-    sources = [f"memory://{dset}" for dset in fs.ls("cache")]
-
-    # In a flow context
-
-    target = "memory://target.zarr"
-    with Flow("test") as flow:
-        result = pangeo_forge.tasks.xarray.combine_and_write(
-            sources, target, concat_dim="time", append_dim="time"
-        )
-        assert isinstance(result, Task)
-    flow.validate()
-
-    result = pangeo_forge.tasks.xarray.combine_and_write.run(
-        sources, target, concat_dim="time", append_dim="time"
-    )
-    assert result == target
-    result = xr.open_zarr(fs.get_mapper("target.zarr"))
-    xr.testing.assert_equal(ds, result)
diff --git a/tests/test_pipeline.py b/tests/test_pipeline.py
deleted file mode 100644
index abfa20d7..00000000
--- a/tests/test_pipeline.py
+++ /dev/null
@@ -1,161 +0,0 @@
-"""Integration testing of pipelines."""
-
-import subprocess
-import time
-
-import fsspec
-import numpy as np
-import pandas as pd
-import pytest
-import xarray as xr
-
-# classes tested here
-from pangeo_forge.pipelines.base import AbstractPipeline
-from pangeo_forge.pipelines.http_xarray_zarr import HttpXarrayZarrMixin
-
-# where to run the http server
-_PORT = "8080"
-_ADDRESS = "127.0.0.1"
-
-
-@pytest.fixture(scope="session")
-def daily_xarray_dataset():
-    """Return a synthetic random xarray dataset."""
-    np.random.seed(1)
-    nt, ny, nx = 10, 18, 36
-    time = pd.date_range(start="2010-01-01", periods=nt, freq="D")
-    lon = (np.arange(nx) + 0.5) * 360 / nx
-    lon_attrs = {"units": "degrees_east", "long_name": "longitude"}
-    lat = (np.arange(ny) + 0.5) * 180 / ny
-    lat_attrs = {"units": "degrees_north", "long_name": "latitude"}
-    foo = np.random.rand(nt, ny, nx)
-    foo_attrs = {"long_name": "Fantastic Foo"}
-    bar = np.random.rand(nt, ny, nx)
-    bar_attrs = {"long_name": "Beautiful Bar"}
-    dims = ("time", "lat", "lon")
-    ds = xr.Dataset(
-        {"foo": (dims, foo, foo_attrs), "bar": (dims, bar, bar_attrs)},
-        coords={
-            "time": ("time", time),
-            "lat": ("lat", lat, lat_attrs),
-            "lon": ("lon", lon, lon_attrs),
-        },
-        attrs={"conventions": "CF 1.6"},
-    )
-    return ds
-
-
-@pytest.fixture(scope="session", params=["D", "2D"])
-def netcdf_local_paths(daily_xarray_dataset, tmpdir_factory, request):
-    """Return a list of paths pointing to netcdf files."""
-    tmp_path = tmpdir_factory.mktemp("netcdf_data")
-    gb = daily_xarray_dataset.resample(time=request.param)
-    _, datasets = zip(*gb)
-    fnames = [f"{n:03d}.nc" for n in range(len(datasets))]
-    paths = [tmp_path.join(fname) for fname in fnames]
-    print(len(paths))
-    xr.save_mfdataset(datasets, [str(path) for path in paths])
-    return paths
-
-
-@pytest.fixture(scope="session")
-def netcdf_http_server(netcdf_local_paths):
-    first_path = netcdf_local_paths[0]
-    # assume that all files are in the same directory
-    basedir = first_path.dirpath()
-    print(basedir)
-    fnames = [path.basename for path in netcdf_local_paths]
-
-    # this feels very hacky
-    command_list = ["python", "-m", "http.server", _PORT, "--bind", _ADDRESS]
-    p = subprocess.Popen(command_list, cwd=basedir)
-    url = f"http://{_ADDRESS}:{_PORT}"
-    time.sleep(0.1)  # let the server start up
-    yield url, fnames
-    p.kill()
-
-
-# tests that our fixtures work
-
-
-def test_fixture_local_files(daily_xarray_dataset, netcdf_local_paths):
-    paths = [str(path) for path in netcdf_local_paths]
-    ds = xr.open_mfdataset(paths, combine="nested", concat_dim="time")
-    assert ds.identical(daily_xarray_dataset)
-
-
-def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
-    url, paths = netcdf_http_server
-    urls = ["/".join([url, str(path)]) for path in paths]
-    open_files = [fsspec.open(url).open() for url in urls]
-    ds = xr.open_mfdataset(open_files, combine="nested", concat_dim="time")
-    assert ds.identical(daily_xarray_dataset)
-
-
-# a pipeline to load that data
-
-
-class MyPipeline(HttpXarrayZarrMixin, AbstractPipeline):
-    repo = "pangeo-forge/pangeo-forge"
-
-    def __init__(
-        self,
-        name,
-        cache_path,
-        target_path,
-        concat_dim,
-        append_dim,
-        files_per_chunk,
-        url_base,
-        nfiles,
-    ):
-        self.name = name
-        self.cache_location = f"{cache_path}/{name}-cache/"
-        self.target_location = f"{target_path}/{name}.zarr"
-        self.append_dim = append_dim
-        self.concat_dim = concat_dim
-        self.files_per_chunk = files_per_chunk
-
-        # needed to build up sources
-        self._url_base = url_base
-        self._nfiles = nfiles
-
-    @property
-    def sources(self):
-        keys = range(self._nfiles)
-        source_url_pattern = self._url_base + "/{n:03d}.nc"
-        source_urls = [source_url_pattern.format(n=key) for key in keys]
-        return source_urls
-
-    @property
-    def targets(self):
-        return [self.target_location]
-
-
-# a basic pipeline test
-def test_pipeline(daily_xarray_dataset, netcdf_http_server, tmpdir):
-    name = "TEST_DATASET"
-    cache_dir = tmpdir.mkdir("cache")
-    target_dir = tmpdir.mkdir("target")
-    concat_dim = "time"
-    append_dim = "time"
-    files_per_chunk = 5
-
-    url_base, paths = netcdf_http_server
-    nfiles = len(paths)
-
-    pipeline = MyPipeline(
-        name, cache_dir, target_dir, concat_dim, append_dim, files_per_chunk, url_base, nfiles
-    )
-    pipeline.flow.run()
-
-    ds_test = xr.open_zarr(pipeline.targets[0])
-    assert ds_test.identical(daily_xarray_dataset)
-
-
-def test_storage_environment_set():
-    pipeline = MyPipeline(
-        "name", "cache", "target", "concat", "append", "files_per_chunk", "url_base", "nfiles"
-    )
-    assert pipeline.storage
-    assert pipeline.environment
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 7b266473..b91ea2c5 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -2,7 +2,7 @@
 import xarray as xr
 import zarr
 
-from pangeo_forge.recipe import recipe
+from pangeo_forge import recipe
 
 from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache
 

From 885634450a2d04c8e6a5ae5ea498ed995d530f24 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 21:36:34 -0500
Subject: [PATCH 11/34] lint and fix tests

---
 .github/workflows/main.yaml |  2 +-
 pangeo_forge/__init__.py    |  2 +-
 pangeo_forge/recipe.py      | 82 ++++++++++++++-----------------------
 pangeo_forge/storage.py     |  5 ++-
 pangeo_forge/utils.py       |  1 +
 5 files changed, 37 insertions(+), 55 deletions(-)

diff --git a/.github/workflows/main.yaml b/.github/workflows/main.yaml
index f77811c3..ef75c257 100644
--- a/.github/workflows/main.yaml
+++ b/.github/workflows/main.yaml
@@ -29,7 +29,7 @@ jobs:
           path: ~/conda_pkgs_dir
           key: ${{ runner.os }}-conda-${{ env.CACHE_NUMBER }}-${{ hashFiles('ci/py${{ matrix.python-version }}.yml') }}
       - name: setup miniconda
-        uses: goanpeca/setup-miniconda@v1
+        uses: conda-incubator/setup-miniconda@v2
         with:
           activate-environment: pangeo-forge
           environment-file: ci/py${{ matrix.python-version }}.yml
diff --git a/pangeo_forge/__init__.py b/pangeo_forge/__init__.py
index 30bf1a56..7e3d3772 100644
--- a/pangeo_forge/__init__.py
+++ b/pangeo_forge/__init__.py
@@ -1,6 +1,6 @@
 from pkg_resources import DistributionNotFound, get_distribution
 
-#from pangeo_forge.pipelines import AbstractPipeline
+# from pangeo_forge.pipelines import AbstractPipeline
 
 try:
     __version__ = get_distribution(__name__).version
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index 88a13b1b..d265be8e 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -3,19 +3,19 @@
 """
 
 import logging
-from dataclasses import dataclass, field
 from contextlib import contextmanager
-from typing import Optional, Iterable, Callable, Any
+from dataclasses import dataclass, field
+from typing import Any, Callable, Iterable, Optional
 
+import fsspec
 import numpy as np
 import xarray as xr
-import fsspec
 import zarr
 
+from .storage import InputCache, Target
 from .utils import chunked_iterable
-from .storage import Target, InputCache
 
-#logger = logging.getLogger(__name__)
+# logger = logging.getLogger(__name__)
 logger = logging.getLogger("recipe")
 
 ### How to manually execute a recipe: ###
@@ -46,7 +46,7 @@
 
 
 @dataclass
-class DatasetRecipe():
+class DatasetRecipe:
     target: Target
     chunk_preprocess_funcs: Iterable[Callable]
 
@@ -54,6 +54,7 @@ class DatasetRecipe():
     def prepare(self):
         def _prepare():
             pass
+
         return _prepare
 
     def iter_inputs(self):
@@ -69,7 +70,7 @@ def iter_inputs(self):
     # this only gets run when iterating, not preparing!
     def preprocess_chunk(self, ds):
         for f in self.chunk_preprocess_funcs:
-             ds = f(ds)
+            ds = f(ds)
         return ds
 
     def iter_chunks(self):
@@ -88,14 +89,16 @@ def iter_chunks(self):
     #         pass
     #     return _finalize
 
+
 # Notes about dataclasses:
 # - https://www.python.org/dev/peps/pep-0557/#inheritance
 # - https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
 # This means that, for now, I can't get default arguments to work.
 
+
 @dataclass
 class FSSpecFileOpenerMixin:
-    #input_open_kwargs: dict #= field(default_factory=dict)
+    # input_open_kwargs: dict #= field(default_factory=dict)
 
     @contextmanager
     def input_opener(self, fname, **kwargs):
@@ -106,7 +109,7 @@ def input_opener(self, fname, **kwargs):
 
 @dataclass
 class InputCachingMixin(FSSpecFileOpenerMixin):
-    require_cache: bool #= False
+    require_cache: bool  # = False
     input_cache: InputCache
 
     # returns a function that takes one input, the input_key
@@ -115,6 +118,7 @@ class InputCachingMixin(FSSpecFileOpenerMixin):
     def cache_input(self):
 
         opener = super().input_opener
+
         def cache_func(fname: str) -> None:
             logger.info(f"Caching input '{fname}'")
             with opener(fname, mode="rb") as source:
@@ -127,7 +131,7 @@ def cache_func(fname: str) -> None:
     def input_opener(self, fname):
         if self.input_cache.exists(fname):
             logger.info(f"Input '{fname}' found in cache")
-            with self.input_cache.open(fname, mode='rb') as f:
+            with self.input_cache.open(fname, mode="rb") as f:
                 yield f
         elif self.require_cache:
             # this creates an error on prepare because nothing is cached
@@ -139,7 +143,6 @@ def input_opener(self, fname):
                 yield f
 
 
-
 @dataclass
 class XarrayInputOpener:
     xarray_open_kwargs: dict
@@ -167,7 +170,7 @@ def open_chunk(self, chunk_key):
         logger.debug(f"{ds}")
 
         # do we really want to just delete all encoding?
-        #for v in ds.variables:
+        # for v in ds.variables:
         #    ds[v].encoding = {}
 
         # TODO: maybe do some chunking here?
@@ -176,10 +179,8 @@ def open_chunk(self, chunk_key):
 
 @dataclass
 class ZarrXarrayWriterMixin:
-
     @property
     def store_chunk(self) -> Callable:
-
         def _store_chunk(chunk_key):
             ds_chunk = self.open_chunk(chunk_key)
             ds_chunk = self.preprocess_chunk(ds_chunk)
@@ -190,26 +191,21 @@ def _store_chunk(chunk_key):
 
         return _store_chunk
 
-
     def open_target(self):
         target_mapper = self.target.get_mapper()
         return xr.open_zarr(target_mapper)
 
-
     def initialize_target(self, ds, **expand_dims):
         logger.info(f"Creating a new dataset in target")
         target_mapper = self.target.get_mapper()
-        ds.to_zarr(target_mapper, mode='w', compute=False)
-
+        ds.to_zarr(target_mapper, mode="w", compute=False)
 
     def expand_target_dim(self, dim, dimsize):
         target_mapper = self.target.get_mapper()
         zgroup = zarr.open_group(target_mapper)
 
         ds = self.open_target()
-        sequence_axes = {v: ds[v].get_axis_num(dim)
-                         for v in ds.variables
-                         if dim in ds[v].dims}
+        sequence_axes = {v: ds[v].get_axis_num(dim) for v in ds.variables if dim in ds[v].dims}
 
         for v, axis in sequence_axes.items():
             arr = zgroup[v]
@@ -218,15 +214,12 @@ def expand_target_dim(self, dim, dimsize):
             arr.resize(shape)
 
 
-
-
 @dataclass
-class ZarrConsolidatorMixin():
-    consolidate_zarr: bool #= True
+class ZarrConsolidatorMixin:
+    consolidate_zarr: bool  # = True
 
     @property
     def finalize(self):
-
         def _finalize():
             if self.consolidate_zarr:
                 logger.info(f"Consolidating Zarr metadata")
@@ -243,62 +236,50 @@ class SequenceRecipe(DatasetRecipe):
     inputs_per_chunk: int = 1
     nitems_per_input: int = 1
 
-
     def __post_init__(self):
-        self._chunks_inputs = {k: v for k, v in
-                              enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))}
+        self._chunks_inputs = {
+            k: v for k, v in enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))
+        }
 
         def drop_vars(ds):
             # writing a region means that all the variables MUST have sequence_dim
-            to_drop = [v for v in ds.variables
-                       if self.sequence_dim not in ds[v].dims]
+            to_drop = [v for v in ds.variables if self.sequence_dim not in ds[v].dims]
             return ds.drop_vars(to_drop)
 
         self.chunk_preprocess_funcs.append(drop_vars)
 
-
     def inputs_for_chunk(self, chunk_key):
         return self._chunks_inputs[chunk_key]
 
-
     def iter_inputs(self):
         for chunk_key in self.iter_chunks():
             for input in self.inputs_for_chunk(chunk_key):
                 yield input
 
-
     def nitems_for_chunk(self, chunk_key):
         return self.nitems_per_input * len(self.inputs_for_chunk(chunk_key))
 
-
     def region_for_chunk(self, chunk_key):
         # return a dict suitable to pass to xr.to_zarr(region=...)
         # specifies where in the overall array to put this chunk's data
         stride = self.nitems_per_input * self.inputs_per_chunk
         start = chunk_key * stride
-        return {
-            self.sequence_dim:
-                slice(start, start + self.nitems_for_chunk(chunk_key))
-        }
-
+        return {self.sequence_dim: slice(start, start + self.nitems_for_chunk(chunk_key))}
 
     def sequence_len(self):
         # tells the total size of dataset along the sequence dimension
         return sum([self.nitems_for_chunk(k) for k in self.iter_chunks()])
 
-
     def sequence_chunks(self):
         # chunking
         return {self.sequence_dim: self.inputs_per_chunk * self.nitems_per_input}
 
-
     def iter_chunks(self):
         for k in self._chunks_inputs:
             yield k
 
     @property
     def prepare(self):
-
         def _prepare():
 
             target_store = self.target.get_mapper()
@@ -313,7 +294,7 @@ def _prepare():
 
                 # make sure the concat dim has a valid fill_value to avoid
                 # overruns when writing chunk
-                ds[self.sequence_dim].encoding = {'_FillValue': -1}
+                ds[self.sequence_dim].encoding = {"_FillValue": -1}
                 # actually not necessary if we use decode_times=False
                 self.initialize_target(ds)
 
@@ -324,12 +305,12 @@ def _prepare():
 
 @dataclass
 class StandardSequentialRecipe(
-        SequenceRecipe,
-        InputCachingMixin,
-        XarrayConcatChunkOpener,
-        ZarrXarrayWriterMixin,
-        ZarrConsolidatorMixin
-    ):
+    SequenceRecipe,
+    InputCachingMixin,
+    XarrayConcatChunkOpener,
+    ZarrXarrayWriterMixin,
+    ZarrConsolidatorMixin,
+):
     pass
 
 
@@ -338,7 +319,6 @@ class StandardSequentialRecipe(
 # only needed because of
 # https://github.com/pydata/xarray/issues/4631
 def _fix_scalar_attr_encoding(ds):
-
     def _fixed_attrs(d):
         fixed = {}
         for k, v in d.items():
diff --git a/pangeo_forge/storage.py b/pangeo_forge/storage.py
index 04e949a0..b3777097 100644
--- a/pangeo_forge/storage.py
+++ b/pangeo_forge/storage.py
@@ -1,7 +1,8 @@
-from dataclasses import dataclass
+import os
 from contextlib import closing, contextmanager
+from dataclasses import dataclass
+
 import fsspec
-import os
 
 
 @dataclass
diff --git a/pangeo_forge/utils.py b/pangeo_forge/utils.py
index c49b3014..c865e521 100644
--- a/pangeo_forge/utils.py
+++ b/pangeo_forge/utils.py
@@ -1,5 +1,6 @@
 import itertools
 
+
 # https://alexwlchan.net/2018/12/iterating-in-fixed-size-chunks/
 def chunked_iterable(iterable, size):
     it = iter(iterable)

From b3a42ed950716775a391d4030396a3dc2a327786 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 21:45:22 -0500
Subject: [PATCH 12/34] update requirements

---
 ci/py3.7.yml     | 3 ++-
 ci/py3.8.yml     | 3 ++-
 requirements.txt | 4 ++--
 3 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/ci/py3.7.yml b/ci/py3.7.yml
index 8d5f0961..c310ff4c 100644
--- a/ci/py3.7.yml
+++ b/ci/py3.7.yml
@@ -25,4 +25,5 @@ dependencies:
   - scipy
   - setuptools
   - toolz
-  - zarr
+  - xarray>=0.16.2
+  - zarr>=2.6.0
diff --git a/ci/py3.8.yml b/ci/py3.8.yml
index ada7aae7..739ce1ef 100644
--- a/ci/py3.8.yml
+++ b/ci/py3.8.yml
@@ -25,4 +25,5 @@ dependencies:
   - scipy
   - setuptools
   - toolz
-  - zarr
+  - xarray>=0.16.2
+  - zarr>=2.6.0
diff --git a/requirements.txt b/requirements.txt
index dbf84c5f..11a36cbe 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,6 +2,6 @@ setuptools
 click
 dask
 distributed
-xarray
-zarr
+xarray >= 0.16.2
+zarr >= 2.6.0
 fsspec[http]

From ec95d9d7bb23606ecc9f6f5bb90568286fca096d Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 1 Dec 2020 21:51:07 -0500
Subject: [PATCH 13/34] more linting

---
 tests/fixtures.py    | 32 +++++++++++++++++---------------
 tests/test_recipe.py | 21 +++++++++++++++++----
 tests/test_utils.py  | 26 +++++++++++++++++++++++---
 3 files changed, 57 insertions(+), 22 deletions(-)

diff --git a/tests/fixtures.py b/tests/fixtures.py
index 1ee9427e..5ac6738d 100644
--- a/tests/fixtures.py
+++ b/tests/fixtures.py
@@ -7,7 +7,7 @@
 import pytest
 import xarray as xr
 
-from pangeo_forge.storage import Target, InputCache
+from pangeo_forge.storage import InputCache, Target
 
 # where to run the http server
 _PORT = "8080"
@@ -46,7 +46,7 @@ def netcdf_local_paths(daily_xarray_dataset, tmpdir_factory, request):
     """Return a list of paths pointing to netcdf files."""
     tmp_path = tmpdir_factory.mktemp("netcdf_data")
     items_per_file = {"D": 1, "2D": 2}
-    daily_xarray_dataset.attrs['items_per_file'] = items_per_file[request.param]
+    daily_xarray_dataset.attrs["items_per_file"] = items_per_file[request.param]
     gb = daily_xarray_dataset.resample(time=request.param)
     _, datasets = zip(*gb)
     fnames = [f"{n:03d}.nc" for n in range(len(datasets))]
@@ -76,6 +76,7 @@ def netcdf_http_server(netcdf_local_paths):
 @pytest.fixture()
 def tmp_target(tmpdir_factory):
     import fsspec
+
     fs = fsspec.get_filesystem_class("file")()
     path = str(tmpdir_factory.mktemp("target"))
     return Target(fs, path)
@@ -85,9 +86,10 @@ def tmp_target(tmpdir_factory):
 def tmp_cache(tmpdir_factory):
     path = str(tmpdir_factory.mktemp("cache"))
     fs = fsspec.get_filesystem_class("file")()
-    cache = InputCache(fs, prefix='cache')
+    cache = InputCache(fs, prefix="cache")
     return cache
 
+
 # tests that our fixtures work
 
 
@@ -107,19 +109,19 @@ def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
 
 def test_target(tmp_target):
     mapper = tmp_target.get_mapper()
-    mybytes = b'bar'
-    mapper['foo'] = b'bar'
-    with open(tmp_target.path + '/foo') as f:
+    mybytes = b"bar"
+    mapper["foo"] = b"bar"
+    with open(tmp_target.path + "/foo") as f:
         res = f.read()
-    assert res == 'bar'
+    assert res == "bar"
 
 
 def test_cache(tmp_cache):
-    assert not tmp_cache.exists('foo')
-    with tmp_cache.open('foo', mode='w') as f:
-        f.write('bar')
-    assert tmp_cache.exists('foo')
-    with tmp_cache.open('foo', mode='r') as f:
-        assert f.read() == 'bar'
-    tmp_cache.rm('foo')
-    assert not tmp_cache.exists('foo')
+    assert not tmp_cache.exists("foo")
+    with tmp_cache.open("foo", mode="w") as f:
+        f.write("bar")
+    assert tmp_cache.exists("foo")
+    with tmp_cache.open("foo", mode="r") as f:
+        assert f.read() == "bar"
+    tmp_cache.rm("foo")
+    assert not tmp_cache.exists("foo")
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index b91ea2c5..d55b4710 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -4,15 +4,28 @@
 
 from pangeo_forge import recipe
 
-from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache
+from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_cache, tmp_target
 
 dummy_fnames = ["a.nc", "b.nc", "c.nc"]
+
+
 @pytest.mark.parametrize(
     "file_urls, files_per_chunk, expected_keys, expected_filenames",
     [
         (dummy_fnames, 1, [0, 1, 2], [("a.nc",), ("b.nc",), ("c.nc",)]),
-        (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",)])
-    ]
+        (
+            dummy_fnames,
+            2,
+            [0, 1],
+            [
+                (
+                    "a.nc",
+                    "b.nc",
+                ),
+                ("c.nc",),
+            ],
+        ),
+    ],
 )
 def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames, tmp_target):
 
@@ -47,7 +60,7 @@ def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_c
         input_urls=netcdf_local_paths,
         sequence_dim="time",
         inputs_per_chunk=1,
-        nitems_per_input=daily_xarray_dataset.attrs['items_per_file']
+        nitems_per_input=daily_xarray_dataset.attrs["items_per_file"],
     )
 
     # this is the cannonical way to manually execute a recipe
diff --git a/tests/test_utils.py b/tests/test_utils.py
index 03e6be7f..0d9b8ee8 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -8,9 +8,29 @@
     [
         ([1, 2, 3], 1, [(1,), (2,), (3,)]),
         ([1, 2, 3], 2, [(1, 2), (3,)]),
-        ([1, 2, 3], 3, [(1, 2, 3,)]),
-        ([1, 2, 3], 4, [(1, 2, 3,)])
-    ]
+        (
+            [1, 2, 3],
+            3,
+            [
+                (
+                    1,
+                    2,
+                    3,
+                )
+            ],
+        ),
+        (
+            [1, 2, 3],
+            4,
+            [
+                (
+                    1,
+                    2,
+                    3,
+                )
+            ],
+        ),
+    ],
 )
 def test_chunked_iterable(iterable, size, expected):
     actual = list(pangeo_forge.utils.chunked_iterable(iterable, size))

From 79ab04fcdcbda4041cdda9d39043a679014d0dfd Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Wed, 2 Dec 2020 14:00:24 -0500
Subject: [PATCH 14/34] added executors

---
 pangeo_forge/executors/__init__.py |  0
 pangeo_forge/executors/prefect.py  | 38 ++++++++++++++++++++++++++++++
 pangeo_forge/executors/python.py   | 29 +++++++++++++++++++++++
 3 files changed, 67 insertions(+)
 create mode 100644 pangeo_forge/executors/__init__.py
 create mode 100644 pangeo_forge/executors/prefect.py
 create mode 100644 pangeo_forge/executors/python.py

diff --git a/pangeo_forge/executors/__init__.py b/pangeo_forge/executors/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/pangeo_forge/executors/prefect.py b/pangeo_forge/executors/prefect.py
new file mode 100644
index 00000000..cea2c367
--- /dev/null
+++ b/pangeo_forge/executors/prefect.py
@@ -0,0 +1,38 @@
+from typing import Any, Iterable, Tuple
+
+import prefect
+
+from ..recipe import DatasetRecipe
+
+class PrefectExecutor():
+
+    def prepare_plan(self, r: DatasetRecipe) -> prefect.Flow:
+
+        # wrap our functions as prefect tasks
+        @prefect.task
+        def prepare() -> None:
+            r.prepare()
+
+        @prefect.task
+        def cache_input(input: Any) -> None:
+            r.cache_input(input)
+
+        @prefect.task
+        def store_chunk(input: Any) -> None:
+            r.store_chunk(input)
+
+        @prefect.task
+        def finalize() -> None:
+            r.finalize()
+
+        with prefect.Flow("Pangeo-Forge") as flow:
+            prepare()
+            cache_input.map(list(r.iter_inputs()))
+            store_chunk.map(list(r.iter_chunks()))
+            finalize()
+
+        return flow
+
+
+    def execute_plan(self, plan: prefect.Flow, **kwargs):
+        return plan.run(**kwargs)
diff --git a/pangeo_forge/executors/python.py b/pangeo_forge/executors/python.py
new file mode 100644
index 00000000..4be76073
--- /dev/null
+++ b/pangeo_forge/executors/python.py
@@ -0,0 +1,29 @@
+from functools import partial
+
+#from ..types import Pipeline, Stage, Task
+from ..recipe import DatasetRecipe
+
+from typing import Callable, Iterable
+
+Task = Callable[[], None]
+
+class PythonExecutor:
+
+    def prepare_plan(self, r: DatasetRecipe) -> Task:
+        tasks = []
+        tasks.append(r.prepare)
+        for input in r.iter_inputs():
+            tasks.append(partial(r.cache_input, input))
+        for chunk in r.iter_chunks():
+            tasks.append(partial(r.store_chunk, chunk))
+        tasks.append(r.finalize)
+
+        return partial(_execute_all, tasks)
+
+    def execute_plan(self, plan: Task, **kwargs):
+        plan()
+
+
+def _execute_all(tasks: Iterable[Task]) -> None:
+    for task in tasks:
+        task()

From e6b32a98282e823a73c7f4f8b718063a158a1615 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 3 Dec 2020 20:41:35 -0500
Subject: [PATCH 15/34] linting and stuff

---
 pangeo_forge/executors/prefect.py  |  5 ++-
 pangeo_forge/executors/python.py   |  7 ++---
 pangeo_forge/recipe.py             | 50 ++++++++++++------------------
 pangeo_forge/storage.py            |  2 +-
 pangeo_forge/utils.py              | 21 +++++++++++++
 tests/{fixtures.py => conftest.py} | 39 +----------------------
 tests/test_recipe.py               |  3 --
 7 files changed, 47 insertions(+), 80 deletions(-)
 rename tests/{fixtures.py => conftest.py} (69%)

diff --git a/pangeo_forge/executors/prefect.py b/pangeo_forge/executors/prefect.py
index cea2c367..afe13b8a 100644
--- a/pangeo_forge/executors/prefect.py
+++ b/pangeo_forge/executors/prefect.py
@@ -1,11 +1,11 @@
-from typing import Any, Iterable, Tuple
+from typing import Any
 
 import prefect
 
 from ..recipe import DatasetRecipe
 
-class PrefectExecutor():
 
+class PrefectExecutor:
     def prepare_plan(self, r: DatasetRecipe) -> prefect.Flow:
 
         # wrap our functions as prefect tasks
@@ -33,6 +33,5 @@ def finalize() -> None:
 
         return flow
 
-
     def execute_plan(self, plan: prefect.Flow, **kwargs):
         return plan.run(**kwargs)
diff --git a/pangeo_forge/executors/python.py b/pangeo_forge/executors/python.py
index 4be76073..5e1d1c64 100644
--- a/pangeo_forge/executors/python.py
+++ b/pangeo_forge/executors/python.py
@@ -1,14 +1,13 @@
 from functools import partial
+from typing import Callable, Iterable
 
-#from ..types import Pipeline, Stage, Task
+# from ..types import Pipeline, Stage, Task
 from ..recipe import DatasetRecipe
 
-from typing import Callable, Iterable
-
 Task = Callable[[], None]
 
-class PythonExecutor:
 
+class PythonExecutor:
     def prepare_plan(self, r: DatasetRecipe) -> Task:
         tasks = []
         tasks.append(r.prepare)
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index d265be8e..e651e834 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -4,21 +4,20 @@
 
 import logging
 from contextlib import contextmanager
-from dataclasses import dataclass, field
-from typing import Any, Callable, Iterable, Optional
+from dataclasses import dataclass
+from typing import Callable, Iterable
 
 import fsspec
-import numpy as np
 import xarray as xr
 import zarr
 
 from .storage import InputCache, Target
-from .utils import chunked_iterable
+from .utils import chunked_iterable, fix_scalar_attr_encoding
 
 # logger = logging.getLogger(__name__)
 logger = logging.getLogger("recipe")
 
-### How to manually execute a recipe: ###
+# How to manually execute a recipe: ###
 #
 #   t = PangeoForgeTarget()
 #   r = MyRecipe(target=t, **opts) # 1
@@ -29,8 +28,6 @@
 #   for chunk_key in r.iter_chunks():
 #       r.store_chunk(chunk_key) # 5
 #   r.finalize() # 6
-#
-###
 
 
 # 1) Initialize the Recipe object
@@ -152,7 +149,7 @@ def open_input(self, fname):
             logger.info(f"Opening input with Xarray '{fname}'")
             ds = xr.open_dataset(f, **self.xarray_open_kwargs).load()
             # do we always want to remove encoding? I think so.
-        ds = _fix_scalar_attr_encoding(ds)
+        ds = fix_scalar_attr_encoding(ds)
         logger.debug(f"{ds}")
         return ds
 
@@ -196,7 +193,7 @@ def open_target(self):
         return xr.open_zarr(target_mapper)
 
     def initialize_target(self, ds, **expand_dims):
-        logger.info(f"Creating a new dataset in target")
+        logger.info("Creating a new dataset in target")
         target_mapper = self.target.get_mapper()
         ds.to_zarr(target_mapper, mode="w", compute=False)
 
@@ -222,7 +219,7 @@ class ZarrConsolidatorMixin:
     def finalize(self):
         def _finalize():
             if self.consolidate_zarr:
-                logger.info(f"Consolidating Zarr metadata")
+                logger.info("Consolidating Zarr metadata")
                 target_mapper = self.target.get_mapper()
                 zarr.consolidate_metadata(target_mapper)
 
@@ -231,10 +228,21 @@ def _finalize():
 
 @dataclass
 class SequenceRecipe(DatasetRecipe):
+    """There are many inputs (a.k.a. files, granules), arranged in a sequence
+    along the dimension `sequence_dim`. Each file may contain multiple variables.
+    """
+
     input_urls: Iterable[str]
+    """The inputs used to generate the dataset."""
+
     sequence_dim: str
+    """The dimension name along which the inputs will be concatenated."""
+
     inputs_per_chunk: int = 1
+    """The number of inputs to use in each chunk."""
+
     nitems_per_input: int = 1
+    """The length of each input along the `sequence_dim` dimension."""
 
     def __post_init__(self):
         self._chunks_inputs = {
@@ -282,11 +290,9 @@ def iter_chunks(self):
     def prepare(self):
         def _prepare():
 
-            target_store = self.target.get_mapper()
-
             try:
                 ds = self.open_target()
-                logger.info(f"Found an existing dataset in target")
+                logger.info("Found an existing dataset in target")
                 logger.debug(f"{ds}")
             except (IOError, zarr.errors.GroupNotFoundError):
                 first_chunk_key = next(self.iter_chunks())
@@ -315,21 +321,3 @@ class StandardSequentialRecipe(
 
 
 # helper utilities
-
-# only needed because of
-# https://github.com/pydata/xarray/issues/4631
-def _fix_scalar_attr_encoding(ds):
-    def _fixed_attrs(d):
-        fixed = {}
-        for k, v in d.items():
-            if isinstance(v, np.ndarray) and len(v) == 1:
-                fixed[k] = v[0]
-        return fixed
-
-    ds = ds.copy()
-    ds.attrs.update(_fixed_attrs(ds.attrs))
-    ds.encoding.update(_fixed_attrs(ds.encoding))
-    for v in ds.variables:
-        ds[v].attrs.update(_fixed_attrs(ds[v].attrs))
-        ds[v].encoding.update(_fixed_attrs(ds[v].encoding))
-    return ds
diff --git a/pangeo_forge/storage.py b/pangeo_forge/storage.py
index b3777097..742bb9f2 100644
--- a/pangeo_forge/storage.py
+++ b/pangeo_forge/storage.py
@@ -1,5 +1,5 @@
 import os
-from contextlib import closing, contextmanager
+from contextlib import contextmanager
 from dataclasses import dataclass
 
 import fsspec
diff --git a/pangeo_forge/utils.py b/pangeo_forge/utils.py
index c865e521..1ac84c7b 100644
--- a/pangeo_forge/utils.py
+++ b/pangeo_forge/utils.py
@@ -1,5 +1,7 @@
 import itertools
 
+import numpy as np
+
 
 # https://alexwlchan.net/2018/12/iterating-in-fixed-size-chunks/
 def chunked_iterable(iterable, size):
@@ -9,3 +11,22 @@ def chunked_iterable(iterable, size):
         if not chunk:
             break
         yield chunk
+
+
+# only needed because of
+# https://github.com/pydata/xarray/issues/4631
+def fix_scalar_attr_encoding(ds):
+    def _fixed_attrs(d):
+        fixed = {}
+        for k, v in d.items():
+            if isinstance(v, np.ndarray) and len(v) == 1:
+                fixed[k] = v[0]
+        return fixed
+
+    ds = ds.copy()
+    ds.attrs.update(_fixed_attrs(ds.attrs))
+    ds.encoding.update(_fixed_attrs(ds.encoding))
+    for v in ds.variables:
+        ds[v].attrs.update(_fixed_attrs(ds[v].attrs))
+        ds[v].encoding.update(_fixed_attrs(ds[v].encoding))
+    return ds
diff --git a/tests/fixtures.py b/tests/conftest.py
similarity index 69%
rename from tests/fixtures.py
rename to tests/conftest.py
index 5ac6738d..c2584809 100644
--- a/tests/fixtures.py
+++ b/tests/conftest.py
@@ -86,42 +86,5 @@ def tmp_target(tmpdir_factory):
 def tmp_cache(tmpdir_factory):
     path = str(tmpdir_factory.mktemp("cache"))
     fs = fsspec.get_filesystem_class("file")()
-    cache = InputCache(fs, prefix="cache")
+    cache = InputCache(fs, prefix=path)
     return cache
-
-
-# tests that our fixtures work
-
-
-def test_fixture_local_files(daily_xarray_dataset, netcdf_local_paths):
-    paths = [str(path) for path in netcdf_local_paths]
-    ds = xr.open_mfdataset(paths, combine="nested", concat_dim="time")
-    assert ds.identical(daily_xarray_dataset)
-
-
-def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
-    url, paths = netcdf_http_server
-    urls = ["/".join([url, str(path)]) for path in paths]
-    open_files = [fsspec.open(url).open() for url in urls]
-    ds = xr.open_mfdataset(open_files, combine="nested", concat_dim="time")
-    assert ds.identical(daily_xarray_dataset)
-
-
-def test_target(tmp_target):
-    mapper = tmp_target.get_mapper()
-    mybytes = b"bar"
-    mapper["foo"] = b"bar"
-    with open(tmp_target.path + "/foo") as f:
-        res = f.read()
-    assert res == "bar"
-
-
-def test_cache(tmp_cache):
-    assert not tmp_cache.exists("foo")
-    with tmp_cache.open("foo", mode="w") as f:
-        f.write("bar")
-    assert tmp_cache.exists("foo")
-    with tmp_cache.open("foo", mode="r") as f:
-        assert f.read() == "bar"
-    tmp_cache.rm("foo")
-    assert not tmp_cache.exists("foo")
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index d55b4710..28e58bc3 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -1,11 +1,8 @@
 import pytest
 import xarray as xr
-import zarr
 
 from pangeo_forge import recipe
 
-from .fixtures import daily_xarray_dataset, netcdf_local_paths, tmp_cache, tmp_target
-
 dummy_fnames = ["a.nc", "b.nc", "c.nc"]
 
 

From b993dabd5f3db6fbc86826cb18a6372f120e4a20 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 3 Dec 2020 21:19:47 -0500
Subject: [PATCH 16/34] testing executors

---
 pangeo_forge/executors/__init__.py |  8 ++++++
 pangeo_forge/recipe.py             |  3 ---
 setup.cfg                          |  3 +--
 tests/conftest.py                  | 19 ++++++++++++++
 tests/test_executors.py            | 14 +++++++++++
 tests/test_fixtures.py             | 40 ++++++++++++++++++++++++++++++
 tests/test_recipe.py               |  2 ++
 7 files changed, 84 insertions(+), 5 deletions(-)
 create mode 100644 tests/test_executors.py
 create mode 100644 tests/test_fixtures.py

diff --git a/pangeo_forge/executors/__init__.py b/pangeo_forge/executors/__init__.py
index e69de29b..e0aa6809 100644
--- a/pangeo_forge/executors/__init__.py
+++ b/pangeo_forge/executors/__init__.py
@@ -0,0 +1,8 @@
+"""
+Executors know how to run recipes.
+"""
+
+from .prefect import PrefectExecutor
+from .python import PythonExecutor
+
+__all__ = [PythonExecutor, PrefectExecutor]
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index e651e834..d0a6a6ab 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -318,6 +318,3 @@ class StandardSequentialRecipe(
     ZarrConsolidatorMixin,
 ):
     pass
-
-
-# helper utilities
diff --git a/setup.cfg b/setup.cfg
index 7468400a..9c7719f5 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -11,5 +11,4 @@ combine_as_imports=True
 line_length=100
 
 [tool:pytest]
-log_cli = True
-log_level = INFO
+log_cli = False
diff --git a/tests/conftest.py b/tests/conftest.py
index c2584809..6389e837 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -7,6 +7,7 @@
 import pytest
 import xarray as xr
 
+from pangeo_forge import recipe
 from pangeo_forge.storage import InputCache, Target
 
 # where to run the http server
@@ -88,3 +89,21 @@ def tmp_cache(tmpdir_factory):
     fs = fsspec.get_filesystem_class("file")()
     cache = InputCache(fs, prefix=path)
     return cache
+
+
+@pytest.fixture
+def sequential_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
+    r = recipe.StandardSequentialRecipe(
+        consolidate_zarr=True,
+        xarray_open_kwargs={},
+        xarray_concat_kwargs={},
+        require_cache=False,
+        input_cache=tmp_cache,
+        target=tmp_target,
+        chunk_preprocess_funcs=[],
+        input_urls=netcdf_local_paths,
+        sequence_dim="time",
+        inputs_per_chunk=1,
+        nitems_per_input=daily_xarray_dataset.attrs["items_per_file"],
+    )
+    return r, daily_xarray_dataset, tmp_target
diff --git a/tests/test_executors.py b/tests/test_executors.py
new file mode 100644
index 00000000..eb90e2a0
--- /dev/null
+++ b/tests/test_executors.py
@@ -0,0 +1,14 @@
+import pytest
+import xarray as xr
+
+from pangeo_forge.executors import PrefectExecutor, PythonExecutor
+
+
+@pytest.mark.parametrize("Executor", [PythonExecutor, PrefectExecutor])
+def test_recipe_w_executor(Executor, sequential_recipe):
+    rec, ds_expected, target = sequential_recipe
+    ex = Executor()
+    plan = ex.prepare_plan(rec)
+    ex.execute_plan(plan)
+    ds_actual = xr.open_zarr(target.get_mapper()).load()
+    assert ds_actual.identical(ds_expected)
diff --git a/tests/test_fixtures.py b/tests/test_fixtures.py
new file mode 100644
index 00000000..c867f17b
--- /dev/null
+++ b/tests/test_fixtures.py
@@ -0,0 +1,40 @@
+import fsspec
+import xarray as xr
+
+from pangeo_forge.utils import fix_scalar_attr_encoding
+
+
+def test_fixture_local_files(daily_xarray_dataset, netcdf_local_paths):
+    paths = [str(path) for path in netcdf_local_paths]
+    ds = xr.open_mfdataset(paths, combine="nested", concat_dim="time")
+    assert ds.identical(daily_xarray_dataset)
+
+
+def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
+    url, paths = netcdf_http_server
+    urls = ["/".join([url, str(path)]) for path in paths]
+    open_files = [fsspec.open(url).open() for url in urls]
+    ds = xr.open_mfdataset(open_files, combine="nested", concat_dim="time").load()
+    ds = fix_scalar_attr_encoding(ds)
+    print(ds)
+    print(daily_xarray_dataset)
+    assert ds.identical(daily_xarray_dataset)
+
+
+def test_target(tmp_target):
+    mapper = tmp_target.get_mapper()
+    mapper["foo"] = b"bar"
+    with open(tmp_target.path + "/foo") as f:
+        res = f.read()
+    assert res == "bar"
+
+
+def test_cache(tmp_cache):
+    assert not tmp_cache.exists("foo")
+    with tmp_cache.open("foo", mode="w") as f:
+        f.write("bar")
+    assert tmp_cache.exists("foo")
+    with tmp_cache.open("foo", mode="r") as f:
+        assert f.read() == "bar"
+    tmp_cache.rm("foo")
+    assert not tmp_cache.exists("foo")
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 28e58bc3..11675e17 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -46,6 +46,8 @@ def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_fil
 
 def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
 
+    # the same recipe is created as a fixture in conftest.py
+    # I left it here explicitly because it makes the test easier to read.
     r = recipe.StandardSequentialRecipe(
         consolidate_zarr=True,
         xarray_open_kwargs={},

From 0e150cb6a5c1291799c7588f69ca209cb1b7b6e9 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 21 Dec 2020 11:45:45 -0500
Subject: [PATCH 17/34] major simplification of recipe class

---
 pangeo_forge/executors/prefect.py |  15 +-
 pangeo_forge/executors/python.py  |   6 +-
 pangeo_forge/recipe.py            | 302 ++++++++++++------------------
 tests/conftest.py                 |  13 +-
 tests/test_executors.py           |   4 +-
 tests/test_recipe.py              |  14 +-
 6 files changed, 140 insertions(+), 214 deletions(-)

diff --git a/pangeo_forge/executors/prefect.py b/pangeo_forge/executors/prefect.py
index afe13b8a..ce91017e 100644
--- a/pangeo_forge/executors/prefect.py
+++ b/pangeo_forge/executors/prefect.py
@@ -2,11 +2,9 @@
 
 import prefect
 
-from ..recipe import DatasetRecipe
-
 
 class PrefectExecutor:
-    def prepare_plan(self, r: DatasetRecipe) -> prefect.Flow:
+    def prepare_plan(self, r) -> prefect.Flow:
 
         # wrap our functions as prefect tasks
         @prefect.task
@@ -26,10 +24,13 @@ def finalize() -> None:
             r.finalize()
 
         with prefect.Flow("Pangeo-Forge") as flow:
-            prepare()
-            cache_input.map(list(r.iter_inputs()))
-            store_chunk.map(list(r.iter_chunks()))
-            finalize()
+            prep_task = prepare()
+            cache_task = cache_input.map(list(r.iter_inputs()))
+            cache_task.set_dependencies(upstream_tasks=[prep_task])
+            store_task = store_chunk.map(list(r.iter_chunks()))
+            store_task.set_dependencies(upstream_tasks=[cache_task])
+            finalize_task = finalize()
+            finalize_task.set_dependencies(upstream_tasks=[store_task])
 
         return flow
 
diff --git a/pangeo_forge/executors/python.py b/pangeo_forge/executors/python.py
index 5e1d1c64..0661d153 100644
--- a/pangeo_forge/executors/python.py
+++ b/pangeo_forge/executors/python.py
@@ -1,14 +1,10 @@
 from functools import partial
 from typing import Callable, Iterable
 
-# from ..types import Pipeline, Stage, Task
-from ..recipe import DatasetRecipe
-
 Task = Callable[[], None]
 
-
 class PythonExecutor:
-    def prepare_plan(self, r: DatasetRecipe) -> Task:
+    def prepare_plan(self, r):
         tasks = []
         tasks.append(r.prepare)
         for input in r.iter_inputs():
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index d0a6a6ab..af389317 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -4,8 +4,8 @@
 
 import logging
 from contextlib import contextmanager
-from dataclasses import dataclass
-from typing import Callable, Iterable
+from dataclasses import dataclass, field
+from typing import Callable, Iterable, Optional
 
 import fsspec
 import xarray as xr
@@ -14,8 +14,7 @@
 from .storage import InputCache, Target
 from .utils import chunked_iterable, fix_scalar_attr_encoding
 
-# logger = logging.getLogger(__name__)
-logger = logging.getLogger("recipe")
+logger = logging.getLogger(__name__)
 
 # How to manually execute a recipe: ###
 #
@@ -41,123 +40,159 @@
 #    Might be coming from the cache or might be read directly.
 # 6)
 
+@contextmanager
+def input_opener(fname, **kwargs):
+    logger.info(f"Opening input '{fname}'")
+    with fsspec.open(fname, **kwargs) as f:
+        yield f
+
+# Notes about dataclasses:
+# - https://www.python.org/dev/peps/pep-0557/#inheritance
+# - https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
+# This means that, for now, I can't get default arguments to work.
 
 @dataclass
-class DatasetRecipe:
-    target: Target
-    chunk_preprocess_funcs: Iterable[Callable]
+class NetCDFtoZarrSequentialRecipe:
+    """There are many inputs (a.k.a. files, granules), arranged in a sequence
+    along the dimension `sequence_dim`. Each file may contain multiple variables.
+    """
+    input_urls: Iterable[str]
+    """The inputs used to generate the dataset."""
 
-    @property
-    def prepare(self):
-        def _prepare():
-            pass
+    sequence_dim: str
+    """The dimension name along which the inputs will be concatenated."""
 
-        return _prepare
+    inputs_per_chunk: int = 1
+    """The number of inputs to use in each chunk."""
 
-    def iter_inputs(self):
-        return []
-
-    # need to figure out what's going on with these methods and inheritance
-    # @property
-    # def cache_input(self):
-    #     def _cache_input(input_key):
-    #         raise NotImplementedError
-    #     return _cache_input
-
-    # this only gets run when iterating, not preparing!
-    def preprocess_chunk(self, ds):
-        for f in self.chunk_preprocess_funcs:
-            ds = f(ds)
-        return ds
+    nitems_per_input: int = 1
+    """The length of each input along the `sequence_dim` dimension."""
 
-    def iter_chunks(self):
-        raise NotImplementedError
+    target: Optional[Target] = None
+    """A location in which to put the dataset."""
 
-    # @property
-    # def store_chunk(self):
-    #     def _store_chunk(chunk_key):
-    #         raise NotImplementedError
-    #     return _store_chunk
+    input_cache: Optional[InputCache] = None
+    """The length of each input along the `sequence_dim` dimension."""
 
-    # @property
-    # def finalize(self):
-    #
-    #     def _finalize():
-    #         pass
-    #     return _finalize
+    consolidate_zarr: bool = True
+    """Whether to consolidate the resulting Zarr dataset."""
 
+    xarray_open_kwargs: dict = field(default_factory=dict)
+    """Extra options for opening the inputs with Xarray."""
 
-# Notes about dataclasses:
-# - https://www.python.org/dev/peps/pep-0557/#inheritance
-# - https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
-# This means that, for now, I can't get default arguments to work.
+    xarray_concat_kwargs: dict = field(default_factory=dict)
+    """Extra options to pass to Xarray when concatenating the inputs to form a chunk."""
 
+    def __post_init__(self):
+        self._chunks_inputs = {
+            k: v for k, v in enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))
+        }
 
-@dataclass
-class FSSpecFileOpenerMixin:
-    # input_open_kwargs: dict #= field(default_factory=dict)
+    @property
+    def prepare(self) -> Callable:
+        """Prepare target for storing dataset."""
 
-    @contextmanager
-    def input_opener(self, fname, **kwargs):
-        logger.info(f"Opening input '{fname}'")
-        with fsspec.open(fname, **kwargs) as f:
-            yield f
+        def _prepare():
 
+            try:
+                ds = self.open_target()
+                logger.info("Found an existing dataset in target")
+                logger.debug(f"{ds}")
+            except (IOError, zarr.errors.GroupNotFoundError):
+                first_chunk_key = next(self.iter_chunks())
+                for input_url in self.inputs_for_chunk(first_chunk_key):
+                    self.cache_input(input_url)
+                ds = self.open_chunk(first_chunk_key).chunk()
 
-@dataclass
-class InputCachingMixin(FSSpecFileOpenerMixin):
-    require_cache: bool  # = False
-    input_cache: InputCache
+                # make sure the concat dim has a valid fill_value to avoid
+                # overruns when writing chunk
+                ds[self.sequence_dim].encoding = {"_FillValue": -1}
+                # actually not necessary if we use decode_times=False
+                self.initialize_target(ds)
+
+            self.expand_target_dim(self.sequence_dim, self.sequence_len())
+
+        return _prepare
 
-    # returns a function that takes one input, the input_key
-    # this allows us to parallelize these operations
     @property
-    def cache_input(self):
+    def cache_input(self) -> Callable:
+        """Cache the input.
 
-        opener = super().input_opener
+        Properties
+        ----------
+        url : URL pointing to the input file. Must be openable by fsspec.
+        """
 
         def cache_func(fname: str) -> None:
             logger.info(f"Caching input '{fname}'")
-            with opener(fname, mode="rb") as source:
+            with input_opener(fname, mode="rb") as source:
                 with self.input_cache.open(fname, mode="wb") as target:
                     target.write(source.read())
 
         return cache_func
 
+    @property
+    def store_chunk(self) -> Callable:
+        """Store a chunk in the target.
+
+        Parameters
+        ----------
+        chunk_key : str
+            The identifier for the chunk
+        """
+        def _store_chunk(chunk_key):
+            ds_chunk = self.open_chunk(chunk_key)
+
+            def drop_vars(ds):
+                # writing a region means that all the variables MUST have sequence_dim
+                to_drop = [v for v in ds.variables if self.sequence_dim not in ds[v].dims]
+                return ds.drop_vars(to_drop)
+
+            ds_chunk = drop_vars(ds_chunk)
+            target_mapper = self.target.get_mapper()
+            write_region = self.region_for_chunk(chunk_key)
+            logger.info(f"Storing chunk '{chunk_key}' to Zarr region {write_region}")
+            ds_chunk.to_zarr(target_mapper, region=write_region)
+
+        return _store_chunk
+
+    @property
+    def finalize(self) -> Callable:
+        """Finalize writing of dataset."""
+
+        def _finalize():
+            if self.consolidate_zarr:
+                logger.info("Consolidating Zarr metadata")
+                target_mapper = self.target.get_mapper()
+                zarr.consolidate_metadata(target_mapper)
+
+        return _finalize
+
     @contextmanager
-    def input_opener(self, fname):
-        if self.input_cache.exists(fname):
+    def input_opener(self, fname: str):
+        if self.input_cache is None:
+            logger.info(f"No cache. Opening input `{fname}` directly.")
+            # This will bypass the cache. May be slow.
+            with input_opener(fname, mode="rb") as f:
+                yield f
+        elif self.input_cache.exists(fname):
             logger.info(f"Input '{fname}' found in cache")
             with self.input_cache.open(fname, mode="rb") as f:
                 yield f
-        elif self.require_cache:
-            # this creates an error on prepare because nothing is cached
-            raise IOError("Input can only be opened from cache. Call .cache_input first.")
         else:
-            logger.info(f"Input '{fname}' not found in cache. Opening directly.")
-            # This will bypass the cache. May be slow.
-            with super().input_opener(fname, mode="rb") as f:
-                yield f
-
+            raise ValueError(f"Input '{fname}' has not been cached yet. "
+                             "Call .cache_input() first.")
 
-@dataclass
-class XarrayInputOpener:
-    xarray_open_kwargs: dict
-
-    def open_input(self, fname):
+    def open_input(self, fname: str):
         with self.input_opener(fname) as f:
             logger.info(f"Opening input with Xarray '{fname}'")
-            ds = xr.open_dataset(f, **self.xarray_open_kwargs).load()
-            # do we always want to remove encoding? I think so.
+            ds = xr.open_dataset(f, **self.xarray_open_kwargs)
+            # explicitly load into memory
+            ds = ds.load()
         ds = fix_scalar_attr_encoding(ds)
         logger.debug(f"{ds}")
         return ds
 
-
-@dataclass
-class XarrayConcatChunkOpener(XarrayInputOpener):
-    xarray_concat_kwargs: dict
-
     def open_chunk(self, chunk_key):
         logger.info(f"Concatenating inputs for chunk '{chunk_key}'")
         inputs = self.inputs_for_chunk(chunk_key)
@@ -166,28 +201,9 @@ def open_chunk(self, chunk_key):
         ds = xr.concat(dsets, self.sequence_dim, **self.xarray_concat_kwargs)
         logger.debug(f"{ds}")
 
-        # do we really want to just delete all encoding?
-        # for v in ds.variables:
-        #    ds[v].encoding = {}
-
         # TODO: maybe do some chunking here?
         return ds
 
-
-@dataclass
-class ZarrXarrayWriterMixin:
-    @property
-    def store_chunk(self) -> Callable:
-        def _store_chunk(chunk_key):
-            ds_chunk = self.open_chunk(chunk_key)
-            ds_chunk = self.preprocess_chunk(ds_chunk)
-            target_mapper = self.target.get_mapper()
-            write_region = self.region_for_chunk(chunk_key)
-            logger.info(f"Storing chunk '{chunk_key}' to Zarr region {write_region}")
-            ds_chunk.to_zarr(target_mapper, region=write_region)
-
-        return _store_chunk
-
     def open_target(self):
         target_mapper = self.target.get_mapper()
         return xr.open_zarr(target_mapper)
@@ -210,52 +226,6 @@ def expand_target_dim(self, dim, dimsize):
             shape[axis] = dimsize
             arr.resize(shape)
 
-
-@dataclass
-class ZarrConsolidatorMixin:
-    consolidate_zarr: bool  # = True
-
-    @property
-    def finalize(self):
-        def _finalize():
-            if self.consolidate_zarr:
-                logger.info("Consolidating Zarr metadata")
-                target_mapper = self.target.get_mapper()
-                zarr.consolidate_metadata(target_mapper)
-
-        return _finalize
-
-
-@dataclass
-class SequenceRecipe(DatasetRecipe):
-    """There are many inputs (a.k.a. files, granules), arranged in a sequence
-    along the dimension `sequence_dim`. Each file may contain multiple variables.
-    """
-
-    input_urls: Iterable[str]
-    """The inputs used to generate the dataset."""
-
-    sequence_dim: str
-    """The dimension name along which the inputs will be concatenated."""
-
-    inputs_per_chunk: int = 1
-    """The number of inputs to use in each chunk."""
-
-    nitems_per_input: int = 1
-    """The length of each input along the `sequence_dim` dimension."""
-
-    def __post_init__(self):
-        self._chunks_inputs = {
-            k: v for k, v in enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))
-        }
-
-        def drop_vars(ds):
-            # writing a region means that all the variables MUST have sequence_dim
-            to_drop = [v for v in ds.variables if self.sequence_dim not in ds[v].dims]
-            return ds.drop_vars(to_drop)
-
-        self.chunk_preprocess_funcs.append(drop_vars)
-
     def inputs_for_chunk(self, chunk_key):
         return self._chunks_inputs[chunk_key]
 
@@ -272,7 +242,8 @@ def region_for_chunk(self, chunk_key):
         # specifies where in the overall array to put this chunk's data
         stride = self.nitems_per_input * self.inputs_per_chunk
         start = chunk_key * stride
-        return {self.sequence_dim: slice(start, start + self.nitems_for_chunk(chunk_key))}
+        region_slice = slice(start, start + self.nitems_for_chunk(chunk_key))
+        return {self.sequence_dim: region_slice}
 
     def sequence_len(self):
         # tells the total size of dataset along the sequence dimension
@@ -285,36 +256,3 @@ def sequence_chunks(self):
     def iter_chunks(self):
         for k in self._chunks_inputs:
             yield k
-
-    @property
-    def prepare(self):
-        def _prepare():
-
-            try:
-                ds = self.open_target()
-                logger.info("Found an existing dataset in target")
-                logger.debug(f"{ds}")
-            except (IOError, zarr.errors.GroupNotFoundError):
-                first_chunk_key = next(self.iter_chunks())
-                ds = self.open_chunk(first_chunk_key).chunk()
-
-                # make sure the concat dim has a valid fill_value to avoid
-                # overruns when writing chunk
-                ds[self.sequence_dim].encoding = {"_FillValue": -1}
-                # actually not necessary if we use decode_times=False
-                self.initialize_target(ds)
-
-            self.expand_target_dim(self.sequence_dim, self.sequence_len())
-
-        return _prepare
-
-
-@dataclass
-class StandardSequentialRecipe(
-    SequenceRecipe,
-    InputCachingMixin,
-    XarrayConcatChunkOpener,
-    ZarrXarrayWriterMixin,
-    ZarrConsolidatorMixin,
-):
-    pass
diff --git a/tests/conftest.py b/tests/conftest.py
index 6389e837..6386481a 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -92,18 +92,13 @@ def tmp_cache(tmpdir_factory):
 
 
 @pytest.fixture
-def sequential_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
-    r = recipe.StandardSequentialRecipe(
-        consolidate_zarr=True,
-        xarray_open_kwargs={},
-        xarray_concat_kwargs={},
-        require_cache=False,
-        input_cache=tmp_cache,
-        target=tmp_target,
-        chunk_preprocess_funcs=[],
+def netCDFtoZarr_sequential_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
+    r = recipe.NetCDFtoZarrSequentialRecipe(
         input_urls=netcdf_local_paths,
         sequence_dim="time",
         inputs_per_chunk=1,
         nitems_per_input=daily_xarray_dataset.attrs["items_per_file"],
+        target=tmp_target,
+        input_cache=tmp_cache,
     )
     return r, daily_xarray_dataset, tmp_target
diff --git a/tests/test_executors.py b/tests/test_executors.py
index eb90e2a0..551abaa5 100644
--- a/tests/test_executors.py
+++ b/tests/test_executors.py
@@ -5,8 +5,8 @@
 
 
 @pytest.mark.parametrize("Executor", [PythonExecutor, PrefectExecutor])
-def test_recipe_w_executor(Executor, sequential_recipe):
-    rec, ds_expected, target = sequential_recipe
+def test_recipe_w_executor(Executor, netCDFtoZarr_sequential_recipe):
+    rec, ds_expected, target = netCDFtoZarr_sequential_recipe
     ex = Executor()
     plan = ex.prepare_plan(rec)
     ex.execute_plan(plan)
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 11675e17..c146565e 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -6,6 +6,7 @@
 dummy_fnames = ["a.nc", "b.nc", "c.nc"]
 
 
+@pytest.mark.skip(reason="Removed this class for now")
 @pytest.mark.parametrize(
     "file_urls, files_per_chunk, expected_keys, expected_filenames",
     [
@@ -44,22 +45,17 @@ def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_fil
         assert fnames == expected
 
 
-def test_full_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
+def test_NetCDFtoZarrSequentialRecipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
 
     # the same recipe is created as a fixture in conftest.py
     # I left it here explicitly because it makes the test easier to read.
-    r = recipe.StandardSequentialRecipe(
-        consolidate_zarr=True,
-        xarray_open_kwargs={},
-        xarray_concat_kwargs={},
-        require_cache=False,
-        input_cache=tmp_cache,
-        target=tmp_target,
-        chunk_preprocess_funcs=[],
+    r = recipe.NetCDFtoZarrSequentialRecipe(
         input_urls=netcdf_local_paths,
         sequence_dim="time",
         inputs_per_chunk=1,
         nitems_per_input=daily_xarray_dataset.attrs["items_per_file"],
+        target=tmp_target,
+        input_cache=tmp_cache,
     )
 
     # this is the cannonical way to manually execute a recipe

From 1ec63eb1513b46b6cf649711151e919d820409fc Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 21 Dec 2020 11:49:35 -0500
Subject: [PATCH 18/34] fix precommit again

---
 pangeo_forge/executors/python.py |  1 +
 pangeo_forge/recipe.py           | 10 ++++++++--
 tests/test_recipe.py             | 17 ++++-------------
 3 files changed, 13 insertions(+), 15 deletions(-)

diff --git a/pangeo_forge/executors/python.py b/pangeo_forge/executors/python.py
index 0661d153..8e6625e8 100644
--- a/pangeo_forge/executors/python.py
+++ b/pangeo_forge/executors/python.py
@@ -3,6 +3,7 @@
 
 Task = Callable[[], None]
 
+
 class PythonExecutor:
     def prepare_plan(self, r):
         tasks = []
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index af389317..2af0ebfe 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -40,22 +40,26 @@
 #    Might be coming from the cache or might be read directly.
 # 6)
 
+
 @contextmanager
 def input_opener(fname, **kwargs):
     logger.info(f"Opening input '{fname}'")
     with fsspec.open(fname, **kwargs) as f:
         yield f
 
+
 # Notes about dataclasses:
 # - https://www.python.org/dev/peps/pep-0557/#inheritance
 # - https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
 # This means that, for now, I can't get default arguments to work.
 
+
 @dataclass
 class NetCDFtoZarrSequentialRecipe:
     """There are many inputs (a.k.a. files, granules), arranged in a sequence
     along the dimension `sequence_dim`. Each file may contain multiple variables.
     """
+
     input_urls: Iterable[str]
     """The inputs used to generate the dataset."""
 
@@ -140,6 +144,7 @@ def store_chunk(self) -> Callable:
         chunk_key : str
             The identifier for the chunk
         """
+
         def _store_chunk(chunk_key):
             ds_chunk = self.open_chunk(chunk_key)
 
@@ -180,8 +185,9 @@ def input_opener(self, fname: str):
             with self.input_cache.open(fname, mode="rb") as f:
                 yield f
         else:
-            raise ValueError(f"Input '{fname}' has not been cached yet. "
-                             "Call .cache_input() first.")
+            raise ValueError(
+                f"Input '{fname}' has not been cached yet. " "Call .cache_input() first."
+            )
 
     def open_input(self, fname: str):
         with self.input_opener(fname) as f:
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index c146565e..9b4c5461 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -11,18 +11,7 @@
     "file_urls, files_per_chunk, expected_keys, expected_filenames",
     [
         (dummy_fnames, 1, [0, 1, 2], [("a.nc",), ("b.nc",), ("c.nc",)]),
-        (
-            dummy_fnames,
-            2,
-            [0, 1],
-            [
-                (
-                    "a.nc",
-                    "b.nc",
-                ),
-                ("c.nc",),
-            ],
-        ),
+        (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",)]),
     ],
 )
 def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames, tmp_target):
@@ -45,7 +34,9 @@ def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_fil
         assert fnames == expected
 
 
-def test_NetCDFtoZarrSequentialRecipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
+def test_NetCDFtoZarrSequentialRecipe(
+    daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache
+):
 
     # the same recipe is created as a fixture in conftest.py
     # I left it here explicitly because it makes the test easier to read.

From a4bf88a820f1f2444d6aa0bb898b665a0caa4666 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 21 Dec 2020 20:50:05 -0500
Subject: [PATCH 19/34] finally

---
 tests/test_recipe.py |  2 +-
 tests/test_utils.py  | 24 ++----------------------
 2 files changed, 3 insertions(+), 23 deletions(-)

diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 9b4c5461..2a6a412d 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -11,7 +11,7 @@
     "file_urls, files_per_chunk, expected_keys, expected_filenames",
     [
         (dummy_fnames, 1, [0, 1, 2], [("a.nc",), ("b.nc",), ("c.nc",)]),
-        (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",)]),
+        (dummy_fnames, 2, [0, 1], [("a.nc", "b.nc",), ("c.nc",),],),  # noqa: E231
     ],
 )
 def test_sequence_recipe(file_urls, files_per_chunk, expected_keys, expected_filenames, tmp_target):
diff --git a/tests/test_utils.py b/tests/test_utils.py
index 0d9b8ee8..d486b7af 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -8,28 +8,8 @@
     [
         ([1, 2, 3], 1, [(1,), (2,), (3,)]),
         ([1, 2, 3], 2, [(1, 2), (3,)]),
-        (
-            [1, 2, 3],
-            3,
-            [
-                (
-                    1,
-                    2,
-                    3,
-                )
-            ],
-        ),
-        (
-            [1, 2, 3],
-            4,
-            [
-                (
-                    1,
-                    2,
-                    3,
-                )
-            ],
-        ),
+        ([1, 2, 3], 3, [(1, 2, 3,)],),
+        ([1, 2, 3], 4, [(1, 2, 3,)],),
     ],
 )
 def test_chunked_iterable(iterable, size, expected):

From dbf6b132bbd066ff4afacd59943d391fe34aded6 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 18 Jan 2021 12:39:19 -0500
Subject: [PATCH 20/34] cleanup

---
 pangeo_forge/executors/__init__.py |  8 -------
 pangeo_forge/executors/prefect.py  | 38 ------------------------------
 pangeo_forge/executors/python.py   | 25 --------------------
 pangeo_forge/recipe.py             |  9 +++++++
 setup.cfg                          |  2 +-
 tests/test_executors.py            | 14 +++++++----
 6 files changed, 20 insertions(+), 76 deletions(-)
 delete mode 100644 pangeo_forge/executors/__init__.py
 delete mode 100644 pangeo_forge/executors/prefect.py
 delete mode 100644 pangeo_forge/executors/python.py

diff --git a/pangeo_forge/executors/__init__.py b/pangeo_forge/executors/__init__.py
deleted file mode 100644
index e0aa6809..00000000
--- a/pangeo_forge/executors/__init__.py
+++ /dev/null
@@ -1,8 +0,0 @@
-"""
-Executors know how to run recipes.
-"""
-
-from .prefect import PrefectExecutor
-from .python import PythonExecutor
-
-__all__ = [PythonExecutor, PrefectExecutor]
diff --git a/pangeo_forge/executors/prefect.py b/pangeo_forge/executors/prefect.py
deleted file mode 100644
index ce91017e..00000000
--- a/pangeo_forge/executors/prefect.py
+++ /dev/null
@@ -1,38 +0,0 @@
-from typing import Any
-
-import prefect
-
-
-class PrefectExecutor:
-    def prepare_plan(self, r) -> prefect.Flow:
-
-        # wrap our functions as prefect tasks
-        @prefect.task
-        def prepare() -> None:
-            r.prepare()
-
-        @prefect.task
-        def cache_input(input: Any) -> None:
-            r.cache_input(input)
-
-        @prefect.task
-        def store_chunk(input: Any) -> None:
-            r.store_chunk(input)
-
-        @prefect.task
-        def finalize() -> None:
-            r.finalize()
-
-        with prefect.Flow("Pangeo-Forge") as flow:
-            prep_task = prepare()
-            cache_task = cache_input.map(list(r.iter_inputs()))
-            cache_task.set_dependencies(upstream_tasks=[prep_task])
-            store_task = store_chunk.map(list(r.iter_chunks()))
-            store_task.set_dependencies(upstream_tasks=[cache_task])
-            finalize_task = finalize()
-            finalize_task.set_dependencies(upstream_tasks=[store_task])
-
-        return flow
-
-    def execute_plan(self, plan: prefect.Flow, **kwargs):
-        return plan.run(**kwargs)
diff --git a/pangeo_forge/executors/python.py b/pangeo_forge/executors/python.py
deleted file mode 100644
index 8e6625e8..00000000
--- a/pangeo_forge/executors/python.py
+++ /dev/null
@@ -1,25 +0,0 @@
-from functools import partial
-from typing import Callable, Iterable
-
-Task = Callable[[], None]
-
-
-class PythonExecutor:
-    def prepare_plan(self, r):
-        tasks = []
-        tasks.append(r.prepare)
-        for input in r.iter_inputs():
-            tasks.append(partial(r.cache_input, input))
-        for chunk in r.iter_chunks():
-            tasks.append(partial(r.store_chunk, chunk))
-        tasks.append(r.finalize)
-
-        return partial(_execute_all, tasks)
-
-    def execute_plan(self, plan: Task, **kwargs):
-        plan()
-
-
-def _execute_all(tasks: Iterable[Task]) -> None:
-    for task in tasks:
-        task()
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index 2af0ebfe..5c9b322a 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -10,6 +10,7 @@
 import fsspec
 import xarray as xr
 import zarr
+from rechunker.types import MultiStagePipeline, ParallelPipelines, Stage
 
 from .storage import InputCache, Target
 from .utils import chunked_iterable, fix_scalar_attr_encoding
@@ -92,6 +93,14 @@ def __post_init__(self):
             k: v for k, v in enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))
         }
 
+    def to_pipelines(self) -> ParallelPipelines:
+        pipeline = []  # type: MultiStagePipeline
+        pipeline.append(Stage(self.prepare))
+        pipeline.append(Stage(self.cache_input, list(self.iter_inputs())))
+        pipeline.append(Stage(self.store_chunk, list(self.iter_chunks())))
+        pipeline.append(Stage(self.finalize))
+        return [pipeline]  # type: ParallelPipelines
+
     @property
     def prepare(self) -> Callable:
         """Prepare target for storing dataset."""
diff --git a/setup.cfg b/setup.cfg
index 9c7719f5..d2063367 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -3,7 +3,7 @@ max-line-length = 100
 
 [isort]
 known_first_party=pangeo_forge
-known_third_party=click,fsspec,numpy,pandas,pkg_resources,prefect,pytest,setuptools,sphinx_pangeo_theme,xarray,zarr
+known_third_party=click,fsspec,numpy,pandas,pkg_resources,prefect,pytest,rechunker,setuptools,sphinx_pangeo_theme,xarray,zarr
 multi_line_output=3
 include_trailing_comma=True
 force_grid_wrap=0
diff --git a/tests/test_executors.py b/tests/test_executors.py
index 551abaa5..b21ad56a 100644
--- a/tests/test_executors.py
+++ b/tests/test_executors.py
@@ -1,14 +1,20 @@
 import pytest
 import xarray as xr
+from rechunker.executors import (
+    DaskPipelineExecutor,
+    PrefectPipelineExecutor,
+    PythonPipelineExecutor,
+)
 
-from pangeo_forge.executors import PrefectExecutor, PythonExecutor
 
-
-@pytest.mark.parametrize("Executor", [PythonExecutor, PrefectExecutor])
+@pytest.mark.parametrize(
+    "Executor", [PythonPipelineExecutor, DaskPipelineExecutor, PrefectPipelineExecutor]
+)
 def test_recipe_w_executor(Executor, netCDFtoZarr_sequential_recipe):
     rec, ds_expected, target = netCDFtoZarr_sequential_recipe
+    pipeline = rec.to_pipelines()
     ex = Executor()
-    plan = ex.prepare_plan(rec)
+    plan = ex.pipelines_to_plan(pipeline)
     ex.execute_plan(plan)
     ds_actual = xr.open_zarr(target.get_mapper()).load()
     assert ds_actual.identical(ds_expected)

From c2927be41f69f63ec4b52992d66fe2dc02f0f600 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 18 Jan 2021 12:41:20 -0500
Subject: [PATCH 21/34] add rechunker to CI

---
 ci/py3.7.yml | 2 ++
 ci/py3.8.yml | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/ci/py3.7.yml b/ci/py3.7.yml
index c310ff4c..807f4192 100644
--- a/ci/py3.7.yml
+++ b/ci/py3.7.yml
@@ -27,3 +27,5 @@ dependencies:
   - toolz
   - xarray>=0.16.2
   - zarr>=2.6.0
+  - pip:
+    - git+https://github.com/rabernat/rechunker.git@refactor-executors
diff --git a/ci/py3.8.yml b/ci/py3.8.yml
index 739ce1ef..bbdd3a6e 100644
--- a/ci/py3.8.yml
+++ b/ci/py3.8.yml
@@ -27,3 +27,5 @@ dependencies:
   - toolz
   - xarray>=0.16.2
   - zarr>=2.6.0
+  - pip:
+    - git+https://github.com/rabernat/rechunker.git@refactor-executors

From 9a1d11f41a7be9b508496ed6d70c185acdfc431d Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 18 Jan 2021 12:42:46 -0500
Subject: [PATCH 22/34] add rechunker to requirements.txt

---
 requirements.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/requirements.txt b/requirements.txt
index 11a36cbe..83520897 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,6 +2,7 @@ setuptools
 click
 dask
 distributed
+rechunker
 xarray >= 0.16.2
 zarr >= 2.6.0
 fsspec[http]

From 1879acb2a2a6ecf928c6fba9264f736f5c833e1b Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 18 Jan 2021 16:55:03 -0500
Subject: [PATCH 23/34] create ABC for Recipe

---
 docs/{design.md => concepts.md} | 13 ++++++-
 docs/index.md                   |  5 +--
 pangeo_forge/recipe.py          | 61 +++++++++++++++++++++++++++------
 3 files changed, 65 insertions(+), 14 deletions(-)
 rename docs/{design.md => concepts.md} (93%)

diff --git a/docs/design.md b/docs/concepts.md
similarity index 93%
rename from docs/design.md
rename to docs/concepts.md
index fa81f088..65da4a82 100644
--- a/docs/design.md
+++ b/docs/concepts.md
@@ -1,8 +1,19 @@
-# Design
+# Concepts
 
 pangeo-forge is modeled after [conda-forge], a community-led collection of recipes
 for building conda packages.
 
+## Recipes
+
+The most important concept in Pangeo Forge is a ``Recipe``.
+A recipe defines how to transform data in one format / location into another format / location.
+
+
+## Storage
+
+
+
+
 ## Components
 
 The high-level components of pangeo-forge are:
diff --git a/docs/index.md b/docs/index.md
index 14b7ad31..6b629e03 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -4,6 +4,7 @@
 :maxdepth: 2
 :caption: Contents
 
-contribute
-design
+concepts
+contribte
+
 ```
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index 5c9b322a..adee9124 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -3,9 +3,10 @@
 """
 
 import logging
+from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from dataclasses import dataclass, field
-from typing import Callable, Iterable, Optional
+from typing import Callable, Hashable, Iterable, NoReturn, Optional
 
 import fsspec
 import xarray as xr
@@ -49,14 +50,60 @@ def input_opener(fname, **kwargs):
         yield f
 
 
+class BaseRecipe(ABC):
+    @property
+    @abstractmethod
+    def prepare(self) -> Callable[[], NoReturn]:
+        pass
+
+    @abstractmethod
+    def iter_inputs(self) -> Iterable[Hashable]:
+        pass
+
+    @property
+    @abstractmethod
+    def cache_input(self) -> Callable[[Hashable], NoReturn]:
+        pass
+
+    @abstractmethod
+    def iter_chunks(self) -> Iterable[Hashable]:
+        pass
+
+    @property
+    @abstractmethod
+    def store_chunk(self) -> Callable[[Hashable], NoReturn]:
+        pass
+
+    @property
+    @abstractmethod
+    def finalize(self) -> Callable[[], NoReturn]:
+        pass
+
+    def to_pipelines(self) -> ParallelPipelines:
+        """Translate recipe to pipelines
+
+        Returns
+        -------
+        pipeline : ParallelPipelines
+        """
+
+        pipeline = []  # type: MultiStagePipeline
+        pipeline.append(Stage(self.prepare))
+        pipeline.append(Stage(self.cache_input, list(self.iter_inputs())))
+        pipeline.append(Stage(self.store_chunk, list(self.iter_chunks())))
+        pipeline.append(Stage(self.finalize))
+        pipelines = []  # type: ParallelPipelines
+        pipelines.append(pipeline)
+        return pipelines
+
+
 # Notes about dataclasses:
 # - https://www.python.org/dev/peps/pep-0557/#inheritance
 # - https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
-# This means that, for now, I can't get default arguments to work.
 
 
 @dataclass
-class NetCDFtoZarrSequentialRecipe:
+class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     """There are many inputs (a.k.a. files, granules), arranged in a sequence
     along the dimension `sequence_dim`. Each file may contain multiple variables.
     """
@@ -93,14 +140,6 @@ def __post_init__(self):
             k: v for k, v in enumerate(chunked_iterable(self.input_urls, self.inputs_per_chunk))
         }
 
-    def to_pipelines(self) -> ParallelPipelines:
-        pipeline = []  # type: MultiStagePipeline
-        pipeline.append(Stage(self.prepare))
-        pipeline.append(Stage(self.cache_input, list(self.iter_inputs())))
-        pipeline.append(Stage(self.store_chunk, list(self.iter_chunks())))
-        pipeline.append(Stage(self.finalize))
-        return [pipeline]  # type: ParallelPipelines
-
     @property
     def prepare(self) -> Callable:
         """Prepare target for storing dataset."""

From 0fa41ee1390616f6d2889712b1646b70b62f6084 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Mon, 18 Jan 2021 22:49:56 -0500
Subject: [PATCH 24/34] start working on docs

---
 docs/bakeries.md       |  1 +
 docs/concepts.md       | 75 ------------------------------------------
 docs/conf.py           |  4 +++
 docs/execution.md      |  1 +
 docs/index.md          | 38 +++++++++++++++++++--
 docs/recipes.md        | 15 +++++++++
 pangeo_forge/recipe.py | 51 +++++++++++++++-------------
 7 files changed, 85 insertions(+), 100 deletions(-)
 create mode 100644 docs/bakeries.md
 delete mode 100644 docs/concepts.md
 create mode 100644 docs/execution.md
 create mode 100644 docs/recipes.md

diff --git a/docs/bakeries.md b/docs/bakeries.md
new file mode 100644
index 00000000..efdfd4d7
--- /dev/null
+++ b/docs/bakeries.md
@@ -0,0 +1 @@
+# Bakeries
diff --git a/docs/concepts.md b/docs/concepts.md
deleted file mode 100644
index 65da4a82..00000000
--- a/docs/concepts.md
+++ /dev/null
@@ -1,75 +0,0 @@
-# Concepts
-
-pangeo-forge is modeled after [conda-forge], a community-led collection of recipes
-for building conda packages.
-
-## Recipes
-
-The most important concept in Pangeo Forge is a ``Recipe``.
-A recipe defines how to transform data in one format / location into another format / location.
-
-
-## Storage
-
-
-
-
-## Components
-
-The high-level components of pangeo-forge are:
-
-* https://github.com/pangeo-forge/staged-recipes: The birthplace for new recipes.
-  Anyone interested in adding a dataset to pangeo-forge can submit a new recipe
-  through a pull request, using the examples there as a starting point.
-* Recipes: Metadata describing a dataset and code for transforming it from
-  raw input to analysis-ready dataset.
-* pangeo-forge: A Python library containing useful pipeline helpers and the
-  [pangeo-forge command-line-interface][cli] for validating pipelines.
-* https://github.com/pangeo-forge/docker-images: A repository for building docker
-  images for pangeo-forge.
-
-## Pipeline Structure
-
-pangeo-forge uses [Prefect] for
-
-1. Pipeline definitions: the code that expresses the transformation from raw to analysis-ready data.
-2. Orchestration: everything involved with taking a pipeline definition and actually running it.
-
-This imposes a few constraints which we build upon.
-
-All recipe definitions must be named `recipe/pipeline.py` and must contain an instances
-of a `prefect.Flow` at the top-level of the module.
-
-Additionally, pangeo-forge imposes additional strucutre. All pipelines modules should include a `Pipeline`
-class that inherits from `pangeo_forge.AbstractPipeline`. So most pipeline modules will have
-
-
-```
-class Pipeline(pangeo_forge.AbstractPipeline):
-    @property
-    def flow(self):
-        with prefect.Flow(self.name) as flow:
-            ...
-        return flow
-
-
-pipeline = Pipeline()
-flow = pipeline.flow
-```
-
-## Lifecycle of a recipe
-
-A maintainer contributes a recipe to [staged-recipes] through a pull request. We
-use GitHub Actions to perform some initial validation. The actions are at https://github.com/pangeo-forge/staged-recipes/blob/master/.github/workflows/main.yaml and the logs are available at https://github.com/pangeo-forge/staged-recipes/actions.
-
-When a PR is merged into `staged-recipes` the [Create Repository](https://github.com/pangeo-forge/staged-recipes/blob/master/.github/workflows/create-repository.yaml) action is run, which
-
-1. Creates a new git repository with the contents of `staged-recipes/examples/<new-pipeline>`
-   inserted, along with a few other files provided by pangeo-forge, including the `register_pipeline` action definition.
-2. Pushes that new repository to `https://pangeo-forge/<new-pipeline>`, for example https://github.com/pangeo-forge/example-pipeline/.
-3. Executes the [Register Pipeline](https://github.com/pangeo-forge/staged-recipes/blob/master/.github/workflows/scripts/register_pipeline.yaml) action, which registers the new pipeline with Prefect and executes it.
-
-[conda-forge]: https://conda-forge.github.io
-[cli]: https://github.com/pangeo-forge/pangeo-forge/blob/master/pangeo_forge/cli.py
-[Prefect]: https://docs.prefect.io
-[staged-recipes]: https://github.com/pangeo-forge/staged-recipes/
diff --git a/docs/conf.py b/docs/conf.py
index c46ec4db..2dfe4699 100644
--- a/docs/conf.py
+++ b/docs/conf.py
@@ -11,6 +11,9 @@
 
 extensions = [
     "myst_parser",
+    "sphinx.ext.autodoc",
+    # "numpydoc",
+    "sphinx_autodoc_typehints",
 ]
 
 templates_path = ["_templates"]
@@ -21,5 +24,6 @@
 
 html_theme = "pangeo"
 html_static_path = ["_static"]
+html_sidebars = {"index": [], "**": ["localtoc.html"]}
 
 myst_heading_anchors = 2
diff --git a/docs/execution.md b/docs/execution.md
new file mode 100644
index 00000000..7bb5b258
--- /dev/null
+++ b/docs/execution.md
@@ -0,0 +1 @@
+# Recipe Execution
diff --git a/docs/index.md b/docs/index.md
index 6b629e03..2718b2bd 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,10 +1,42 @@
-# pangeo-forge
+# Pangeo Forge
+
+Pangeo Forge is an open source tool for data Extraction, Transformation, and Loading (ETL).
+The goal of Pangeo Forge is to make it easy to extract data from traditional data
+repositories and deposit in cloud object storage in analysis-ready, cloud-optimize (ARCO) format.
+
+Pangeo Forge is inspired by [Conda Forge](https://conda-forge.org/), a
+community-led collection of recipes for building conda packages.
+We hope that Pangeo Forge can play the same role for datasets.
+
+## Recipes
+
+The most important concept in Pangeo Forge is a ``recipe``.
+A recipe defines how to transform data in one format / location into another format / location.
+The primary way people contribute to Pangeo Forge is by writing / maintaining recipes.
+Recipes developed by the community are stored in GitHub repositories.
+For information about how to write a recipe, see {doc}`recipes`.
+
+## Recipe Execution
+
+There are several different ways to execute recipes.
+See {doc}`execution` for details.
+
+## Bakeries
+
+Bakeries are cloud-based environments for executing recipes.
+Each Bakery is coupled to one or more cloud storage buckets where the ARCO data is stored.
+Bakeries use [Prefect](https://prefect.io/) to orchestrate the various steps
+of the recipe.
+For more information, see {doc}`bakeries`.
+
 
 ```{toctree}
 :maxdepth: 2
 :caption: Contents
 
-concepts
-contribte
+recipes
+execution
+bakeries
+contribute
 
 ```
diff --git a/docs/recipes.md b/docs/recipes.md
new file mode 100644
index 00000000..f053c335
--- /dev/null
+++ b/docs/recipes.md
@@ -0,0 +1,15 @@
+# Recipes
+
+## The Base Recipe Class
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.recipe.BaseRecipe
+    :members:
+```
+
+## Specific Recipe Classes
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe
+    :show-inheritance:
+```
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index adee9124..e42fbb7d 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -51,40 +51,53 @@ def input_opener(fname, **kwargs):
 
 
 class BaseRecipe(ABC):
+    """Base recipe class from which all other Recipes inherit.
+    """
+
     @property
     @abstractmethod
     def prepare(self) -> Callable[[], NoReturn]:
+        """Prepare the recipe for execution by initializing the target.
+        Attribute that returns a callable function.
+        """
         pass
 
     @abstractmethod
     def iter_inputs(self) -> Iterable[Hashable]:
+        """Iterate over all inputs."""
         pass
 
     @property
     @abstractmethod
     def cache_input(self) -> Callable[[Hashable], NoReturn]:
+        """Copy an input from its source location to the cache.
+        Attribute that returns a callable function.
+        """
         pass
 
     @abstractmethod
     def iter_chunks(self) -> Iterable[Hashable]:
+        """Iterate over all target chunks."""
         pass
 
     @property
     @abstractmethod
     def store_chunk(self) -> Callable[[Hashable], NoReturn]:
+        """Store a chunk of data in the target.
+        Attribute that returns a callable function.
+        """
         pass
 
     @property
     @abstractmethod
     def finalize(self) -> Callable[[], NoReturn]:
+        """Final step to finish the recipe after data has been written.
+        Attribute that returns a callable function.
+        """
         pass
 
     def to_pipelines(self) -> ParallelPipelines:
-        """Translate recipe to pipelines
-
-        Returns
-        -------
-        pipeline : ParallelPipelines
+        """Translate recipe to pipeline for execution.
         """
 
         pipeline = []  # type: MultiStagePipeline
@@ -106,34 +119,28 @@ def to_pipelines(self) -> ParallelPipelines:
 class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     """There are many inputs (a.k.a. files, granules), arranged in a sequence
     along the dimension `sequence_dim`. Each file may contain multiple variables.
+
+    :param input_urls: The inputs used to generate the dataset.
+    :param sequence_dim: The dimension name along which the inputs will be concatenated.
+    :param inputs_per_chunk: The number of inputs to use in each chunk.
+    :param nitems_per_input: The length of each input along the `sequence_dim` dimension.
+    :param target: A location in which to put the dataset. Can also be assigned at run time.
+    :param input_cache: A location in which to cache temporary data.
+    :param consolidate_zarr: Whether to consolidate the resulting Zarr dataset.
+    :param xarray_open_kwargs: Extra options for opening the inputs with Xarray.
+    :param xarray_concat_kwargs: Extra options to pass to Xarray when concatenating
+       the inputs to form a chunk.
     """
 
     input_urls: Iterable[str]
-    """The inputs used to generate the dataset."""
-
     sequence_dim: str
-    """The dimension name along which the inputs will be concatenated."""
-
     inputs_per_chunk: int = 1
-    """The number of inputs to use in each chunk."""
-
     nitems_per_input: int = 1
-    """The length of each input along the `sequence_dim` dimension."""
-
     target: Optional[Target] = None
-    """A location in which to put the dataset."""
-
     input_cache: Optional[InputCache] = None
-    """The length of each input along the `sequence_dim` dimension."""
-
     consolidate_zarr: bool = True
-    """Whether to consolidate the resulting Zarr dataset."""
-
     xarray_open_kwargs: dict = field(default_factory=dict)
-    """Extra options for opening the inputs with Xarray."""
-
     xarray_concat_kwargs: dict = field(default_factory=dict)
-    """Extra options to pass to Xarray when concatenating the inputs to form a chunk."""
 
     def __post_init__(self):
         self._chunks_inputs = {

From 4ee634a4806b569310d43f87d04f40a1a4222d27 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Tue, 19 Jan 2021 10:39:46 -0500
Subject: [PATCH 25/34] writing more docs

---
 docs/_static/pangeo-forge-logo-blue.png | Bin 0 -> 17602 bytes
 docs/api.md                             |  26 +++++++++++++
 docs/conf.py                            |  13 ++++---
 docs/index.md                           |   1 +
 docs/recipes.md                         |  48 ++++++++++++++++++++++--
 pangeo_forge/recipe.py                  |  20 +---------
 pangeo_forge/storage.py                 |  33 +++++++---------
 setup.cfg                               |   2 +-
 8 files changed, 96 insertions(+), 47 deletions(-)
 create mode 100644 docs/_static/pangeo-forge-logo-blue.png
 create mode 100644 docs/api.md

diff --git a/docs/_static/pangeo-forge-logo-blue.png b/docs/_static/pangeo-forge-logo-blue.png
new file mode 100644
index 0000000000000000000000000000000000000000..467758a8305bbc71aac9b174d10be530bc8ed8c1
GIT binary patch
literal 17602
zcmdSB^;?wR7dA==DAGtt2@EAQLpLhj0!sG~j&wHz(k<O7$OmbpbI1{q?oNm9?)UL~
z&iV2E1I}C*7hE$ud#}CrUiYedPsDpQd3@ZLxM*l-_=*ZLnrLVkENE!xI#02IcT#ji
zI)Go#92N9j(9r0*{{5iakGM$!Z&JC+>bhz<Sh;$bIa{K6czAHb>}_12W{#HJ4$jt@
z2jVZ$&|aY_%DnyHnYFj*;cM`B_VIYw4>UWJLsat%jrZyMSC$|_*eCY&d@TK!q5%i*
zlymvr6W&c%y5*|9n)zNk^8IwXuxRA?eO?*1x!9AZSW2&a5+v-9T4POcI9v1+Ld8u5
z`Gk(cY2`}t-Gynt-kNgdP2*RL|NnlqeR(E|P8`oYXz0lqOB+@94RRq|MeKNkX)5yc
zfT2tz7!srVGw^0!5GSkJ6_wD`%P-q!Mit|P4A{oRLnFmT6cMk>L59=}FF%@(mW$ti
z^Xp{cdaGfUFl_Y!ueV2jRVI>eE-r?a@x|x->k&0=&RDD3O4lqthbIiNNGEJ&H@}5F
z%$9V^$y>gx*?82SXVo@)Y=$njV|HaW;dtIT%*hZFmnI8)$RNsd;TM9gQlNGWX^UEw
z9c)KA*3KrI72xxMFBmzkgG6YBS&Aq_u}zt$9&l_@h~gv=Ct@%85-MHo<a$5RaQB@M
z(5NAK$(|OI_t@i{idWI3hUp4!>HX)*7hkbKPR4Tm-=4B`)jk}1DgAHGI;FemHDL;w
z>7%UAz!)Mj*#$)aB6*NVQAT6!or0{`GX_a^C;YC*v+yXSBC}pUvNpB-ZHHfJ-WXam
zseYB947;8*J;1;&9C?lFPV6RNa?BjYbdnC|?(-)ZpfRl)9-W7XM-H^r3O&Y?MH%?U
z@>Hg4KwgB!q0!@fx6TWbpnM2$f)9<&oBVm>+fV#j<wQg8x}h3X`nDJyB$==O5bXwk
zcfUVuQFmAqDP?;7wc*tT>%^q*RBmvyhIX~M6U8glQ0%M`D)FXI--^Y#(+yOho<bw1
z*$G3sKG^31!#=-P1&aQr2Arx^SznV>TKjZ91o*H1h+e)&XXD6?+06>AW~vJ2T{7W~
zc7s?^IC+o_go>{X(1xc!kL&H}c;ZXGwJu`pOoyyCv4*J{rQ9TPYMLd)DRDgD8q&GG
z%=Ub7#p^aXY}HzvZX%9OYJi{fxFY{7oz%1Db16N{t)#YqYHJtrdwt@g3EE686R6!`
zcg)^lPrju5T6J#UWFT-x9o#ygl~p^tue<qtLT53vn)~-_d*n$-bnj)*s{Q9usVw(f
z5tnzi^kNiy4T1g-%qa;z5y6j{%nb*4|H1nMJ~+UYxxwtQE!M+W2xp~%(sYgZRdu!M
zxb-$R^l+kv=^LVA8w&2du*-rcE7?iOChP29V(?EVmfTwJ_Xe{bsZ{oQS0YK!_;8_*
zfBr<8@ZF0_8K*K0w&zQwQ2WO13Jrv@WqSDP+YP6Sh3-RBzLc{x0wXCER3|45b#NkR
zdaQwMiyzz;Ga^WHe#dEz`r<^j(`|Yc9(T-Zkmg!@b?L3AHTWl_Up+@c|JQj+l+*Za
zCd67TeC%2ZG1GgcRp@2in6l=tEMmR%)MUNymPY+IsYH-D%l8Efx!&2}Zgot)yJ1!?
z>CT6cnn)h3?C>k;kO+&=zaOGoqHC@bedCPuHLwdGaO!Nd*vdUyfN5credV~K5fdLo
zZRziySGfcYZE)dPlenjVZ=;ac<l}vs=^tn~r$G$~`vR7yL!GO{rWRHtN|`}_eUh|&
z`L<+w#K>3@6q%|oG^_^I{g{<JIF~~CYRLGYdy`2YTZOhb6SU_X#nR%iv*m=oj}YJD
z{B+o&cV!gWUpOhz-DDWc7S5(MDK0ZN9#ISzdaVK)?^~Ob;w!KDmHKJ?brjfeTe(b#
z&o^y*TGlr%lRbqV)k_W2yABI{BoK|EqZ$I~8<M)d6xr<xxqq52wf^XA|ENc<QJ=@T
z#SiiWhCAi7<|`b{z30IeXs0+54E*VW)Es+UqzcbuN0M?q<mM%AP5r@IuR;~SEsp=9
z&YS`sFyKd0ptIs+%O!N_UN%sr&v5GHcgfUr&t<07Ieys`W(nf>#7iZ{hl2bR*I{|p
zp6nE4Tpf>ns%4eB`Ct@iD_VJ{vcd-pSKyR%Hf=E@UIaW{6r6S|piT^6AQtxpYqpwh
zh3SoIi7GZ;N)XgPAyNAMW^yXV;d9X#{%96d5>B>^T12M7ku}UeWi}*7yGqgYLz=<`
zrB7Y)Bq4+36N?kWJd#MUrQFV|ho4zWijN_0i;u~jSTCKGQtDhz%^%bRvxE-$Ri@(F
zkCf?E?oQvbGL&eUG0uKa;ZJ!`OBC$Bjd^445Y-Cp=xq9{6*RQw_7fG?JL9`_^qGxT
zHutG$M46*X>-^^8h~bG(*wJn}A1mSE#hJZV&9kpB@VapVc9Q0q>mgFe$KWM%!OfJo
z&5{gNy{!SBUSSmXif^l}&S7ffJ=%l+)@+nOoT=x_Zow)Xw5sn<p-AYyo+&r;WMX&=
z^iOesm~T{SY1zIv$0e}Ch#N5)YrUvd$ci4=2V9T!G>8^Mr8`5@l>FiTF52{VMi0^9
z8_BAd*H!58xy#wQL5Gs23K@`&u5wXsJs^^_BplFekfHf`v!`-!EVd!E>YC{2p7ZQ4
zeYY4J*W(K}0jcMrWOtbc&P2Kj79)q=;^;U39GFZ6a^`StICx$w%}*Mhvf%YAOt<Mr
za(zLak)*){oLsj<8_^bCM>a8i7nKvasrQ_fD9(@e57KXYc824<S%S`ov@^L-?42Kp
zxz)znPLPz|0D9`85J#vrW?a!opXs8FX8SkWJ^3mW>zmJFPAs9(C#7I?uoW@iYdQ<P
z^Lm7HBjC?s=+rd)V^%+Ajg%FtYxy6=y073Q4Z<;O=y5{5hZyFha>Kyh&O|n-nl)4?
zccr6kiImZ-6fgLpy1P<in56QWm5l>VW8)t(-NH(q`#_t(^;)ISx^9vXavNn$8Y6+_
zDfQS_T^z4u#tVJGt70u29mTEkQLF<8CW-(jHf2gY)jJox29Hq&AIw@s^ODIr;9T_W
z?Al!Ef2uWXUO(r4P_P*tF*rU(q&5+U(raX{8MRHH#k9&q4X!c~nEQOPTkMSN9i`(I
zpAH*JadAA7FDczqXtJ~$bYLr`4Et+9=i*5{Ub#CE*~>+@u8{H{IDUh($&GdMqc2`6
z^HNay#08(z*n&w<ezNdw;KZf6Qxo~G%C0Af6Lc$WLp8tiWzmjTG?+pz_tcjbWt-Ia
zgc>G4IrpdB8YH^@Tzf!me~^}Qr4>=y7uE03!!93>3am_J=k{H~E}u>I>HYf6ADJ*6
zjKjq^cSQB7P3tS6e7^c)VgQAB&sWZ|A2f<fV>_$}@pXx>w25a$5#@W8N~M{eSFciN
zVRgN}X|u`KdX3ZvJI;w<X&bp)cP&m)G3el-1QL891`;ElIl>KnlxK6raQk?eSHJU6
zI8XkgXW0Lwqd0qBVc)JUhmo_yw%a)lDUPUHmq30k(=cN!slHI8{{rfe@A(?4tmx4G
zhRLmg-%Mqv8~qtmINhIOuAOCD5G_h9Taq-!`og{b*o-crPTeb3L#K^I*~3Gk-mD3K
zI!n>+ynjL)F1CHm+4gp~ef`K=e(?(M&5qHR!fLqmp&&;91O-Cn4k0Qhr7U9-=hKd!
z82_n04)%$@J%Ewm++NliP)quFUQROUWzLGRtoT7|Vo`&V8|Vy|rAOa$b)+$;?l5eY
zWf$|-ODc_j)%<|pj_Kq(MYeXtazCha2_Co~zNN8COO*!z>T~pYyJd?&bzud^?HJL@
z*xAjB<6Co`qDeZ$%z%is{d@V=uvcf&g!{CHr!H1HDKy8Ej!xfM(Yz&u%b6n-)BE+j
zZ=;0Wac4LnDA%r_P%V>l_8l9#u-|xUwd_yo<2H%>eZ$eKP<fY>*X~wYhbg(xEfN}=
z?^i^HB25FL6^x5qTU?Lmh*d8_7l~Vv$X5ERk8{MWRR@%K9Dy*HT$)Pl>HtE>_))3Q
zfnyX~8RdR@=54R;tgMt+S8Ijim|F?)33}i9bobn)^xQfpmb9j)WjRkU;zOiwziiFB
zkO+_istd{>Iq&o8r&QDK*i6fYK&;U-&{CQbhdVmcSY)EG)0NIP^J;uI9w*^fe>03D
z=HIBt*ts79Xq+X{8v50lt{wCkd=$~$)k~M36+iy*izxnV-0$7NsQ0}~Cc%2GVdY+1
zg3?1PZaT|Al1}A@nyFXXL0`fmHAwMtFJ<;mhExh`Y5n(X<z~(z%3H6p>w+*gSzVv!
zj=j4kkTT>Cuwt$Rp$$Co++O;7$gEH_+tK1DSzu>K{sq#|J&)=OXnT)Kwwr`8DK9<q
z{c5NLXN%qW?xSTkt2oW;&6N}46%_xx4+M+UBm)2N$(U3!6?d^~XG=-TJ~3LE5suY&
z9xAK(vTJC>EEXjT`U5@8A$+}t8k)<(do@MZ&^27qK2rSTy%{5xYKWqPMY)ToIc&P?
z*nI8ay`cy8{58hRxbLo%MUL9qk5Y-VswAO9X=3Np74I}nv;Dd!T4};Wl2ljtD}O7V
zt!VSLoVccZ=&Q2*HE?QftrHqScSq4zSpD2F!8q3sx+Gm|_7p4y^@~s_u6<ps!~3mj
z|F*?X1}f&bn2Z=kdyFKR86HDltW09hh67RGJ@;Oa{N3hQ=dg4$bvZgXhU#?qXhhPF
zTPtGT<c0|Lk~W4O0|7CjUq~5P!%+{yBWHGSQzXA7iSGvV=T7yI(xW+ecEPRa7$I{Y
zb)uZi9eOL)A+ES>A9a{)`Fr(Q+>#Kj%U%o!n<JqKn(R|Xl|A`@$Mkb9HyjNxK|p6N
z4a4dk+!z-@bfW)R_m_)_d96K;#XUnz2HI*k?DdB|UKZy-_0)>W))-ShR(8+R7Mkj=
z`>nYt-D`}U$}+AA0fCmP&41Ka?;CWI&N`k@9^}Uj5mrW>vaMLUR6`>K3|I2~yrUIq
z3I$(1h#i@+st$bVejJ{sVqs*YDga`T?G(B&r)VSt>AGus%x%q%c2%5rq+S1mW2_}V
z-|L&B7217-ourH}#m{y}lBxf|#PL~$z+-=xqrJ~{7L6hI-`~#gH{{_U+Rtv}%(G8L
zvs|!>gb0o(nLdUpuvLF(H2DOn8g>yEx`R3T2}f(l^uU3YQ;P=dHg4t|TqbkJ8e<j0
z<-!Eu`w!Pg%bnfW=OYvahnFLg5>7(ZRCC=?8J$Z9A~|*R*6;7gsw8P*Ss2%NfK}r^
z|F6`)9$6Y)JSiNebLaZaf7Je{k&!ukl?Letb*ZaQnGfKiW1>^m+TeAhJED}A>G1$&
z{>j;(rL%a--UJvNrU5Fy*lyxurNWx1L$^riX}SQsteCRId5}7!AMi+&w*teE@>?pJ
zOZuF3i$u4&^%4c1ylfL@EzUFt0Ec37c(AIbSIXb}r_^9sU-;_za!ifa7tbyiEX^?u
z*8xciR}on*&XA(ie@$o$A*OSRmrpire_AOUAL#BD>-MZeS{R$ufndA~edXuG$9wi|
z@;-0szQBR%gTIQd8`ac^xwwSEt{&n0xFNLlp>EOhLX$sKhMV1pj({kMzaBClNH0C-
z{HEcFDIYszV7<Jip{WGoeDy&+!=ClGZJBe2*2N~fFSOAu*vjDQfk3S1Uqks2UVW;k
zpLmCD%(t11rP6Px-!7J({<m1m<kZ)_`KQC;R_VpX-=yTnCpMCs3`ggb?rSBn!jW~x
z7m?^&gvn|Y1HZZK1}WZ;T{7vqFpk`NsfQ|~KYB!!mp+hw^&QQQqbbIk9uRjzj18w6
z792lFs)&BNU?A-$A!At!b1!QmTrl`gd3|s$xT1I31Ig*wNd8OB%$$AaY<Ki3^=TLn
zR;)VT36U+K!vQ-V7oCZ6w-=6lRc%x@Oo!8+!O351Sc!v<ffN7|qT<=pLD_Yhb!F~8
zy7g<dVxXR&&VT)Jtd8*N`ONT2PoJjAd3^1`t3MTZL2bVu&v^f2?Ad8Jh&Dkns_tUt
zs&WUF2>MUkC);+j{z}uidKa^kg$lw?eo4z>y02c3e>n%^?5IYsOT-R#Gf{d&79rc&
z8P#{(A-kCY+BEI~vtQQBHu`0${rW*uo)B;Wa@TqI=^SC}^DAren;mcG&qjkmN!oYh
z4mk8T$e9?)g*$Jak!yl>c8j%c#U53<`OQk^WV@u4Lo~kM(n);j|MG|qvOeS0YwO>Z
ziO8o3vaeOVnYz(;epC{Q;-i)M#PE=62CW2FFmr@NZ2Pip3J@26`Ok;J(H7;h9q2aY
zJ>N}$FdmGKR7&|+3>SP4yY5{%b~Ux}zAU!2nKD1pOcU-|U&gLb6MMIPH&At9ODOVY
zZD?3D4g!kePh#p6u|9YMKExM|Ih(!2s*=4o;A*ntN$~<o32#6TcuO3KlEN4E-#@yW
ze08$~Y9(N)<21}S9!TmlZ1roRb*4XAQc~q9e6Q7>(5p>`3UHP#9ZJ*mlPp5wS7hGN
zBN7(B?cGmOkowRc?dFl@x8K-X4L!c<Gba8-r~H{fQ~wwjJZSJrt8XP#RB6*ldU3xr
z(N0FMmrU0`c)Eodt?xms+`ULBwyUS6L)x*XY%vO|Q(gPLVtzNe<kC{JxjKrCD_6$g
zShcLNi6Nc*PubP0P<o8+R#wwguQ-NT0J6S7RSzDH)l%xF6P|}Wd-JOeeY!O$CuP)X
z>2b9QjJ~qJ(yF4@8>D**7e3$%erNFM^JmlEH8udt5HfL{8SE=5fnM}eNXyL%KhARZ
z?|z(9*X<5o7GG>kt$MR3bl~j+Ha8-1qcjfa!}N4`X}j#(S5Ga;Jc7J{BoA*iWwA_m
zp({&-;etjgICiFAcu0klzI%8dv&^N#p&G>TZv%ijfjF3Bi{JN1Z75xgWebF+tN#<T
zJN1hiCUR-ry0#}``~ul%g{9#^$b2u#R$fJ9*Hc5cCx!5(f?syczPn$c@QY6M!OQ*I
z+GNDD?dPe%hb=X%^)lcupYAQ|J1ZvU`waH-@M=Gm=>J-9*2M$=`dr<L3Xc14y`^7Q
zBnM(|)0vUU7k$II(!zJEd5NAzcIWDW@|c5Wut68fxG-ky!88;1>V5+Tr|=L~SD%n5
zdKd9#(%+w4yAKHu{g}K8iIylqJWce^L^0|8IJcfG^^o8i$*+hR&1IP0UKYkQ@+5M*
zn1tsqN}8D5gJ<UyM&RV@8>ho&mH)Eg!`2hN&a5?v5@}Uu7*adV3rz)Tb-gXwbBs)k
z68ut*#~*gI;uCO3gA2J(Q&&ztjZj6&C?{^b&1&kq%PDGTqj4oq5LNY0-g7yIj_vCF
zSNf2@$u?dY^?uBOB8N3P!{%3!dWu_6DVv?A)LUB5hC*Js*)PM;Mb`e(n9qCeY}f81
zdK*d2KRUFukK=j_b6ZEBtWpQ5y&L-`#*$$4lcU>Iasv{{Z1I3nmgFFpulCHXbGUly
zg0gUL6=zQ7=AOo+4J>-^W5D%Qy<yDA@V37!dnHr7pn4d?=ZhK>Y|1yV*`OXLJG4+9
zfP9^^m6pJDaXG!A#E<@Xogm@6-Y_<%SwWN$u;AgEok!C@ml%Wj<<~5ExA(lxA$2BJ
zv-`8c=C2+(JJvsPPNN;?X(@IMmsH+-98pxb`4^7pA#cI4b^b3%9KuhD{8h*W+VM25
zcM*ugFMjM*4?3K&YtB?h?@UENy;JW^T$F*u*52%dK`ux=muH1sqOVyop+<*;j9za6
z@rL)PZ=idYK!tO67(8T?^08rUI9=!vdVzoM7rvgGoJ`w@(t2{O=%IJ+Z@cvlJp3d@
zdK&PG8ai6Uu8dy!^(Z0TQ}!=3jC*s-PYGMU;2eG;6eT{E1BO<wy;91{1uBt6FY4LU
znGgIVpKQyQZTSJ~p03#`?p#_`=7f&Te`^ICIwPhvddE34p%mpygnvNukILJZLI+n>
zaXu-9r`}UK_qe<5)HD0=cd%-NM+x<rf69k?tn6sd;x3RBv&)jqbQ$sjc4^})jO@K`
zx}?kwcfG?G-Vjl<=AcMfn8ho6WFLBYm<XBIZ^w0S+Z0X-f;AXv&dM2`*HOBVRSmds
zFD-;Y)ru<m;=Y_IOrd^LyAQ9iv`yzfT(ClVwW>HTM818n>k&BoUU5^}W-8rkEFGdK
zSz~)dY4(%HtMK>CPjR^!Y0jJHCe9*r31WnSO+b?1@|o477SVSmee?c3fC*oA4dbS@
zX77Fh87iI1@rUSpf{6Q1HNV6oRu_S)=l1TC9igatTTDVM!;9KgxpBCJO$R*SW@p!F
z5Q}oh?+*u+PXIN{Ml!_m<90R-B%%yRfGZ+}F?Rogws&RN?*0SSx+MwfWtk8Az~QGe
zuH1&F!#KC{;;IKfmBsFBRt?^8#sp-B2j8>zrF@hK6?m@b(H`;lnVkLZ-E!)WxkLKA
zdX$Zf7>ns3#;ZhXGXV@K<Xrm1$F`ukoj>-o9p+9i8Xp`xnto928JhU#0z=5zxv{5t
zm=Dqis1OvFAu_1LYt;)YZnW)$Owvw1tuu*q?O}CCG*PT&JTt{P^K8qf?G^+p|5vV`
z7=pL_Sx%N7kHS1H93m?rRHfRuhpdWiemiWruf6Z}mnkCiXJEc*=nBzxb<A%4N}A4p
zv}c~FUPJoA=tAz?6?`?DX$K63f}=|(Jj;()IR=u+UtNep*EU`@hK0|&y&&|8W=`II
zq0=woumo!yl0}3d<^wxna*-H3=*kGmFE%MOIV)L4seiaoZ$2?HYzjvY#z@#3?vmi{
zgHZN=W-cw<uT2Y3@9-#~UN~ULzHTTs=Rh?oxA7oI{NrgPJVL2L(~Oa^sie%D5sXh6
zWyzniN;yurXCqtQ>r5ogm&S!4Fq|3@c4X$@aB-ScIOK2f1#4{LRgprhIA=6;v7xF@
zptp%^AVTzgQ+M;dQUv+JQPjk8GkRrr-2c6Sgrz<ZBHO%)rh{gkzf5r~|H2bZm)sTj
z{{*<dyg5CV#M#m5Dn_on%+kADG}&ft+zIUL;1^|9rah~VWbxgu+OZ1L-?6d-Uex*{
z8uJo~`UPIA$TX)=Z)Pemj1Q;%()f@T9v-eR^b{vMK%pHHB)=Y=M&tT2L4HA19Xd~l
zl^v>J3?z36jN=Y$`J);BjQuVRZ9sY>lTgrA2u{K<!Dh*a*ht~;ot0vjTa$K(qq#oG
zmO+PM=U}#@+Y$&<C??`2-d(z{IOVZz$#_2x>ZPQ1eW^*=_ayrbx+s|^t0V3Rr6YF&
zHqd-Ibor01UzCi6QN)h($bUr#eUkwO6x$x<u(IDcK)Z;eL!CibEHC7WY^SyzpgF*H
zBorJ~ue%{-G)Z1Uh#SmwMqp=xH&<VafklD(gPCMl1`$-EyjveY!!j(b`15lLdy}c1
zKmH@zj>S#H_~<)#9UQu)_B~fcQOfaMUgk-{YbcwfCAqBDP)KZj-cNS6(HI`0#S_gD
zSa{rrv={$fwc7=U#&&JbUq|W?_e+wpe+ftZxMlV?Wwl?=$r~riEdjH?ueP%Zw=+@x
zH1jF4FIUMkcnFs?9%H31Vt-Bs*+&sUKhm20uCNRA=+-8ZSl4seuC%+P<8(v#{nH&k
zBb3Fg>!yc=@w(SeLdNw}NwmJKi$#cLDrY+j#K&(V^2mio&g@hDOg=n1Tjlr)CVY=$
z4dx;yJ~a1f7_Z)DE)Wwc5_P&VH1)EoNl*crF^`dfQkaK|b!Jy-kQb_Hauz+upP&T9
z{bYi4OgB6FqJm|hrCO;;6!(!7Lg;>2rRMFST*!Jh!f>Un(9<|}+~HO4y88D|3l=Jc
z!c6@U;b%Q|_Jw}q1Fp8lAu->5gGU%vrV+G=*UO2CK!r=wQx1I^Jr^gmcx)v2os67n
z;YQ*3Q+%3?_zfAESn~OX#0z|m?U{B+wx977rj*2;!u3Vu1|1noNGavjtM}+xIiJ^$
zR#so+I^y<1)zQ|ka<anD>;=h-)+f(v%i$58>wzCKTR;RnMtBQ2J@-TEU%ml#;WUCj
zn37LqG+|pwXGNY}FO*nvqO}YYfH?~<!|6W*;(neJtmu-{G8P_xx`4kx(7Na2oInbc
zaJiA_^h^*6OV`)<F{A~bCXUZUjz4sD#||nuGiv6QF*jR_mG5egRYR3UiKoy!Fr-)^
zsWP}}xQR<8PaCs(jFQGaP=|%#6<e%XL?W|;pB3>BYs+{OQ^$<8yGfMW=#V&wr+q%1
z4z=+|TZkL_vL)~#?UyS_qGcu4P2t!G`?hrdRpOS**oQ31d)|bs@QqwgEzI&=<9xgI
zm$sZKpHHz@;DIkqza91~4gu+1r3m@e6Gu!(UqIUR_x#>8erjndqfcpR*v2^Yly76p
zYb(6cV~ys=BdTz(AV++8v(pQpN+9t2783A?(T-;L_4E3#uV!Z=PQ1+P>J#8N1s+o|
z(NBxDu_==r5kke{_uW`@uOR8?5Pm}@;IV^jbYJpD?5#{C>Egm*NmBKK!w9XX=exOp
zW<kQEn75yd8ZxG0cB_w*!ff0zJ3x?pf%4fZ%|;+x;l)O$$N3$2%wESIx~2i$67qU2
zi-e`ggDL?@FLYqW_qeoT<1Z`p-kKlAFQb*Wqgb$j##Lb#L-qN+_==|H-Ssm<8kyxp
zmhsCT#k4Addh&h>djk$G>Z#EhZ?fZ*q++-$8k|YvtN4zOv!)|3Wiw0s%*8|E8zrcK
z+D-r}<NxCV6oyN(X1#f2XC>XKG^2|*BwOr=L)F=QTCAd@R`+t5%m!&s-w}CEUTzec
zmVk*Xu@h!A>S@w4G?f7`I1s*3Pgm+sgQWjbmC>{?O06$95UJfE6LmjSW9`8u@rk=b
z55LLCI>UXf(!)a+=7l|W*h9K41S3(?&|o<sjkoVHFo5ZG<Vcrl6Fa~Z>%7E{Jn~-Z
zqI{gkN&y|K2oqoNN5ssB0_Q=7@hlUZ^Pp8sLM6^d+7$6C({*ninXjWgTH{1I<)&rV
z9@MiCPmkf_r>^GB{I3Ux89<mQoVl6wfjTw%rDBI7>!`=VvFX{j*J~8iapN^GK2;7A
zmS(BHiGJui3VRYK6tDi)buliEn)vjm@RDTKw-GA96xBQtXgBJT@ehe=UF=i9sc2r~
zDyPwJOZon>)M;K*a<&1#5l8n)>z0o*V^|3|-ZoRgJbZH;QT5?v#{2Mg(mu%6L}COU
zRTgk8a5Zi#7Wvy-Gg1Z(5UPN3@j<}h$KE@Y5Q+Z=EH^I+O$YsXz__NJq4*jlwp!-9
zWhW>k)<0S{79|JeuUwWHA1K`x{(92ie$?kU==bKgO^EDRQO8!o^Y4kegQsSiG^l6i
z``rQTth`4Xior^ed<j_!k`aP*1+a6`4DIr6yF-OKFVYChb}2ONo182K65wQTQ#~_O
zUkhG`YQjFfBe-XYg^rYHvAs+gd-p2-&YG)y-ea`(GwmRC!xvSrMk}r(7ZPbf*kCBx
zaTG3TzicnU@(mW4>2F_m81GS%;+d2aO`nsmx*ABQkyeACd|72(*=9?4<g2sGiW7;4
z{DC~{tx_!DGMns0Bl>ITnISyPu>5vryfps!x5X-q3zh0}38c^7)#gWprcRC2dSlx2
z|D9}(K`ep<r49Z9N~Qyp><jtQ^9=cjUswxobE3IXr5s$*V%2h9I8xbQN${E)tFaTp
zf7DYwMb)ZZ7~wC-n!Bm~W<8Fu_$RTDEaETdpx~dmuR%k3w>C&Ag<;|>G$#@w#2rnv
z8ZUM1Hdf%;q;@JzIgWui3S|{uc=*!_bb!p)dn*5fC(W0o&(7J+Mv|W4NuJAe-QQY)
zM2LAr3HY5I|9ZbPQM749y*jeV!<_+;l#vR^zi~TFSD+x4fmr2|BFGuFPy$17w#;%h
z6K{N`!bw??*C_!(Yhy}EdCOSD5zP_lO2eCf;%;D&v1lsLkg!t2(LE{HAABk3`GPx(
zvk-{>bw~a`OnNUI?21>DV`%P%MGI8fr%7oM%L<(5eiN4)GBqk85ifr9gD#}6UySf1
zhpK4=ztuwWot-6y<EfOCC@9Uj)32=j#^*1ai;gL3ZQ9{p(Luxol>2Y8kh}n<-NKI7
z9pocS`wH0kmJmueG{>Luo;CXSo6lp^vyIB*q5CwxFwDVi{OPh+P97W6Ac84Ub~d2F
z*5DZ6k-+wzo4GglzJLLf(tAz@TRk&?R1r4qON?gDWk=)d#fM|8D}VOh=EtB3bwvsK
z4q&mMFyQ<zeR>^n|GokXyKoHrfyV5p_8FeUDTwjD4wiXV<~EIfQLZaJ-fX8kty*`j
zy@Euj2THWFvL*xphuA?dpr<qNG$2t6w)H|**nX-6Peh<@lwn(<Y`4bx`t=33Nr>wG
zUIrPcKJpQh=p3*(8lb}fI`(lS22(Vq^;D^AVyjaPY&{>Iu(ERw{-pIh^RYR)UV=T_
zWE|s;;ELhMT#kga9}|e}VqF;{CdG3>5T|<i`#wDD1v0*;Wb(WjX&YhnZ85gZ4cy6I
zAnHId#jJzUM8BeGz&VXIJs<=?r~;MU?X?6bwo_Nwke1WfK;41luTZHe5{E;lv)+Um
zmyIxx+J&muNrXZ^5WbE5{C`kr%UMya1JW(B`a@HNZ9D!I`>WCK+EWh)&k9r*m)J#~
z+|e~+da`pwL~Kw2(P4k4k~sl_%E$=3u>qlZTxzi9Vd{T&T+g=ae+jbl$!u=fi$iRV
zIL0V60C8a9cgib#aB=B#xgp7P;!U;R#&RT$_;7u>^-yjhWtYH^H*OcaB>g6H>`EYk
zeMDG*AHrI!#dI^Vlv4PC@+OuhYqAGK*^+C#tfg_HLE<qd1A5jBWlW$8>nXn`biJ_1
zg<|vm)jC87XE<p75`C%f>2<u`ILbZTlrC>BMMA8YZ~zE_WUhAEV>3mKs@EUDEQWtl
zQ#uVj(^FTevEJ?Gao0)ukY6AJvIJB=HjQnF)ry8EcGFK=XCsU{s}5GG4Q?c&oomhw
zturE67035wkJ=7@foG?~zNSig^xE<7sQOG=6sbEU*x%E9<U`ySZ`-@wj_prdt}sUq
zNA#+OpRBTlK7h<?i-&>g@xpCaPg`<ldw$iBI(22(Xewkh&=$f5+ADG%xT_@p4GowF
z<3l5{XXrPUB&5EW@X?8TXosB&SWi)I&$zm&wD<x|Ku<-sa0O@>93yd87*-O#<<6WQ
za?!ZlZ0%WtsE7~4tKokyDstLGdJZZW6IlKYc+HUgf<@JUfJkgY&Eky{8hga@%c+Bd
z&&+BZM^T)`O#hsC_|+zZDF^r>yi$K4P(V!NVjqc<kaT79_=8(#Z2@O**!(pkmSRXm
zINWfh60z_t_tQ*ODzSBppUv}r9`ct{9!)c-;aNbm_kbrqo?hNA1jFW>iN5)zoEpzR
zSebLtW=mp;XZ>}R!A-y=C5B62^$72Hlzyv@iS5&l#7TaVn0!?L+wH8#@h14x|8Fv7
zV6qMZbRG<Ax_g=39O)!HZGhOOf9cw^V&x@xO2lpfY}#Ezf*Nk(#XT1#6jMR|Zp=S#
zAd`{Dl8-R{9bkezaB8tz^Z7vS?9J>={YwwMG}~WRbc`VYfQ3H_x?S{c<-0FJ`dz9~
zpLEpSn#jt+3xP^X_@n5e;$TeQhqR3*Px-RaVty}$9@c-8@zGqt5u0K@yi3H>39I51
zhg+^lodD}VgTTW?T~SU(7+_gzMi4OJ)k|>RKxx?|ZKCh=`M@;A750(FN))0^@3%tX
z$`+?Pu>&_#HRHNfGlx<@;5s%lOUF)pLmcpz+sSV_GmHpYQ+gb=-EDJXFOEfBa>e2P
zHv*1kycC3zW?+Zgv+dwW>GVM*-XC5%d^~p>UULL20{Wyl<cX9^^QogfGPvPD;xmDV
zZjbRgpdg1k_ACaxTJS8IojS<rY*+i~+P;(EqYYms{d_dFOCrqqYtKUB=e+UzJ=qQX
z8)fzc8l1+^o+n7CPxFj$sK#dOM)0F(V8KItG%M@ubWL%{`Qr?tV$*|l-a-IO@A=9_
zFJeOqIA#)=_!DRj%BZ#c#8B|HAL}UmLq{|VbCcQQ4rW=q16P>4AnW{w21#O($&jHo
z<2A6q3e$aP%UBhvsI_+F*=pgXibW5V#M1kC4Cs1FxpG7GL1#>le#^q{#$W-$2!zVz
zx8fIx7ys5Jvr&G7Zh}qzwWm*Sb&ZIW`z5mR6uA=o9G-7b#wAxK5F67XV*s6LWS*i!
z#xzqRP2YtoEQ9+X!BNt}vJ73!4DVz^q>dZDl?1t6I#|rt(9UdVhaNs@$GGG9IuT43
z|E?IA10TCFqerH9RKnQSbV3B_H0HyqDgGbQtIDfr{e4oXYE11OtjNuTn(I~{Ha{)5
z`ygm;XJpSiRbKEwe1pf4KU^hwMy3BQ*Npj$el|4_Nyye%P;@*gUDj;Pg)=1tn%|%z
z=@uK{O-)wt|Ci`Hk~@llnbM<h)G2>Wh_Ld0GHftN-H4GE%b4w`2n>yhp_!Rc7&w?J
zJGZ0h4uITKiJQF&V#TwVDeMf<*wULhUZB%{bd#{VolmswJX9>ZnM)a6ka14KCEeZ+
zMUPV0o0(k~R_R|cXEA{UYFNkox?RrszCXC<YfmVqy_fH$lUj_j%7%bF*Ivoxc;PV(
zQL6Q?sKwF!ramS}jfu0h>}E0u_W%j#kK_5q^?dhJRMjPB2YEddf9y$Y2ifQ6E{un$
z^_Z<;L4{Y>s=eA2^~<=1rkb;Djhcg`#axOhw<96ojo`P@V(=HpY>0T}<rkRP$PmY+
zy(uzMV$sg-(W-+%0~~es`uf&@du?$N;o#G})yNmdcmDg?8EsAn&aZmfJ=5{^`I#D5
z*3p!^JEA2I&u65|5*ZM_IHnIt9f{GGm3AwCrR?&|2J&;b>p=C5%URcDGR|Ru4hGvX
ze2Yt2abmZT{q$MCs``U|rX_Tk?>kN1L!VRAFNRW|?xS^Tn+ioRSK_}B8^NGqbOL{R
z1||we3sRtQ_S{>2yeRwhCYb$@dN^%t{p6z7%4T)4IlJs;fJHQ;VA1G?V5Oh4h&h4u
z<ZN2Mhr$H2-R+Db>Lx7R_gXkM*UiwFkk%IN&cpKvW3t0Z$nf;yf8L5Aj#Y*wh>=Mk
z<z7V0BV%rocI@D-!UUQ1R1tkfkOL0Ts+0_TyM9wVF3#Gc4=rngb7ayjFMV7UruJlP
z(TjM(0e{?y^qhv7Igk1D@&1i;LN8p8`=lJb39k;#9zImB+y04=Qh31xc%OA<>b|dR
z@09`v6kl*DLa^0ZT{#nZ#3U=De%%;6cDGMvb93dy0GhJwddvDM!3NJX+M%0H&&_e1
zA*aoTjLRpnUHOFOh;@a|RJxC3cd0b&wj-=(HQXBAC7?85y~XAeS6LgJ1322e_s#t;
z2q&6k7_YhE)?XtmA{Bi}h8vLee|`SDUxFvFD;cgxnzDeiduk@~A``ypt-t_ulp*Qu
zAD0lAyH&ze_dWH>Ft#JOb5q!jUv>DQAPJCoPRPks{3Qf=7W-YKjDo0l-E*U-1&=cD
z3>6O!d0(<RQT(MLk$$;I&Sr-r_Fv0iW&-=qxOe9Y*YO)SX7NjJLLz=?U_72>twF~-
z+)wT*r!HLu2b2-lpcIA5fxgR8YxS>bVJB3k4muQW{K)@W+?autU_CS!o{Vsj#tQ)+
zcRLs&$V1nXvpgKtu{g7cWB+^sH|y<#dVy<TF1l7A+}%y1Okn6_UT3tXLEvbbbv7)#
z6)5`u()i+^sB|&fSeI_?-%7sge%<Fz)QHg--!(Vq%X4IcnBHEKS-SYnrv9;#aq8vs
zJGX!JwU+>-W<R}|@>eGmLY@3EvS|_9_7IB5&+M>1Y)SdZ9Z|G2{dHz*pX+5q-AR{&
zAohbk^ZBox1~pIvtf}gk28jcC(>?n|k(7PDlR+vg+ToKWMf0HAcb7OMV#5#dY7aSf
zCG!JY4TNTAdeT30bd^4&hqtPEsK|&yoiCkCp?te^OS2zQTnNjv*n$m#OLUX1oqqV^
zXVU5m&y`Ap20~)5Y#_d(;Yvl`bG07L4=$>@--9AZ6MZsU0<;2$Y42qWq?0Be8?v=3
zlx#L{Ir~R*h2qh+VgY^NiGVlrqL4!r)H5y3C8ZN^X>)~d4f<$hJ9`lVga?J|?>UdA
zD@*^W4-hEU>K8zkvN>2Doz(IP^wLvdrl?Dp+1yZ!efcV)p`{FNK`~xIL#ALZq3OhU
zBDx467KvoI`ui<gBEzGh+Qt&91A~(af9-Y9s`!C*u)i09A9vV4JFV>&@r&9ODYL=m
zGflIW+~Fb8#1Fom7VtQqP+KVmW|{Vj>PfkkZ(VNK+g-U{Zc{o&EZqK_Zf#)NdLbhN
z!tssiDeLxy`wkpBsG#e<JF1MLkMp>?^CRx}dik5UAKvEGWutTOkKwNv_YZ%X7X$_3
zY(Y+Ma%r|5-m<PFzPq9#Q;hickv4XA03jWrXdun`>nhPLka%SL<0CdPv0@<RUC_!V
zapCO@LG<Nqrsf-N^tq=GZc3NOA0IoI0hUAQL}9~oOnNjmLecma``G@f=8q)Xb1zH@
zMtHVlG>I_jm?42WGXK|CA&x#bXF|oc5PpFJa$EPi85K}o@1w)%Lz70UqUj>SA_!Q-
zwt7BQ(fQ&&^Iw}S1MVQ?vRX|fMH^Pq7~N|I)Z#%YUkT9)5DY75Q~ZH;U8|OcxAber
z$($YE2EOIU{$mFBWCC{u%<&I<GuU;-*tT4kK#|=9StCLJ=<z(n#U>Q_>Kz{FsisX&
z)Ak^1YMgN^GdM6V{{S)}p{uWd<7@73DCerbqwl##&&m|;GXtzqw)K0Kb%2K~{G#Ov
zlpCA-NuPIgBl>`}t_yuHX0EBC<)1K4S#Jq|8=(s8T@Jkb$B5R440%l~)idq`{bN$6
zCiLz)MD;_FdoXzw9D9>{=SR*7*Js%EGQtpL&^pl@_>uBfz05_*4Kp`DNvM?S52q9@
zav|Vv5CVaBUPCxQ@3fFv(&ncedPB5L24smhzhASM>(RTul&InS-hOkFG@s4qffs)?
zUF}N~ewCv$P1jRx0tGt0IR$_8g#$f(gVaT?EipsIV0P_Mz$FCUxpA-OYJF}USgmn;
z_vO2YncQ{Z<@Np_<y5<t0C_#_>fBTfE!3(kccM4Ih;ru{;Z4W^qoU})sg*W!2?gm1
z;(gMZccL^Km)`3;NjUPQ-v%!x3a5QGymbTmjNi%f%X|f=?glnn?`kk@><im>8rtP^
zwDo$v_Ennhrv@@Eud^0obpy>G*RVs6aG+KG)zOm-JT~3v|8qgj&&_KDrtmPj-L=*m
zkVmSZdEzw$aPj#zmvI(k9c7=rdAj$GCdV8QbUOsxbaxZoGqVhKc`y;Su7EFZp8l5S
zZCq6>yT|S%V@C6<o&f(`b!dOMMSHh;+J9c~1aY*Qr}gy?C}(U4@KMky*EW(vE1!xc
z3yI*bOAdPdwZ_g?kfZd$T6L88iSo$~$K6a(@69L<bZKQpk#_q$4NGOwM!W8*hWbV&
z>mk7{HJ~jJ=n;Tybm806Zt$^uKm%5`5^HR)03qMEwTV2^{;#l}{cnFB)|_DSY?)FK
zU%l{@x!Ym6p%GGQv#Ft1B!gqfRN-{%@VdXs6$#Qnegv>xNV%ISgt@jpG()0^2*Ey`
zNB_oA>@mLt{M`gA)i5`Gv>W}ahA^S;X7h)ZLs{=fBEmQRbyY?~9n3uCffmUx-q5sr
zt`Je!B5z`Xe=>W+uA{$~r<+oRSnfU)g_9kuPyuB0Z3gC^7G@(b(BNDYTlv+e5tzlu
zJEcwL4kx4SJ3`b<R$h*8)L0}8Aetu(0Q-N4nh{g9eVe0hf8kdof53)uI#^hd<!@vX
zZIra?d#8<e2dG2FIUPV6e9`5N64!@o)892&QS@>|1V$9DB<awKjfInA3FE2aLsn<k
z8XI*bc8*+HR}#i1rI3g-jQZ2=ld{3(V}B?L(!#rZa0cQ~`E>?1{J3pe@=HyW%5*U1
zZSkvwQb3edK2E=Z>7h&?ywE7V5rY>qXk}<QSW{@K?U&|M((>J6esP&%6?+%6={Uc6
z`B%@;BBuBBOtC_B0%GybTp)ZMj!M{Ts1!#i|JDo8<BJ_u?sx)I>J%wrs=FzOpU{ba
z)x+H#^~);MbmM~RF=A6!bJl$WE194OU#l908{I}Ayp-!dNV8&QV*WQrOT^&-!r2B@
zFCNYq`Dl8-vn#v8ZAfRSci!9}F4n@UvWIZB-TGXWi+(r5|Hm{dDlpYIJUJ5&U&N)R
zc7u$Jg@hBScIm5Yq;LD$dee22A^ILvH!0TKUN}}1n~YGh`tX5x?N+yarOjsf0pEyk
zXx>LW-Rs89Z+rYu*zlA)I-5}C_?z{@FT9n*Y^i<Yz3v~|lt*McKPX0xk`sl)?M14x
zsK}%cow*VYWqgw^(S}z0q<|)`6{SVS3<i&AI*$nFGI*dcRUz4a4Zwyq4I~Jm9qUbS
z%7N&}QJ}GJ)w@Rixm4)A$+?!{P>*wmdc2s4witlV4|wZoJeoRw93>0hdfIKV;iuf3
zHf<gbkNYBW2N7nTWZjE0j3v2STwe}=0#}kJpE~^3MPcDPff&7Y@7h`4hz$-rdn{Br
za9jd)>~5M_OzE$$z4-eqvm(C`iV2?3;<Q0}+_;fu9$ec;tinE*ejfYWvQgjmC+*QG
z+DOTp{y-m$3X5Lw*3rM8p5(+hu}iO27z4bmH+@({4pjG4D`lZx=pHSZ`$%Hgi<16<
z%iqAs1v5>`GZxxBXTB8I!h8{ylfhH^1DU}cn%;aSKEq=!P5NJXbA-=9gwOA{g>SZF
zBhDoIVv*kWJ^K&cE_B4VU=7mlfx~>W)=hEV&XY1i>dnht#tYBk>UES;95LD@Iap#}
zw1pT7k9&6ikJz%;rXJq~WP0B8yLhLq1>N{Q@I*aLm#<@yc8JDWs4m|h7rJ!Q1vz?_
zVsa0<d4f|)AD)39ZG`a2Y0hu8YfdX$mh(Q<cM)IjC6sM;j-9T*Dmdf;!!gWHL(x)4
zGcT_baQ8a3%*J|+Lh|&iDf@PBtL?V$cKddB8<wHBLc$j>*HEaQY>pO;fp4`9;#n69
zOvFH-+xE364oP_GM8Ans>+{)8oI)?%x#E)Odvz4+jVG?nv4cZQPh{CU$qyQ!U4WS9
zO^@7tl4zI^^qAs;ib2c}ddEVy?<$hcb+um_s_)B!z5#mwLbi$V8|7nSx!UJpmVMs-
zhxY4P@{%xGfyek{&ty{C<Pvlvf9&KEIykGhKW^%D{e0j@?$GBB#%v%p?c_ySl@YpN
z#w1>?>+4sCH-D!6nxYC&fQYumG5{`sjKh=M6tI|J+DA?oRPc<`9iE{W9V@A{Wj)!n
z{r67__p3&hao%%=!3?*>@?wTxrv2Pxd;v}2MxeJV==H{xlrIsv@?v#14{K)ZF|@aH
zOGZG=7ac-!pZ(-eo>@BLaUAabnt#rZ9Ub_o8`tVX7P4k%O64peRAJ;<cmOI*fuY9T
zd4xS&KJ>2<-Abw=EzWWf52;5YFD=qArKnxzBovpFU3{v4uuTGl#eVo>hY)#erZf9t
zhVzm&;r1VFIV~(0lrWZj>UQ&|dcV9hbg7t5m9Y3A<D3f=tNJ0Rh6$cIT32#L5<CFT
z6O=Qa9!mk}%h(C*7Gu)}hqzvy006NdVW3I`(pgYm%o76$PBLEr{*{b#r&gok^W-^h
z<HA$`+(VE%oJr!Rui`H}HhbitQ|)f7+Qj{@fx39WJN?mkdUIWS7Tzu1l!fi(h>oSb
zr=sZbOKJvV%K1)CiwDEHPBGJ`FqiQDU*EU&MuOdzWr1smg$rYz3v5g7r-6Ucigb6!
zIHYRRQF(ks=)j=5b5*EP&eENSV6&iUh9Dwfe*l7TBT^WxH%H%Kd`+6*EoA^{25zBo
z+RE`c-`U|w6L9$?P2}-)Rc<$!V9TDu3={7nNx4{Kki(SzT*Cg+udmZrtkf^k6-aHK
z3}jF{?&+&T{>SDhEP|18kh31k+3Abt456y*7Ql!&hHp^uIzfvYEqkUTeRM&hr`P1O
zb7I1kcf=`Eg~Gn(21o7#rMHVlyQK#}a1Qx>wtJ&FHvD)9H~NVGe%;3~UL3d(_Jzzj
z^=B=rsq6x_;HMi06;a!>XeDR)Ie<fISzAomcW^q<j=UNYD{13$`hCdk5+(9Nk|LD{
z?eLP}=8fSg*UC71U5!;PGQV*(3y8`F4m~Hek`g)@!o6|PGi*7WYmJ3tU$^B5WVRw5
zxzA~sK6Cf%hG7n{410lgJH_Vo^zWazP3oB26{%}Tbw)>OX?-^Hy%LgVJUpXeT7vYs
z;C(t_2vq=X-ZHku4PYO>-E;e63>Lq?NG3S4JcL~xJ19!~4GIN5o--r_LAA2xO{aYh
zr~H>(!mFAVT=2S$haob!(SUpQOZn)~eOogU*t1{)xo=j`FCQRyAara-(jZRCHYvBA
z%}t&VcoSN1e{+3kQZ%ve<`DOQiDZsNM%K;`|0hg5pt^JlUP9?aym&>%GG@SaO4I4g
z&V2tc1h9M(?fJeH5L%KoKnHasE(r|?DD*mYy^(a_hAYxP{zBWzq^yLMr}^UsY?*M<
zlA%JbY`7$W`kJ3?Y_BMXL?02maK_#1w#3z3T@w~Z<7gXg4gF&q!&96{97!Z_7ng~r
z#Ffo1EmA<obnjfZ?`FhdZJDUH{#SOAt8YY4BrVU7wN-D(gpO3!!e3ObsruQBHKk35
zI_z3NcT=saTHfmfCwuvST+Jyjp(Em|rOject8e2kB~`s%pK8NhEdu|artW5T66cff
zp7j%<?d*S%kA*qBGZY>y)q#l_2;8`ORIO!Dg7s>zpoI#jyd44KQm)(@f8vH;2${ar
zURZcDdq)r5kI8{jwPT9E<8`nSd|-`U_sdZzSl0qysI4)ul?$Wop^v{Y-`0kk@~5Cb
zoSn0zgpJu)vj(NFy9i3wn5HSidOz&^=pc3#On?LI(zfw2v5Ug{zLB=4VB)mczqY2l
z7AvZQN3Ugr=q@GfMWYQcYj26R_y2Hq@Ro4whE3e?J-grvMN35FV71bwy<;(!LfakK
zST($;&LjO_1nkX&Tq1YHh5fv6Px*lAqoY)|alg7?vH8=G#>X0=y71ZvZ?gT2D!PcX
z^g(?@(PCz6!&o_<320p|mDXE5tK#3{AFGPe^>rFiq+y^j>7f?u@rH!hK_xRLlFPX|
zp48urSYJ@xp66-(`U3We-6nM<Nr3;;rPUw9J1^<M@0q&(^3Z<|Lttl;CgxP_oy;yi
zlG1EWcZhc;H-9R82q}6Q$DRQPtYuQW4Q;BJxl-?#$ZfKNKr}*)YSSv0G3Fcl-E3jS
z&(?$?G_)r||GpQX44|<eQRLnXhDywg;QzmoYq$(HoOS>RdE$f1WwLDgCz;bC_}uP9
z$fi%otd>X4*%1!p)wzF0WOO^+Ocx0Y>tu8bDF7p@0Vt+r>q;xVd0N|jlD2mpy3Pjn
z4rca8yA}P_GEoWh>28XHZ{Hm-doB+OC9Y9QzP;a9$IcG_Uv!OebQX6(?7{;Qib|BA
z++^H=A8C$6oc&jSqj3bw^o5c;UTlUjyX_}R=|-&^26WCxbf%96Lg>HO-(0?BWlSIh
zl7UvM>;0DS+SCLC7jk2Ss5LDeOF!Y1?{cm%UIAOeotL1pgAWyS7(3^r2oAUHZ2SoX
zF05NJGAALpq{m&i@%vLE8`5+Sdd`~+01dszQ6e-HjW5FN50?yfw(`AQ=J<0nKTkjN
zk`Br9kC*5J5|{X<gOi2MDT8Afw6y}m813k;R|(SIoJTH(LwOhO@OFwy7iZ{WhD9yK
z=6s}3KRR$D!>{lAb?9}oaCw;X@l;Duy*I?^yO|!I%+G~K%tn;$=Tdv`-6k=1$F3!l
zT>p4iWBRB80fbuIJ!Z7E`*mKDewi*!QN;~Be11I8r~ksjYyhfjpk#Dj41|+Xm=YT&
zh8aR)2*3~LC4Rk2oGZy-l;_}jq)r<cv$6G;Gn%x%Q)R=n;>vm>{vfjQ4H1raK8CRn
zG?2|%>Oa@;UlAD`i(CEY`EW_;gX^5%upyXW@6qAV-`F;~OE>^N{D(G14;=C<j)c09
zMNff~h56^#@;gkeBy)=rD~{(%$2>E-k>+NBx~7(qNeS^W^bx%4LlWePF`CnD6dpUL
zdbyVp>e)l{BEwpTF6>2gikqo<VNGx+2P6d4|4>prF?C!>qkK7t4c}Bw*kvy-Z5#f2
zXa2a&RpOQ7f9R&|%yh|3_poeMbtarPl6|}__kSj4C+nA;0BnrCPqPlNQ!7>b9(K=T
zS#-VSp?f!NEvKZO=1*DRvt)|%%A3>PeY=n|$$I6^sXJ56KmVwC@9i$T*56LG^4jLF
zlJ>jx&OW=hWSzh6>LBAw*NT_!J9FXB);oJI%`UqC?&L(CexMnhS7dxP?L2M1&+Fve
zC3pND-#$C<=*_;`VE4&YdMB@`9R8nk@7o^#PYP$Bb=vH@(=QqLYfdv~N!Iz3qM0Vg
zf=-{CvdhqPcF-m>(`HS{SecN0YNd83+f_DuR?dCm+Qv5j$-CCLDONA8{Fr(5=e$_W
zwX!Qyle5g0TrrJ2<MJox&Uef1dbYahLHCwjt(t6QF!{>z`jtMPpC9w@U1oVl{{E-d
ko%yHCfhU5{80hHQGjLSB4s(@A>;hTk>FVdQ&MBb@0MQy3l>h($

literal 0
HcmV?d00001

diff --git a/docs/api.md b/docs/api.md
new file mode 100644
index 00000000..f363d279
--- /dev/null
+++ b/docs/api.md
@@ -0,0 +1,26 @@
+# API Reference
+
+
+## Storage
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.storage.Target
+    :members:
+```
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.storage.InputCache
+    :members:
+```
+
+## Recipes
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.recipe.BaseRecipe
+    :members:
+```
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe
+    :show-inheritance:
+```
diff --git a/docs/conf.py b/docs/conf.py
index 2dfe4699..a0e1904c 100644
--- a/docs/conf.py
+++ b/docs/conf.py
@@ -1,8 +1,9 @@
-import sphinx_pangeo_theme  # noqa
+# import sphinx_pangeo_theme  # noqa
+import sphinx_book_theme  # noqa
 
 # -- Project information -----------------------------------------------------
 
-project = "pangeo-forge"
+project = "Pangeo Forge"
 copyright = "2020, Pangeo Community"
 author = "Pangeo Community"
 
@@ -22,8 +23,10 @@
 
 # -- Options for HTML output -------------------------------------------------
 
-html_theme = "pangeo"
-html_static_path = ["_static"]
-html_sidebars = {"index": [], "**": ["localtoc.html"]}
+html_theme = "sphinx_book_theme"
+html_logo = "_static/pangeo-forge-logo-blue.png"
+# html_theme = "pangeo"
+# html_static_path = ["_static"]
+# html_sidebars = {"index": [], "**": ["localtoc.html"]}
 
 myst_heading_anchors = 2
diff --git a/docs/index.md b/docs/index.md
index 2718b2bd..f35e3bfb 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -38,5 +38,6 @@ recipes
 execution
 bakeries
 contribute
+api
 
 ```
diff --git a/docs/recipes.md b/docs/recipes.md
index f053c335..a45789fc 100644
--- a/docs/recipes.md
+++ b/docs/recipes.md
@@ -1,15 +1,57 @@
 # Recipes
 
+A recipe defines how to transform data in one format / location into another format / location.
+The primary way people contribute to Pangeo Forge is by writing / maintaining recipes.
+
+```{warning}
+The Recipe API is still in flux and may change. Make sure the version of the documentation
+you are reading matches your installed version of pangeo_forge.
+```
+
+## Storage
+
+Recipes need a place to store data.
+The location where the final dataset produced by the recipe is stored is called the
+``Target``. Pangeo forge has a special class for this: {class}`pangeo_forge.storage.Target`
+
+Creating a Target requires two arguments:
+- The ``fs`` argument is an [fsspec](https://filesystem-spec.readthedocs.io/en/latest/)
+  filesystem. Fsspec supports many different types of storage via its
+  [built in](https://filesystem-spec.readthedocs.io/en/latest/api.html#built-in-implementations)
+  and [third party](https://filesystem-spec.readthedocs.io/en/latest/api.html#other-known-implementations)
+  implementations.
+- The `path` argument specifies the specific path where the data should be stored.
+
+For example, creating a storage target for AWS S3 might look like this:
+```{code-block} python
+import s3fs
+fs = s3fs.S3FileSystem(key="MY_AWS_KEY", secret="MY_AWS_SECRET")
+target_path = "pangeo-forge-bucket/my-dataset-v1.zarr"
+target = Target(fs=fs, path=target_path)
+```
+
+Temporary data can be stored in an {class}`pangeo_forge.storage.InputCache` object.
+``InputCache`` is similar to ``Target``, but instead of specifying a ``path``,
+you specify ``prefix``.
+
+
 ## The Base Recipe Class
 
-```{eval-rst}
-.. autoclass:: pangeo_forge.recipe.BaseRecipe
-    :members:
+A recipe is initialized from a recipe class.
+```{code-block} python
+recipe = Recipe(option1='foo', option2=)
 ```
 
+All recipes follow the same basic steps.
+
+
+
+
+
 ## Specific Recipe Classes
 
 ```{eval-rst}
 .. autoclass:: pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe
     :show-inheritance:
+    :noindex:
 ```
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index e42fbb7d..9d07c79f 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -119,6 +119,7 @@ def to_pipelines(self) -> ParallelPipelines:
 class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     """There are many inputs (a.k.a. files, granules), arranged in a sequence
     along the dimension `sequence_dim`. Each file may contain multiple variables.
+    This class uses Xarray to read and write data.
 
     :param input_urls: The inputs used to generate the dataset.
     :param sequence_dim: The dimension name along which the inputs will be concatenated.
@@ -149,8 +150,6 @@ def __post_init__(self):
 
     @property
     def prepare(self) -> Callable:
-        """Prepare target for storing dataset."""
-
         def _prepare():
 
             try:
@@ -175,13 +174,6 @@ def _prepare():
 
     @property
     def cache_input(self) -> Callable:
-        """Cache the input.
-
-        Properties
-        ----------
-        url : URL pointing to the input file. Must be openable by fsspec.
-        """
-
         def cache_func(fname: str) -> None:
             logger.info(f"Caching input '{fname}'")
             with input_opener(fname, mode="rb") as source:
@@ -192,14 +184,6 @@ def cache_func(fname: str) -> None:
 
     @property
     def store_chunk(self) -> Callable:
-        """Store a chunk in the target.
-
-        Parameters
-        ----------
-        chunk_key : str
-            The identifier for the chunk
-        """
-
         def _store_chunk(chunk_key):
             ds_chunk = self.open_chunk(chunk_key)
 
@@ -218,8 +202,6 @@ def drop_vars(ds):
 
     @property
     def finalize(self) -> Callable:
-        """Finalize writing of dataset."""
-
         def _finalize():
             if self.consolidate_zarr:
                 logger.info("Consolidating Zarr metadata")
diff --git a/pangeo_forge/storage.py b/pangeo_forge/storage.py
index 742bb9f2..acc5ad33 100644
--- a/pangeo_forge/storage.py
+++ b/pangeo_forge/storage.py
@@ -1,6 +1,7 @@
 import os
 from contextlib import contextmanager
 from dataclasses import dataclass
+from typing import BinaryIO, NoReturn
 
 import fsspec
 
@@ -8,20 +9,16 @@
 @dataclass
 class Target:
     """Representation of a storage target for Pangeo Forge.
-    Attributes
-    ----------
-    fs : FileSystemSpec.AbtractFileSystem
-        The filesystem we are writing to. Should be instantiated outside this
-        class.
-    path : str
-        The path where the target data will be saved.
+
+    :param fs: The filesystem object we are writing to.
+    :param path: The path where the target data will be saved.
     """
 
     fs: fsspec.AbstractFileSystem
     path: str
 
-    def get_mapper(self):
-        # don't want to use this because we want to use a fancier Zarr FSStore
+    def get_mapper(self) -> fsspec.mapping.FSMap:
+        """Get a mutable mapping object suitable for storing Zarr data."""
         return self.fs.get_mapper(self.path)
 
 
@@ -34,13 +31,8 @@ class InputCache:
     """Representation of an intermediate storage location where remote files
     Can be cached locally.
 
-    Attributes
-    ----------
-    fs : FileSystemSpec.AbtractFileSystem
-        The filesystem we are writing to. Should be instantiated outside this
-        class.
-    prefix : str
-        A path prepended to all paths.
+    :param fs: The filesystem we are writing to.
+    :param prefix: A path prepended to all paths.
     """
 
     fs: fsspec.AbstractFileSystem
@@ -49,14 +41,17 @@ class InputCache:
     def _full_path(self, path):
         return os.path.join(self.prefix, _hash_path(path))
 
-    def exists(self, path):
+    def exists(self, path) -> bool:
+        """Check that the file is in the cache."""
         return self.fs.exists(self._full_path(path))
 
-    def rm(self, path):
+    def rm(self, path) -> NoReturn:
+        """Remove file from the cache."""
         self.fs.rm(self._full_path(path))
 
     @contextmanager
-    def open(self, path, **kwargs):
+    def open(self, path, **kwargs) -> BinaryIO:
+        """Open file with a context manager."""
         with self.fs.open(self._full_path(path), **kwargs) as f:
             yield f
 
diff --git a/setup.cfg b/setup.cfg
index d2063367..6a8506d2 100644
--- a/setup.cfg
+++ b/setup.cfg
@@ -3,7 +3,7 @@ max-line-length = 100
 
 [isort]
 known_first_party=pangeo_forge
-known_third_party=click,fsspec,numpy,pandas,pkg_resources,prefect,pytest,rechunker,setuptools,sphinx_pangeo_theme,xarray,zarr
+known_third_party=click,fsspec,numpy,pandas,pkg_resources,prefect,pytest,rechunker,setuptools,sphinx_book_theme,xarray,zarr
 multi_line_output=3
 include_trailing_comma=True
 force_grid_wrap=0

From fa20daf64b5ccbccd32371d0dd6755a58a1f7b55 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 00:42:18 -0500
Subject: [PATCH 26/34] add tutorial to docs

---
 docs/conf.py               |   4 +-
 docs/index.md              |   1 +
 docs/recipe_tutorial.ipynb | 215 +++++++++++++++++++++++++++++++++++++
 pangeo_forge/recipe.py     |  11 +-
 4 files changed, 226 insertions(+), 5 deletions(-)
 create mode 100644 docs/recipe_tutorial.ipynb

diff --git a/docs/conf.py b/docs/conf.py
index a0e1904c..35c8d26d 100644
--- a/docs/conf.py
+++ b/docs/conf.py
@@ -11,14 +11,14 @@
 # -- General configuration ---------------------------------------------------
 
 extensions = [
-    "myst_parser",
+    "myst_nb",
     "sphinx.ext.autodoc",
     # "numpydoc",
     "sphinx_autodoc_typehints",
 ]
 
 templates_path = ["_templates"]
-exclude_patterns = []
+exclude_patterns = ["_build", "**.ipynb_checkpoints"]
 master_doc = "index"
 
 # -- Options for HTML output -------------------------------------------------
diff --git a/docs/index.md b/docs/index.md
index f35e3bfb..319199e4 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -35,6 +35,7 @@ For more information, see {doc}`bakeries`.
 :caption: Contents
 
 recipes
+recipe_tutorial
 execution
 bakeries
 contribute
diff --git a/docs/recipe_tutorial.ipynb b/docs/recipe_tutorial.ipynb
new file mode 100644
index 00000000..16cc5820
--- /dev/null
+++ b/docs/recipe_tutorial.ipynb
@@ -0,0 +1,215 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Recipe Tutorial\n",
+    "\n",
+    "This tutorial describes how to create a recipe from scratch.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 1: Get to know your source data\n",
+    "\n",
+    "If you are developing a new recipe, you are probably starting from an existing\n",
+    "datset. The first step is to just get to know the dataset. For this tutorial,\n",
+    "our example will be the _NOAA Optimum Interpolation Sea Surface Temperature\n",
+    "(OISST) v2.1_. The authoritative website describing the data is\n",
+    "<https://www.ncdc.noaa.gov/oisst/optimum-interpolation-sea-surface-temperature-oisst-v21>.\n",
+    "This website contains links to the actual data files on the\n",
+    "[data access](https://www.ncdc.noaa.gov/oisst/data-access) page. We will use the\n",
+    "_AVHRR-Only_ version of the data and follow the corresponding link to the\n",
+    "[Gridded netCDF Data](https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/).\n",
+    "Browsing through the directories, we can see that there is one file per day. The\n",
+    "very first day of the dataset is stored at the following URL:\n",
+    "\n",
+    "```text\n",
+    "https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810901.nc\n",
+    "```\n",
+    "\n",
+    "From this example, we can work out the pattern of the file naming conventions.\n",
+    "But first, let's just download one of the files and open it up.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "! curl -O https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810901.nc "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import xarray as xr\n",
+    "\n",
+    "ds = xr.open_dataset(\"oisst-avhrr-v02r01.19810901.nc\")\n",
+    "ds"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can see there are four data variables, all with dimension\n",
+    "`(time, zlev, lat, lon)`. There is a _dimension coordinate_ for each dimension,\n",
+    "and no _non-dimension coordinates_. Each file in the sequence presumably has the\n",
+    "same `zlev`, `lat`, and `lon`, but we expect `time` to be different in each one.\n",
+    "\n",
+    "Let's also check the total size of the dataset in the file.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print(f\"File size is {ds.nbytes/1e6} MB\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The file size is important because it will help us define the _chunk size_\n",
+    "Pangeo Forge will use to build up the target dataset.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 2: Pick a Recipe class\n",
+    "\n",
+    "For our first recipe, we will want to use a pre-defined Recipe class from Pangeo\n",
+    "Forge.\n",
+    "\n",
+    "By examining the {doc}`recipes` documentation page, we see that our scenario is\n",
+    "a good case for the {class}`pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe`\n",
+    "class. Let's examine its documentation string in our notebook.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pangeo_forge.recipe import NetCDFtoZarrSequentialRecipe\n",
+    "NetCDFtoZarrSequentialRecipe?"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 3: Define Recipe parameters\n",
+    "\n",
+    "Our chosen class has only two required parameters: `input_urls` and\n",
+    "`sequence_dim`.\n",
+    "\n",
+    "`input_urls` is a list of URLs pointing to the data. To populate this, we need\n",
+    "to explicitly create this list based on what we know about the file naming\n",
+    "conventions. Let's look again at the first URL\n",
+    "\n",
+    "```text\n",
+    "https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810901.nc\n",
+    "```\n",
+    "\n",
+    "From this we deduce the following format string.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "input_url_pattern = (\n",
+    "    \"https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation\"\n",
+    "    \"/v2.1/access/avhrr/{yyyymm}/oisst-avhrr-v02r01.{yyyymmdd}.nc\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we need a sequence of datetimes. Pandas is the easiest way to get this. At\n",
+    "the time of writing, the latest available data is from 2021-01-05.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "dates = pd.date_range(\"1981-09-01\", \"2021-01-05\", freq=\"D\")\n",
+    "input_urls = [\n",
+    "    input_url_pattern.format(\n",
+    "        yyyymm=day.strftime(\"%Y%m\"), yyyymmdd=day.strftime(\"%Y%m%d\")\n",
+    "    )\n",
+    "    for day in dates\n",
+    "]\n",
+    "print(f\"Found {len(input_urls)} files!\")\n",
+    "input_urls[-1]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "That's a lot of files!\n",
+    "\n",
+    "The other remaining parameter is `sequence_dim`. It's just `\"time\"`. We can now\n",
+    "instantiate our recipe.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "recipe = NetCDFtoZarrSequentialRecipe(\n",
+    "    input_urls=input_urls, sequence_dim=\"time\"\n",
+    ")\n",
+    "recipe"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index 9d07c79f..b962ce1e 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -117,8 +117,13 @@ def to_pipelines(self) -> ParallelPipelines:
 
 @dataclass
 class NetCDFtoZarrSequentialRecipe(BaseRecipe):
-    """There are many inputs (a.k.a. files, granules), arranged in a sequence
-    along the dimension `sequence_dim`. Each file may contain multiple variables.
+    """This class represents a dataset composed of many individual NetCDF files.
+    The files are arraged in a sequence along a single dimension, called the
+    `sequence_dim`. Each file may contain multiple variables.
+
+    The dataset is assembled by concatenating all of these files along `sequence_dim`.
+    The target is written in Zarr format.
+
     This class uses Xarray to read and write data.
 
     :param input_urls: The inputs used to generate the dataset.
@@ -133,7 +138,7 @@ class NetCDFtoZarrSequentialRecipe(BaseRecipe):
        the inputs to form a chunk.
     """
 
-    input_urls: Iterable[str]
+    input_urls: Iterable[str] = field(repr=False)
     sequence_dim: str
     inputs_per_chunk: int = 1
     nitems_per_input: int = 1

From 8d66eb948e7456333c94d8e70bb76231bfae814a Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 11:47:06 -0500
Subject: [PATCH 27/34] refactored storage targets

---
 pangeo_forge/recipe.py  | 31 +++++++++--------
 pangeo_forge/storage.py | 75 ++++++++++++++++++++++++++++++-----------
 tests/conftest.py       |  6 ++--
 tests/test_fixtures.py  |  2 +-
 4 files changed, 76 insertions(+), 38 deletions(-)

diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index b962ce1e..a2ed5aa3 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -13,7 +13,7 @@
 import zarr
 from rechunker.types import MultiStagePipeline, ParallelPipelines, Stage
 
-from .storage import InputCache, Target
+from .storage import AbstractTarget
 from .utils import chunked_iterable, fix_scalar_attr_encoding
 
 logger = logging.getLogger(__name__)
@@ -132,6 +132,9 @@ class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     :param nitems_per_input: The length of each input along the `sequence_dim` dimension.
     :param target: A location in which to put the dataset. Can also be assigned at run time.
     :param input_cache: A location in which to cache temporary data.
+    :param require_cache: Whether to allow opening inputs directly which have not
+      yet been cached. This could lead to very slow behavior if the inputs
+      live on a slow network.
     :param consolidate_zarr: Whether to consolidate the resulting Zarr dataset.
     :param xarray_open_kwargs: Extra options for opening the inputs with Xarray.
     :param xarray_concat_kwargs: Extra options to pass to Xarray when concatenating
@@ -142,8 +145,9 @@ class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     sequence_dim: str
     inputs_per_chunk: int = 1
     nitems_per_input: int = 1
-    target: Optional[Target] = None
-    input_cache: Optional[InputCache] = None
+    target: Optional[AbstractTarget] = None
+    input_cache: Optional[AbstractTarget] = None
+    require_cache: bool = True
     consolidate_zarr: bool = True
     xarray_open_kwargs: dict = field(default_factory=dict)
     xarray_concat_kwargs: dict = field(default_factory=dict)
@@ -217,19 +221,18 @@ def _finalize():
 
     @contextmanager
     def input_opener(self, fname: str):
-        if self.input_cache is None:
-            logger.info(f"No cache. Opening input `{fname}` directly.")
-            # This will bypass the cache. May be slow.
-            with input_opener(fname, mode="rb") as f:
-                yield f
-        elif self.input_cache.exists(fname):
-            logger.info(f"Input '{fname}' found in cache")
+        try:
             with self.input_cache.open(fname, mode="rb") as f:
+                logger.info(f"Opening '{fname}' from cache")
                 yield f
-        else:
-            raise ValueError(
-                f"Input '{fname}' has not been cached yet. " "Call .cache_input() first."
-            )
+        except IOError:  # TODO figure out the excpetion to catch
+            if self.require_cache:
+                raise
+            else:
+                logger.info(f"No cache found. Opening input `{fname}` directly.")
+                # This will bypass the cache. May be slow.
+                with input_opener(fname, mode="rb") as f:
+                    yield f
 
     def open_input(self, fname: str):
         with self.input_opener(fname) as f:
diff --git a/pangeo_forge/storage.py b/pangeo_forge/storage.py
index acc5ad33..87fce491 100644
--- a/pangeo_forge/storage.py
+++ b/pangeo_forge/storage.py
@@ -1,4 +1,7 @@
 import os
+import re
+import unicodedata
+from abc import ABC, abstractmethod
 from contextlib import contextmanager
 from dataclasses import dataclass
 from typing import BinaryIO, NoReturn
@@ -6,20 +9,25 @@
 import fsspec
 
 
-@dataclass
-class Target:
-    """Representation of a storage target for Pangeo Forge.
+class AbstractTarget(ABC):
+    @abstractmethod
+    def get_mapper(self):
+        pass
 
-    :param fs: The filesystem object we are writing to.
-    :param path: The path where the target data will be saved.
-    """
+    @abstractmethod
+    def exists(self, path) -> bool:
+        """Check that the file exists."""
+        pass
 
-    fs: fsspec.AbstractFileSystem
-    path: str
+    @abstractmethod
+    def rm(self, path) -> NoReturn:
+        """Remove file."""
+        pass
 
-    def get_mapper(self) -> fsspec.mapping.FSMap:
-        """Get a mutable mapping object suitable for storing Zarr data."""
-        return self.fs.get_mapper(self.path)
+    @contextmanager
+    def open(self, path, **kwargs) -> BinaryIO:
+        """Open file with a context manager."""
+        pass
 
 
 def _hash_path(path: str) -> str:
@@ -27,19 +35,22 @@ def _hash_path(path: str) -> str:
 
 
 @dataclass
-class InputCache:
-    """Representation of an intermediate storage location where remote files
-    Can be cached locally.
+class FSSpecTarget(AbstractTarget):
+    """Representation of a storage target for Pangeo Forge.
 
-    :param fs: The filesystem we are writing to.
-    :param prefix: A path prepended to all paths.
+    :param fs: The filesystem object we are writing to.
+    :param root_path: The path under which the target data will be stored.
     """
 
     fs: fsspec.AbstractFileSystem
-    prefix: str = ""
+    root_path: str = ""
+
+    def get_mapper(self) -> fsspec.mapping.FSMap:
+        """Get a mutable mapping object suitable for storing Zarr data."""
+        return self.fs.get_mapper(self.root_path)
 
     def _full_path(self, path):
-        return os.path.join(self.prefix, _hash_path(path))
+        return os.path.join(self.root_path, path)
 
     def exists(self, path) -> bool:
         """Check that the file is in the cache."""
@@ -56,5 +67,29 @@ def open(self, path, **kwargs) -> BinaryIO:
             yield f
 
     def __post_init__(self):
-        if not self.fs.isdir(self.prefix):
-            self.fs.mkdir(self.prefix)
+        if not self.fs.isdir(self.root_path):
+            self.fs.mkdir(self.root_path)
+
+
+class FlatFSSpecTarget(FSSpecTarget):
+    """A target that sanitizes all the path names so that everthing is stored
+    in a single directory.
+
+    Designed to be used as a cache for inputs.
+    """
+
+    def _full_path(self, path):
+        slug = _slugify(path)
+        prefix = prefix = hex(hash(path))[2:10]
+        new_path = "-".join([prefix, slug])
+        return os.path.join(self.root_path, new_path)
+
+
+def _slugify(value):
+    # Adopted from
+    # https://github.com/django/django/blob/master/django/utils/text.py
+    # https://stackoverflow.com/questions/295135/turn-a-string-into-a-valid-filename
+    value = str(value)
+    value = unicodedata.normalize("NFKD", value).encode("ascii", "ignore").decode("ascii")
+    value = re.sub(r"[^\w\s-]+", "_", value.lower())
+    return re.sub(r"[-\s]+", "-", value).strip("-_")
diff --git a/tests/conftest.py b/tests/conftest.py
index 6386481a..1525eb57 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -8,7 +8,7 @@
 import xarray as xr
 
 from pangeo_forge import recipe
-from pangeo_forge.storage import InputCache, Target
+from pangeo_forge.storage import FlatFSSpecTarget, FSSpecTarget
 
 # where to run the http server
 _PORT = "8080"
@@ -80,14 +80,14 @@ def tmp_target(tmpdir_factory):
 
     fs = fsspec.get_filesystem_class("file")()
     path = str(tmpdir_factory.mktemp("target"))
-    return Target(fs, path)
+    return FSSpecTarget(fs, path)
 
 
 @pytest.fixture()
 def tmp_cache(tmpdir_factory):
     path = str(tmpdir_factory.mktemp("cache"))
     fs = fsspec.get_filesystem_class("file")()
-    cache = InputCache(fs, prefix=path)
+    cache = FlatFSSpecTarget(fs, path)
     return cache
 
 
diff --git a/tests/test_fixtures.py b/tests/test_fixtures.py
index c867f17b..4e5e34fa 100644
--- a/tests/test_fixtures.py
+++ b/tests/test_fixtures.py
@@ -24,7 +24,7 @@ def test_fixture_http_files(daily_xarray_dataset, netcdf_http_server):
 def test_target(tmp_target):
     mapper = tmp_target.get_mapper()
     mapper["foo"] = b"bar"
-    with open(tmp_target.path + "/foo") as f:
+    with open(tmp_target.root_path + "/foo") as f:
         res = f.read()
     assert res == "bar"
 

From 86f6f929d79a659acbee4993b9afe9641660108e Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 12:10:08 -0500
Subject: [PATCH 28/34] better target testing

---
 .pre-commit-config.yaml |  6 +-----
 pangeo_forge/recipe.py  |  4 ++--
 pangeo_forge/storage.py | 26 ++++++++++++++++++++++++++
 tests/conftest.py       |  7 ++++++-
 tests/test_fixtures.py  | 20 ++++++++++++++++++++
 5 files changed, 55 insertions(+), 8 deletions(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 067b22f0..65a3907e 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -26,16 +26,12 @@ repos:
     rev: v2.2.0
     hooks:
     - id: seed-isort-config
+
 -   repo: https://github.com/pre-commit/mirrors-isort
     rev: v5.2.0
     hooks:
     - id: isort
 
--   repo: https://github.com/deathbeds/prenotebook
-    rev: f5bdb72a400f1a56fe88109936c83aa12cc349fa
-    hooks:
-    - id: prenotebook
-
 -   repo: https://github.com/myint/rstcheck
     rev: master
     hooks:
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index a2ed5aa3..c22c12fa 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -165,7 +165,7 @@ def _prepare():
                 ds = self.open_target()
                 logger.info("Found an existing dataset in target")
                 logger.debug(f"{ds}")
-            except (IOError, zarr.errors.GroupNotFoundError):
+            except (FileNotFoundError, IOError, zarr.errors.GroupNotFoundError):
                 first_chunk_key = next(self.iter_chunks())
                 for input_url in self.inputs_for_chunk(first_chunk_key):
                     self.cache_input(input_url)
@@ -225,7 +225,7 @@ def input_opener(self, fname: str):
             with self.input_cache.open(fname, mode="rb") as f:
                 logger.info(f"Opening '{fname}' from cache")
                 yield f
-        except IOError:  # TODO figure out the excpetion to catch
+        except (IOError, FileNotFoundError):
             if self.require_cache:
                 raise
             else:
diff --git a/pangeo_forge/storage.py b/pangeo_forge/storage.py
index 87fce491..44fc8e15 100644
--- a/pangeo_forge/storage.py
+++ b/pangeo_forge/storage.py
@@ -93,3 +93,29 @@ def _slugify(value):
     value = unicodedata.normalize("NFKD", value).encode("ascii", "ignore").decode("ascii")
     value = re.sub(r"[^\w\s-]+", "_", value.lower())
     return re.sub(r"[-\s]+", "-", value).strip("-_")
+
+
+class UninitializedTarget(AbstractTarget):
+    def get_mapper(self):
+        raise UninitializedTargetError
+
+    def exists(self, path) -> bool:
+        raise UninitializedTargetError
+
+    def rm(self, path) -> NoReturn:
+        raise UninitializedTargetError
+
+    def open(self, path, **kwargs) -> BinaryIO:
+        raise UninitializedTargetError
+
+
+class TargetError(Exception):
+    """Base class for exceptions in this module."""
+
+    pass
+
+
+class UninitializedTargetError(TargetError):
+    """Operation on an uninitialized Target."""
+
+    pass
diff --git a/tests/conftest.py b/tests/conftest.py
index 1525eb57..37b4eaa4 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -8,7 +8,7 @@
 import xarray as xr
 
 from pangeo_forge import recipe
-from pangeo_forge.storage import FlatFSSpecTarget, FSSpecTarget
+from pangeo_forge.storage import FlatFSSpecTarget, FSSpecTarget, UninitializedTarget
 
 # where to run the http server
 _PORT = "8080"
@@ -91,6 +91,11 @@ def tmp_cache(tmpdir_factory):
     return cache
 
 
+@pytest.fixture()
+def uninitialized_target():
+    return UninitializedTarget()
+
+
 @pytest.fixture
 def netCDFtoZarr_sequential_recipe(daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache):
     r = recipe.NetCDFtoZarrSequentialRecipe(
diff --git a/tests/test_fixtures.py b/tests/test_fixtures.py
index 4e5e34fa..cfc64fd1 100644
--- a/tests/test_fixtures.py
+++ b/tests/test_fixtures.py
@@ -1,6 +1,8 @@
 import fsspec
+import pytest
 import xarray as xr
 
+from pangeo_forge.storage import UninitializedTargetError
 from pangeo_forge.utils import fix_scalar_attr_encoding
 
 
@@ -27,6 +29,24 @@ def test_target(tmp_target):
     with open(tmp_target.root_path + "/foo") as f:
         res = f.read()
     assert res == "bar"
+    with pytest.raises(FileNotFoundError):
+        tmp_target.rm("baz")
+    with pytest.raises(FileNotFoundError):
+        with tmp_target.open("baz"):
+            pass
+
+
+def test_uninitialized_target(uninitialized_target):
+    target = uninitialized_target
+    with pytest.raises(UninitializedTargetError):
+        target.get_mapper()
+    with pytest.raises(UninitializedTargetError):
+        target.exists("foo")
+    with pytest.raises(UninitializedTargetError):
+        target.rm("foo")
+    with pytest.raises(UninitializedTargetError):
+        with target.open("foo"):
+            pass
 
 
 def test_cache(tmp_cache):

From 049e692cde62ddfb4cde1dcf5261f615806f76f3 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 14:36:15 -0500
Subject: [PATCH 29/34] change cannonical recipe execution order

---
 pangeo_forge/recipe.py  | 43 ++++++++++++++++++++++++-----------------
 pangeo_forge/storage.py |  6 ++++++
 tests/conftest.py       |  4 ++--
 tests/test_recipe.py    | 21 ++++++++++++++++++--
 4 files changed, 52 insertions(+), 22 deletions(-)

diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index c22c12fa..ebe82dda 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -13,7 +13,7 @@
 import zarr
 from rechunker.types import MultiStagePipeline, ParallelPipelines, Stage
 
-from .storage import AbstractTarget
+from .storage import AbstractTarget, UninitializedTarget
 from .utils import chunked_iterable, fix_scalar_attr_encoding
 
 logger = logging.getLogger(__name__)
@@ -23,12 +23,12 @@
 #   t = PangeoForgeTarget()
 #   r = MyRecipe(target=t, **opts) # 1
 #   # manual execution of recipe
-#   r.prepare() # 3
 #   for input_key in r.iter_inputs():
 #       r.cache_input(input_key) # 4
+#   r.prepare_target() # 3
 #   for chunk_key in r.iter_chunks():
 #       r.store_chunk(chunk_key) # 5
-#   r.finalize() # 6
+#   r.finalize_target() # 6
 
 
 # 1) Initialize the Recipe object
@@ -56,7 +56,7 @@ class BaseRecipe(ABC):
 
     @property
     @abstractmethod
-    def prepare(self) -> Callable[[], NoReturn]:
+    def prepare_target(self) -> Callable[[], NoReturn]:
         """Prepare the recipe for execution by initializing the target.
         Attribute that returns a callable function.
         """
@@ -90,7 +90,7 @@ def store_chunk(self) -> Callable[[Hashable], NoReturn]:
 
     @property
     @abstractmethod
-    def finalize(self) -> Callable[[], NoReturn]:
+    def finalize_target(self) -> Callable[[], NoReturn]:
         """Final step to finish the recipe after data has been written.
         Attribute that returns a callable function.
         """
@@ -101,10 +101,10 @@ def to_pipelines(self) -> ParallelPipelines:
         """
 
         pipeline = []  # type: MultiStagePipeline
-        pipeline.append(Stage(self.prepare))
         pipeline.append(Stage(self.cache_input, list(self.iter_inputs())))
+        pipeline.append(Stage(self.prepare_target))
         pipeline.append(Stage(self.store_chunk, list(self.iter_chunks())))
-        pipeline.append(Stage(self.finalize))
+        pipeline.append(Stage(self.finalize_target))
         pipelines = []  # type: ParallelPipelines
         pipelines.append(pipeline)
         return pipelines
@@ -145,8 +145,8 @@ class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     sequence_dim: str
     inputs_per_chunk: int = 1
     nitems_per_input: int = 1
-    target: Optional[AbstractTarget] = None
-    input_cache: Optional[AbstractTarget] = None
+    target: Optional[AbstractTarget] = field(default_factory=UninitializedTarget)
+    input_cache: Optional[AbstractTarget] = field(default_factory=UninitializedTarget)
     require_cache: bool = True
     consolidate_zarr: bool = True
     xarray_open_kwargs: dict = field(default_factory=dict)
@@ -158,8 +158,8 @@ def __post_init__(self):
         }
 
     @property
-    def prepare(self) -> Callable:
-        def _prepare():
+    def prepare_target(self) -> Callable:
+        def _prepare_target():
 
             try:
                 ds = self.open_target()
@@ -167,8 +167,8 @@ def _prepare():
                 logger.debug(f"{ds}")
             except (FileNotFoundError, IOError, zarr.errors.GroupNotFoundError):
                 first_chunk_key = next(self.iter_chunks())
-                for input_url in self.inputs_for_chunk(first_chunk_key):
-                    self.cache_input(input_url)
+                # for input_url in self.inputs_for_chunk(first_chunk_key):
+                #    self.cache_input(input_url)
                 ds = self.open_chunk(first_chunk_key).chunk()
 
                 # make sure the concat dim has a valid fill_value to avoid
@@ -179,7 +179,7 @@ def _prepare():
 
             self.expand_target_dim(self.sequence_dim, self.sequence_len())
 
-        return _prepare
+        return _prepare_target
 
     @property
     def cache_input(self) -> Callable:
@@ -210,14 +210,14 @@ def drop_vars(ds):
         return _store_chunk
 
     @property
-    def finalize(self) -> Callable:
-        def _finalize():
+    def finalize_target(self) -> Callable:
+        def _finalize_target():
             if self.consolidate_zarr:
                 logger.info("Consolidating Zarr metadata")
                 target_mapper = self.target.get_mapper()
                 zarr.consolidate_metadata(target_mapper)
 
-        return _finalize
+        return _finalize_target
 
     @contextmanager
     def input_opener(self, fname: str):
@@ -227,7 +227,11 @@ def input_opener(self, fname: str):
                 yield f
         except (IOError, FileNotFoundError):
             if self.require_cache:
-                raise
+                raise FileNotFoundError(
+                    f"You are trying to open input {fname}, but the file is "
+                    "not cached yet. First call `cache_input` or set "
+                    "`require_cache=False`."
+                )
             else:
                 logger.info(f"No cache found. Opening input `{fname}` directly.")
                 # This will bypass the cache. May be slow.
@@ -249,7 +253,10 @@ def open_chunk(self, chunk_key):
         inputs = self.inputs_for_chunk(chunk_key)
         dsets = [self.open_input(i) for i in inputs]
         # CONCAT DELETES ENCODING!!!
+        # OR NO IT DOESN'T! Not in the latest version of xarray?
         ds = xr.concat(dsets, self.sequence_dim, **self.xarray_concat_kwargs)
+        for var in ds.variables:
+            ds[var].encoding = {}
         logger.debug(f"{ds}")
 
         # TODO: maybe do some chunking here?
diff --git a/pangeo_forge/storage.py b/pangeo_forge/storage.py
index 44fc8e15..43b11a2d 100644
--- a/pangeo_forge/storage.py
+++ b/pangeo_forge/storage.py
@@ -85,6 +85,12 @@ def _full_path(self, path):
         return os.path.join(self.root_path, new_path)
 
 
+class CacheFSSpecTarget(FlatFSSpecTarget):
+    """Alias for FlatFSSpecTarget"""
+
+    pass
+
+
 def _slugify(value):
     # Adopted from
     # https://github.com/django/django/blob/master/django/utils/text.py
diff --git a/tests/conftest.py b/tests/conftest.py
index 37b4eaa4..af505ade 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -8,7 +8,7 @@
 import xarray as xr
 
 from pangeo_forge import recipe
-from pangeo_forge.storage import FlatFSSpecTarget, FSSpecTarget, UninitializedTarget
+from pangeo_forge.storage import CacheFSSpecTarget, FSSpecTarget, UninitializedTarget
 
 # where to run the http server
 _PORT = "8080"
@@ -87,7 +87,7 @@ def tmp_target(tmpdir_factory):
 def tmp_cache(tmpdir_factory):
     path = str(tmpdir_factory.mktemp("cache"))
     fs = fsspec.get_filesystem_class("file")()
-    cache = FlatFSSpecTarget(fs, path)
+    cache = CacheFSSpecTarget(fs, path)
     return cache
 
 
diff --git a/tests/test_recipe.py b/tests/test_recipe.py
index 2a6a412d..05fc2ae5 100644
--- a/tests/test_recipe.py
+++ b/tests/test_recipe.py
@@ -2,6 +2,7 @@
 import xarray as xr
 
 from pangeo_forge import recipe
+from pangeo_forge.storage import UninitializedTargetError
 
 dummy_fnames = ["a.nc", "b.nc", "c.nc"]
 
@@ -50,13 +51,29 @@ def test_NetCDFtoZarrSequentialRecipe(
     )
 
     # this is the cannonical way to manually execute a recipe
-    r.prepare()
     for input_key in r.iter_inputs():
         r.cache_input(input_key)
+    r.prepare_target()
     for chunk_key in r.iter_chunks():
         r.store_chunk(chunk_key)
-    r.finalize()
+    r.finalize_target()
 
     ds_target = xr.open_zarr(tmp_target.get_mapper(), consolidated=True).load()
     ds_expected = daily_xarray_dataset.compute()
     assert ds_target.identical(ds_expected)
+
+
+def test_NetCDFtoZarrSequentialRecipeNoTarget(
+    daily_xarray_dataset, netcdf_local_paths, tmp_target, tmp_cache
+):
+
+    r = recipe.NetCDFtoZarrSequentialRecipe(
+        input_urls=netcdf_local_paths,
+        sequence_dim="time",
+        inputs_per_chunk=1,
+        nitems_per_input=daily_xarray_dataset.attrs["items_per_file"],
+    )
+
+    # this is the cannonical way to manually execute a recipe
+    with pytest.raises(UninitializedTargetError):
+        r.cache_input(next(r.iter_inputs()))

From 49519ef2a5fcb7b1dfe08b7ed61ed66a56917ff3 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 15:38:12 -0500
Subject: [PATCH 30/34] big update

---
 docs/api.md                |    4 +-
 docs/conf.py               |   12 +-
 docs/recipe_tutorial.ipynb | 3458 +++++++++++++++++++++++++++++++++++-
 docs/recipes.md            |   14 +-
 pangeo_forge/recipe.py     |   12 +-
 5 files changed, 3467 insertions(+), 33 deletions(-)

diff --git a/docs/api.md b/docs/api.md
index f363d279..667283b2 100644
--- a/docs/api.md
+++ b/docs/api.md
@@ -4,12 +4,12 @@
 ## Storage
 
 ```{eval-rst}
-.. autoclass:: pangeo_forge.storage.Target
+.. autoclass:: pangeo_forge.storage.FSSpecTarget
     :members:
 ```
 
 ```{eval-rst}
-.. autoclass:: pangeo_forge.storage.InputCache
+.. autoclass:: pangeo_forge.storage.CacheFSSpecTarget
     :members:
 ```
 
diff --git a/docs/conf.py b/docs/conf.py
index 35c8d26d..8ff69e2c 100644
--- a/docs/conf.py
+++ b/docs/conf.py
@@ -15,18 +15,22 @@
     "sphinx.ext.autodoc",
     # "numpydoc",
     "sphinx_autodoc_typehints",
+    "sphinx_copybutton",
 ]
 
 templates_path = ["_templates"]
 exclude_patterns = ["_build", "**.ipynb_checkpoints"]
 master_doc = "index"
 
+# we always have to manually run the notebooks because they are slow / expensive
+jupyter_execute_notebooks = "off"
+
 # -- Options for HTML output -------------------------------------------------
 
 html_theme = "sphinx_book_theme"
 html_logo = "_static/pangeo-forge-logo-blue.png"
-# html_theme = "pangeo"
-# html_static_path = ["_static"]
-# html_sidebars = {"index": [], "**": ["localtoc.html"]}
-
+html_static_path = ["_static"]
 myst_heading_anchors = 2
+html_css_files = [
+    "custom.css",
+]
diff --git a/docs/recipe_tutorial.ipynb b/docs/recipe_tutorial.ipynb
index 16cc5820..3e7b2cfe 100644
--- a/docs/recipe_tutorial.ipynb
+++ b/docs/recipe_tutorial.ipynb
@@ -46,9 +46,474 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div><svg style=\"position: absolute; width: 0; height: 0; overflow: hidden\">\n",
+       "<defs>\n",
+       "<symbol id=\"icon-database\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M16 0c-8.837 0-16 2.239-16 5v4c0 2.761 7.163 5 16 5s16-2.239 16-5v-4c0-2.761-7.163-5-16-5z\"></path>\n",
+       "<path d=\"M16 17c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "<path d=\"M16 26c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "</symbol>\n",
+       "<symbol id=\"icon-file-text2\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M28.681 7.159c-0.694-0.947-1.662-2.053-2.724-3.116s-2.169-2.030-3.116-2.724c-1.612-1.182-2.393-1.319-2.841-1.319h-15.5c-1.378 0-2.5 1.121-2.5 2.5v27c0 1.378 1.122 2.5 2.5 2.5h23c1.378 0 2.5-1.122 2.5-2.5v-19.5c0-0.448-0.137-1.23-1.319-2.841zM24.543 5.457c0.959 0.959 1.712 1.825 2.268 2.543h-4.811v-4.811c0.718 0.556 1.584 1.309 2.543 2.268zM28 29.5c0 0.271-0.229 0.5-0.5 0.5h-23c-0.271 0-0.5-0.229-0.5-0.5v-27c0-0.271 0.229-0.5 0.5-0.5 0 0 15.499-0 15.5 0v7c0 0.552 0.448 1 1 1h7v19.5z\"></path>\n",
+       "<path d=\"M23 26h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 22h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 18h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "</symbol>\n",
+       "</defs>\n",
+       "</svg>\n",
+       "<style>/* CSS stylesheet for displaying xarray objects in jupyterlab.\n",
+       " *\n",
+       " */\n",
+       "\n",
+       ":root {\n",
+       "  --xr-font-color0: var(--jp-content-font-color0, rgba(0, 0, 0, 1));\n",
+       "  --xr-font-color2: var(--jp-content-font-color2, rgba(0, 0, 0, 0.54));\n",
+       "  --xr-font-color3: var(--jp-content-font-color3, rgba(0, 0, 0, 0.38));\n",
+       "  --xr-border-color: var(--jp-border-color2, #e0e0e0);\n",
+       "  --xr-disabled-color: var(--jp-layout-color3, #bdbdbd);\n",
+       "  --xr-background-color: var(--jp-layout-color0, white);\n",
+       "  --xr-background-color-row-even: var(--jp-layout-color1, white);\n",
+       "  --xr-background-color-row-odd: var(--jp-layout-color2, #eeeeee);\n",
+       "}\n",
+       "\n",
+       "html[theme=dark],\n",
+       "body.vscode-dark {\n",
+       "  --xr-font-color0: rgba(255, 255, 255, 1);\n",
+       "  --xr-font-color2: rgba(255, 255, 255, 0.54);\n",
+       "  --xr-font-color3: rgba(255, 255, 255, 0.38);\n",
+       "  --xr-border-color: #1F1F1F;\n",
+       "  --xr-disabled-color: #515151;\n",
+       "  --xr-background-color: #111111;\n",
+       "  --xr-background-color-row-even: #111111;\n",
+       "  --xr-background-color-row-odd: #313131;\n",
+       "}\n",
+       "\n",
+       ".xr-wrap {\n",
+       "  display: block;\n",
+       "  min-width: 300px;\n",
+       "  max-width: 700px;\n",
+       "}\n",
+       "\n",
+       ".xr-text-repr-fallback {\n",
+       "  /* fallback to plain text repr when CSS is not injected (untrusted notebook) */\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-header {\n",
+       "  padding-top: 6px;\n",
+       "  padding-bottom: 6px;\n",
+       "  margin-bottom: 4px;\n",
+       "  border-bottom: solid 1px var(--xr-border-color);\n",
+       "}\n",
+       "\n",
+       ".xr-header > div,\n",
+       ".xr-header > ul {\n",
+       "  display: inline;\n",
+       "  margin-top: 0;\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type,\n",
+       ".xr-array-name {\n",
+       "  margin-left: 2px;\n",
+       "  margin-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-sections {\n",
+       "  padding-left: 0 !important;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 150px auto auto 1fr 20px 20px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input + label {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label {\n",
+       "  cursor: pointer;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label:hover {\n",
+       "  color: var(--xr-font-color0);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary {\n",
+       "  grid-column: 1;\n",
+       "  color: var(--xr-font-color2);\n",
+       "  font-weight: 500;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary > span {\n",
+       "  display: inline-block;\n",
+       "  padding-left: 0.5em;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in + label:before {\n",
+       "  display: inline-block;\n",
+       "  content: '►';\n",
+       "  font-size: 11px;\n",
+       "  width: 15px;\n",
+       "  text-align: center;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label:before {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label:before {\n",
+       "  content: '▼';\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label > span {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary,\n",
+       ".xr-section-inline-details {\n",
+       "  padding-top: 4px;\n",
+       "  padding-bottom: 4px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-inline-details {\n",
+       "  grid-column: 2 / -1;\n",
+       "}\n",
+       "\n",
+       ".xr-section-details {\n",
+       "  display: none;\n",
+       "  grid-column: 1 / -1;\n",
+       "  margin-bottom: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked ~ .xr-section-details {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap {\n",
+       "  grid-column: 1 / -1;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 20px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap > label {\n",
+       "  grid-column: 1;\n",
+       "  vertical-align: top;\n",
+       "}\n",
+       "\n",
+       ".xr-preview {\n",
+       "  color: var(--xr-font-color3);\n",
+       "}\n",
+       "\n",
+       ".xr-array-preview,\n",
+       ".xr-array-data {\n",
+       "  padding: 0 5px !important;\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-array-data,\n",
+       ".xr-array-in:checked ~ .xr-array-preview {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-array-in:checked ~ .xr-array-data,\n",
+       ".xr-array-preview {\n",
+       "  display: inline-block;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list {\n",
+       "  display: inline-block !important;\n",
+       "  list-style: none;\n",
+       "  padding: 0 !important;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li {\n",
+       "  display: inline-block;\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:before {\n",
+       "  content: '(';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:after {\n",
+       "  content: ')';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li:not(:last-child):after {\n",
+       "  content: ',';\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-has-index {\n",
+       "  font-weight: bold;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list,\n",
+       ".xr-var-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > div,\n",
+       ".xr-var-item label,\n",
+       ".xr-var-item > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-even);\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > .xr-var-name:hover span {\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list > li:nth-child(odd) > div,\n",
+       ".xr-var-list > li:nth-child(odd) > label,\n",
+       ".xr-var-list > li:nth-child(odd) > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-odd);\n",
+       "}\n",
+       "\n",
+       ".xr-var-name {\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dims {\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dtype {\n",
+       "  grid-column: 3;\n",
+       "  text-align: right;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-var-preview {\n",
+       "  grid-column: 4;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name,\n",
+       ".xr-var-dims,\n",
+       ".xr-var-dtype,\n",
+       ".xr-preview,\n",
+       ".xr-attrs dt {\n",
+       "  white-space: nowrap;\n",
+       "  overflow: hidden;\n",
+       "  text-overflow: ellipsis;\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name:hover,\n",
+       ".xr-var-dims:hover,\n",
+       ".xr-var-dtype:hover,\n",
+       ".xr-attrs dt:hover {\n",
+       "  overflow: visible;\n",
+       "  width: auto;\n",
+       "  z-index: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  display: none;\n",
+       "  background-color: var(--xr-background-color) !important;\n",
+       "  padding-bottom: 5px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs-in:checked ~ .xr-var-attrs,\n",
+       ".xr-var-data-in:checked ~ .xr-var-data {\n",
+       "  display: block;\n",
+       "}\n",
+       "\n",
+       ".xr-var-data > table {\n",
+       "  float: right;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name span,\n",
+       ".xr-var-data,\n",
+       ".xr-attrs {\n",
+       "  padding-left: 25px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs,\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  grid-column: 1 / -1;\n",
+       "}\n",
+       "\n",
+       "dl.xr-attrs {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 125px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt,\n",
+       ".xr-attrs dd {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  float: left;\n",
+       "  padding-right: 10px;\n",
+       "  width: auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt {\n",
+       "  font-weight: normal;\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt:hover span {\n",
+       "  display: inline-block;\n",
+       "  background: var(--xr-background-color);\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dd {\n",
+       "  grid-column: 2;\n",
+       "  white-space: pre-wrap;\n",
+       "  word-break: break-all;\n",
+       "}\n",
+       "\n",
+       ".xr-icon-database,\n",
+       ".xr-icon-file-text2 {\n",
+       "  display: inline-block;\n",
+       "  vertical-align: middle;\n",
+       "  width: 1em;\n",
+       "  height: 1.5em !important;\n",
+       "  stroke-width: 0;\n",
+       "  stroke: currentColor;\n",
+       "  fill: currentColor;\n",
+       "}\n",
+       "</style><pre class='xr-text-repr-fallback'>&lt;xarray.Dataset&gt;\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 1, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 1981-09-01T12:00:00\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 ...\n",
+       "    err      (time, zlev, lat, lon) float32 ...\n",
+       "    ice      (time, zlev, lat, lon) float32 ...\n",
+       "    sst      (time, zlev, lat, lon) float32 ...\n",
+       "Attributes:\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    id:                         oisst-avhrr-v02r01.19810901.nc\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    cdm_data_type:              Grid\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    product_version:            Version v02r01\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    keywords:                   Earth Science &gt; Oceans &gt; Ocean Temperature &gt; ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments &gt; Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    time_coverage_start:        1981-09-01T00:00:00Z\n",
+       "    time_coverage_end:          1981-09-01T23:59:59Z\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-...</pre><div class='xr-wrap' hidden><div class='xr-header'><div class='xr-obj-type'>xarray.Dataset</div></div><ul class='xr-sections'><li class='xr-section-item'><input id='section-dff4b653-806c-410c-89c9-c522a6a10bcd' class='xr-section-summary-in' type='checkbox' disabled ><label for='section-dff4b653-806c-410c-89c9-c522a6a10bcd' class='xr-section-summary'  title='Expand/collapse section'>Dimensions:</label><div class='xr-section-inline-details'><ul class='xr-dim-list'><li><span class='xr-has-index'>lat</span>: 720</li><li><span class='xr-has-index'>lon</span>: 1440</li><li><span class='xr-has-index'>time</span>: 1</li><li><span class='xr-has-index'>zlev</span>: 1</li></ul></div><div class='xr-section-details'></div></li><li class='xr-section-item'><input id='section-5a8a0f1c-ca68-4d12-b5b7-894acaf21617' class='xr-section-summary-in' type='checkbox'  checked><label for='section-5a8a0f1c-ca68-4d12-b5b7-894acaf21617' class='xr-section-summary' >Coordinates: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lat</span></div><div class='xr-var-dims'>(lat)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>-89.88 -89.62 ... 89.62 89.88</div><input id='attrs-97f3c5de-bfd9-451a-884a-2a21dd4caa63' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-97f3c5de-bfd9-451a-884a-2a21dd4caa63' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-bfac4b1f-90d4-4dd2-b5dc-0a1c5b735bfb' class='xr-var-data-in' type='checkbox'><label for='data-bfac4b1f-90d4-4dd2-b5dc-0a1c5b735bfb' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Latitude</dd><dt><span>units :</span></dt><dd>degrees_north</dd><dt><span>grids :</span></dt><dd>Uniform grid from -89.875 to 89.875 by 0.25</dd></dl></div><div class='xr-var-data'><pre>array([-89.875, -89.625, -89.375, ...,  89.375,  89.625,  89.875],\n",
+       "      dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lon</span></div><div class='xr-var-dims'>(lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.125 0.375 0.625 ... 359.6 359.9</div><input id='attrs-ac56ac67-f800-436f-9ecc-95970a4a5634' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-ac56ac67-f800-436f-9ecc-95970a4a5634' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-723089f1-7ac0-4b8d-8ccd-c1cccf321ff2' class='xr-var-data-in' type='checkbox'><label for='data-723089f1-7ac0-4b8d-8ccd-c1cccf321ff2' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Longitude</dd><dt><span>units :</span></dt><dd>degrees_east</dd><dt><span>grids :</span></dt><dd>Uniform grid from 0.125 to 359.875 by 0.25</dd></dl></div><div class='xr-var-data'><pre>array([1.25000e-01, 3.75000e-01, 6.25000e-01, ..., 3.59375e+02, 3.59625e+02,\n",
+       "       3.59875e+02], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>time</span></div><div class='xr-var-dims'>(time)</div><div class='xr-var-dtype'>datetime64[ns]</div><div class='xr-var-preview xr-preview'>1981-09-01T12:00:00</div><input id='attrs-807259da-3580-473a-a3d3-3d39478ba932' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-807259da-3580-473a-a3d3-3d39478ba932' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-7ba797ec-e18b-4a46-aef9-6bf550c04336' class='xr-var-data-in' type='checkbox'><label for='data-7ba797ec-e18b-4a46-aef9-6bf550c04336' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Center time of the day</dd></dl></div><div class='xr-var-data'><pre>array([&#x27;1981-09-01T12:00:00.000000000&#x27;], dtype=&#x27;datetime64[ns]&#x27;)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>zlev</span></div><div class='xr-var-dims'>(zlev)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.0</div><input id='attrs-08a1f153-0abc-4302-a540-96b238535eb7' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-08a1f153-0abc-4302-a540-96b238535eb7' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-d63f0fd2-4809-46d9-9d57-992424420ce4' class='xr-var-data-in' type='checkbox'><label for='data-d63f0fd2-4809-46d9-9d57-992424420ce4' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea surface height</dd><dt><span>units :</span></dt><dd>meters</dd><dt><span>actual_range :</span></dt><dd>0, 0</dd><dt><span>positive :</span></dt><dd>down</dd></dl></div><div class='xr-var-data'><pre>array([0.], dtype=float32)</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-92ca863a-2bd7-4646-9b5f-5258c82deec7' class='xr-section-summary-in' type='checkbox'  checked><label for='section-92ca863a-2bd7-4646-9b5f-5258c82deec7' class='xr-section-summary' >Data variables: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span>anom</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>...</div><input id='attrs-0d558848-249f-45d8-a102-6c68f1c3fcb1' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-0d558848-249f-45d8-a102-6c68f1c3fcb1' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-f5b43a4a-3679-4ed0-a04c-6c1b2e6f1ed7' class='xr-var-data-in' type='checkbox'><label for='data-f5b43a4a-3679-4ed0-a04c-6c1b2e6f1ed7' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature anomalies</dd><dt><span>valid_min :</span></dt><dd>-1200</dd><dt><span>valid_max :</span></dt><dd>1200</dd><dt><span>units :</span></dt><dd>Celsius</dd></dl></div><div class='xr-var-data'><pre>[1036800 values with dtype=float32]</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>err</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>...</div><input id='attrs-7780c73f-c575-4639-81c4-d41da826750c' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-7780c73f-c575-4639-81c4-d41da826750c' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-f67bc234-31a2-4467-97d7-24f0374450cb' class='xr-var-data-in' type='checkbox'><label for='data-f67bc234-31a2-4467-97d7-24f0374450cb' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Estimated error standard deviation of analysed_sst</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_min :</span></dt><dd>0</dd><dt><span>valid_max :</span></dt><dd>1000</dd></dl></div><div class='xr-var-data'><pre>[1036800 values with dtype=float32]</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>ice</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>...</div><input id='attrs-72ccb88a-5ad8-4203-95d6-ed242cddcb83' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-72ccb88a-5ad8-4203-95d6-ed242cddcb83' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-f564949c-0876-456b-b229-54207fbadd08' class='xr-var-data-in' type='checkbox'><label for='data-f564949c-0876-456b-b229-54207fbadd08' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea ice concentration</dd><dt><span>units :</span></dt><dd>%</dd><dt><span>valid_min :</span></dt><dd>0</dd><dt><span>valid_max :</span></dt><dd>100</dd></dl></div><div class='xr-var-data'><pre>[1036800 values with dtype=float32]</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>sst</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>...</div><input id='attrs-30586a1a-a146-40a1-a29c-1aba22b22288' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-30586a1a-a146-40a1-a29c-1aba22b22288' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-94e832e0-58fc-4ef5-ad0f-7dffdcb0e62a' class='xr-var-data-in' type='checkbox'><label for='data-94e832e0-58fc-4ef5-ad0f-7dffdcb0e62a' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_min :</span></dt><dd>-300</dd><dt><span>valid_max :</span></dt><dd>4500</dd></dl></div><div class='xr-var-data'><pre>[1036800 values with dtype=float32]</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-3b6b8503-a6da-4edc-b5be-eec41e00d968' class='xr-section-summary-in' type='checkbox'  ><label for='section-3b6b8503-a6da-4edc-b5be-eec41e00d968' class='xr-section-summary' >Attributes: <span>(37)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><dl class='xr-attrs'><dt><span>title :</span></dt><dd>NOAA/NCEI 1/4 Degree Daily Optimum Interpolation Sea Surface Temperature (OISST) Analysis, Version 2.1 - Final</dd><dt><span>source :</span></dt><dd>ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfinder_AVHRR, Navy_AVHRR</dd><dt><span>id :</span></dt><dd>oisst-avhrr-v02r01.19810901.nc</dd><dt><span>naming_authority :</span></dt><dd>gov.noaa.ncei</dd><dt><span>summary :</span></dt><dd>NOAAs 1/4-degree Daily Optimum Interpolation Sea Surface Temperature (OISST) (sometimes referred to as Reynolds SST, which however also refers to earlier products at different resolution), currently available as version v02r01, is created by interpolating and extrapolating SST observations from different sources, resulting in a smoothed complete field. The sources of data are satellite (AVHRR) and in situ platforms (i.e., ships and buoys), and the specific datasets employed may change over time. At the marginal ice zone, sea ice concentrations are used to generate proxy SSTs.  A preliminary version of this file is produced in near-real time (1-day latency), and then replaced with a final version after 2 weeks. Note that this is the AVHRR-ONLY DOISST, available from Oct 1981, but there is a companion DOISST product that includes microwave satellite data, available from June 2002</dd><dt><span>cdm_data_type :</span></dt><dd>Grid</dd><dt><span>history :</span></dt><dd>Final file created using preliminary as first guess, and 3 days of AVHRR data. Preliminary uses only 1 day of AVHRR data.</dd><dt><span>date_modified :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>date_created :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>product_version :</span></dt><dd>Version v02r01</dd><dt><span>processing_level :</span></dt><dd>NOAA Level 4</dd><dt><span>institution :</span></dt><dd>NOAA/National Centers for Environmental Information</dd><dt><span>creator_url :</span></dt><dd>https://www.ncei.noaa.gov/</dd><dt><span>creator_email :</span></dt><dd>oisst-help@noaa.gov</dd><dt><span>keywords :</span></dt><dd>Earth Science &gt; Oceans &gt; Ocean Temperature &gt; Sea Surface Temperature</dd><dt><span>keywords_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Earth Science Keywords</dd><dt><span>platform :</span></dt><dd>Ships, buoys, Argo floats, MetOp-A, MetOp-B</dd><dt><span>platform_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Platform Keywords</dd><dt><span>instrument :</span></dt><dd>Earth Remote Sensing Instruments &gt; Passive Remote Sensing &gt; Spectrometers/Radiometers &gt; Imaging Spectrometers/Radiometers &gt; AVHRR &gt; Advanced Very High Resolution Radiometer</dd><dt><span>instrument_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Instrument Keywords</dd><dt><span>standard_name_vocabulary :</span></dt><dd>CF Standard Name Table (v40, 25 January 2017)</dd><dt><span>geospatial_lat_min :</span></dt><dd>-90.0</dd><dt><span>geospatial_lat_max :</span></dt><dd>90.0</dd><dt><span>geospatial_lon_min :</span></dt><dd>0.0</dd><dt><span>geospatial_lon_max :</span></dt><dd>360.0</dd><dt><span>geospatial_lat_units :</span></dt><dd>degrees_north</dd><dt><span>geospatial_lat_resolution :</span></dt><dd>0.25</dd><dt><span>geospatial_lon_units :</span></dt><dd>degrees_east</dd><dt><span>geospatial_lon_resolution :</span></dt><dd>0.25</dd><dt><span>time_coverage_start :</span></dt><dd>1981-09-01T00:00:00Z</dd><dt><span>time_coverage_end :</span></dt><dd>1981-09-01T23:59:59Z</dd><dt><span>metadata_link :</span></dt><dd>https://doi.org/10.25921/RE9P-PT57</dd><dt><span>ncei_template_version :</span></dt><dd>NCEI_NetCDF_Grid_Template_v2.0</dd><dt><span>comment :</span></dt><dd>Data was converted from NetCDF-3 to NetCDF-4 format with metadata updates in November 2017.</dd><dt><span>sensor :</span></dt><dd>Thermometer, AVHRR</dd><dt><span>Conventions :</span></dt><dd>CF-1.6, ACDD-1.3</dd><dt><span>references :</span></dt><dd>Reynolds, et al.(2007) Daily High-Resolution-Blended Analyses for Sea Surface Temperature (available at https://doi.org/10.1175/2007JCLI1824.1). Banzon, et al.(2016) A long-term record of blended satellite and in situ sea-surface temperature for climate monitoring, modeling and environmental studies (available at https://doi.org/10.5194/essd-8-165-2016). Huang et al. (2020) Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version v02r01, submitted.Climatology is based on 1971-2000 OI.v2 SST. Satellite data: Pathfinder AVHRR SST and Navy AVHRR SST. Ice data: NCEP Ice and GSFC Ice.</dd></dl></div></li></ul></div></div>"
+      ],
+      "text/plain": [
+       "<xarray.Dataset>\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 1, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 1981-09-01T12:00:00\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 ...\n",
+       "    err      (time, zlev, lat, lon) float32 ...\n",
+       "    ice      (time, zlev, lat, lon) float32 ...\n",
+       "    sst      (time, zlev, lat, lon) float32 ...\n",
+       "Attributes:\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    id:                         oisst-avhrr-v02r01.19810901.nc\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    cdm_data_type:              Grid\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    product_version:            Version v02r01\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    keywords:                   Earth Science > Oceans > Ocean Temperature > ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments > Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    time_coverage_start:        1981-09-01T00:00:00Z\n",
+       "    time_coverage_end:          1981-09-01T23:59:59Z\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-..."
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
     "import xarray as xr\n",
     "\n",
@@ -70,9 +535,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "File size is 16.597452 MB\n"
+     ]
+    }
+   ],
    "source": [
     "print(f\"File size is {ds.nbytes/1e6} MB\")"
    ]
@@ -101,9 +574,60 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\u001b[0;31mInit signature:\u001b[0m\n",
+       "\u001b[0mNetCDFtoZarrSequentialRecipe\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0minput_urls\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mIterable\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0msequence_dim\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0minputs_per_chunk\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mint\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mnitems_per_input\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mint\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mtarget\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpangeo_forge\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstorage\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mAbstractTarget\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mNoneType\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m<\u001b[0m\u001b[0mfactory\u001b[0m\u001b[0;34m>\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0minput_cache\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpangeo_forge\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstorage\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mAbstractTarget\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mNoneType\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m<\u001b[0m\u001b[0mfactory\u001b[0m\u001b[0;34m>\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mrequire_cache\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mbool\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mconsolidate_zarr\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mbool\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mxarray_open_kwargs\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mdict\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m<\u001b[0m\u001b[0mfactory\u001b[0m\u001b[0;34m>\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mxarray_concat_kwargs\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mdict\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m<\u001b[0m\u001b[0mfactory\u001b[0m\u001b[0;34m>\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m    \u001b[0mdelete_input_encoding\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mbool\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
+       "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+       "\u001b[0;31mDocstring:\u001b[0m     \n",
+       "This class represents a dataset composed of many individual NetCDF files.\n",
+       "The files are arraged in a sequence along a single dimension, called the\n",
+       "`sequence_dim`. Each file may contain multiple variables.\n",
+       "\n",
+       "The dataset is assembled by concatenating all of these files along `sequence_dim`.\n",
+       "The target is written in Zarr format.\n",
+       "\n",
+       "This class uses Xarray to read and write data.\n",
+       "\n",
+       ":param input_urls: The inputs used to generate the dataset.\n",
+       ":param sequence_dim: The dimension name along which the inputs will be concatenated.\n",
+       ":param inputs_per_chunk: The number of inputs to use in each chunk.\n",
+       ":param nitems_per_input: The length of each input along the `sequence_dim` dimension.\n",
+       ":param target: A location in which to put the dataset. Can also be assigned at run time.\n",
+       ":param input_cache: A location in which to cache temporary data.\n",
+       ":param require_cache: Whether to allow opening inputs directly which have not\n",
+       "  yet been cached. This could lead to very slow behavior if the inputs\n",
+       "  live on a slow network.\n",
+       ":param consolidate_zarr: Whether to consolidate the resulting Zarr dataset.\n",
+       ":param xarray_open_kwargs: Extra options for opening the inputs with Xarray.\n",
+       ":param xarray_concat_kwargs: Extra options to pass to Xarray when concatenating\n",
+       "  the inputs to form a chunk.\n",
+       ":param delete_input_encoding: Whether to remove Xarray encoding from variables\n",
+       "  in the input dataset\n",
+       "\u001b[0;31mFile:\u001b[0m           ~/pangeo-forge/pangeo-forge/pangeo_forge/recipe.py\n",
+       "\u001b[0;31mType:\u001b[0m           ABCMeta\n",
+       "\u001b[0;31mSubclasses:\u001b[0m     \n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "from pangeo_forge.recipe import NetCDFtoZarrSequentialRecipe\n",
     "NetCDFtoZarrSequentialRecipe?"
@@ -131,7 +655,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -151,9 +675,27 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 6,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found 14372 files!\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202101/oisst-avhrr-v02r01.20210105.nc'"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
     "import pandas as pd\n",
     "\n",
@@ -174,21 +716,2909 @@
    "source": [
     "That's a lot of files!\n",
     "\n",
-    "The other remaining parameter is `sequence_dim`. It's just `\"time\"`. We can now\n",
-    "instantiate our recipe.\n"
+    "The other remaining parameter is `sequence_dim`. It's just `\"time\"`.  This is enough to initialize the recipe."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 7,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "NetCDFtoZarrSequentialRecipe(sequence_dim='time', inputs_per_chunk=1, nitems_per_input=1, target=<pangeo_forge.storage.UninitializedTarget object at 0x7ff8b0797f10>, input_cache=<pangeo_forge.storage.UninitializedTarget object at 0x7ff870359820>, require_cache=True, consolidate_zarr=True, xarray_open_kwargs={}, xarray_concat_kwargs={}, delete_input_encoding=True)"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
     "recipe = NetCDFtoZarrSequentialRecipe(\n",
-    "    input_urls=input_urls, sequence_dim=\"time\"\n",
+    "    input_urls=input_urls,\n",
+    "    sequence_dim=\"time\"\n",
     ")\n",
     "recipe"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "However, now let's think about the Zarr chunks that this recipe will produce.\n",
+    "Each target chunk corresponds to one input. So each variable chunk will only be a few MB.\n",
+    "That is too small. Let's increase `inputs_per_chunk` to 20.\n",
+    "This means that we will need to be able to hold 20 files like the one we examined above in memory at once.\n",
+    "That's `16MB * 20 = 320MB`. Not a problem!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "NetCDFtoZarrSequentialRecipe(sequence_dim='time', inputs_per_chunk=20, nitems_per_input=1, target=<pangeo_forge.storage.UninitializedTarget object at 0x7ff8904061c0>, input_cache=<pangeo_forge.storage.UninitializedTarget object at 0x7ff8904061f0>, require_cache=True, consolidate_zarr=True, xarray_open_kwargs={}, xarray_concat_kwargs={}, delete_input_encoding=True)"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "recipe = NetCDFtoZarrSequentialRecipe(\n",
+    "    input_urls=input_urls,\n",
+    "    sequence_dim=\"time\",\n",
+    "    inputs_per_chunk=20\n",
+    ")\n",
+    "recipe"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 4: Play with the recipe\n",
+    "\n",
+    "Now we will just explore our recipe a bit to check whether things make sense.\n",
+    "\n",
+    "We can see how many inputs the recipe has like this:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "14372"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "all_inputs = list(recipe.iter_inputs())\n",
+    "len(all_inputs)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "And how many chunks:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "719"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "all_chunks = list(recipe.iter_chunks())\n",
+    "len(all_chunks)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "And we can see their dependencies as follows:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "first chunk ID: 0\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "('https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810901.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810902.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810903.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810904.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810905.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810906.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810907.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810908.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810909.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810910.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810911.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810912.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810913.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810914.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810915.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810916.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810917.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810918.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810919.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810920.nc')"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "print(f'first chunk ID: {all_chunks[0]}')\n",
+    "recipe.inputs_for_chunk(all_chunks[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "last chunk ID: 718\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "('https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201225.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201226.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201227.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201228.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201229.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201230.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202012/oisst-avhrr-v02r01.20201231.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202101/oisst-avhrr-v02r01.20210101.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202101/oisst-avhrr-v02r01.20210102.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202101/oisst-avhrr-v02r01.20210103.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202101/oisst-avhrr-v02r01.20210104.nc',\n",
+       " 'https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/202101/oisst-avhrr-v02r01.20210105.nc')"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "print(f'last chunk ID: {all_chunks[-1]}')\n",
+    "recipe.inputs_for_chunk(all_chunks[-1])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Those are all the files that will go into the first and last chunk. \n",
+    "\n",
+    "We can now try to load the first chunk. This will raise an exception because we have not initialized any targets."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pangeo_forge.storage.UninitializedTargetError'>\n"
+     ]
+    }
+   ],
+   "source": [
+    "try:\n",
+    "    recipe.open_chunk(all_chunks[0])\n",
+    "except Exception as e:\n",
+    "    print(type(e))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 5: Create storage targets\n",
+    "\n",
+    "We in order to run our recipe, we need to define to places to store data:\n",
+    "- The Input Cache, where we will temporarily store the files we have downloaded\n",
+    "- The Target, where the final Zarr dataset will live\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "NetCDFtoZarrSequentialRecipe(sequence_dim='time', inputs_per_chunk=20, nitems_per_input=1, target=FSSpecTarget(fs=<fsspec.implementations.local.LocalFileSystem object at 0x7ff8b0842520>, root_path='/var/folders/n8/63q49ms55wxcj_gfbtykwp5r0000gn/T/tmppoaidedy'), input_cache=CacheFSSpecTarget(fs=<fsspec.implementations.local.LocalFileSystem object at 0x7ff8b0842520>, root_path='/var/folders/n8/63q49ms55wxcj_gfbtykwp5r0000gn/T/tmp61q8ifva'), require_cache=True, consolidate_zarr=True, xarray_open_kwargs={}, xarray_concat_kwargs={}, delete_input_encoding=True)"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import tempfile\n",
+    "from fsspec.implementations.local import LocalFileSystem\n",
+    "from pangeo_forge.storage import FSSpecTarget, CacheFSSpecTarget\n",
+    "\n",
+    "fs_local = LocalFileSystem()\n",
+    "\n",
+    "cache_dir = tempfile.TemporaryDirectory()\n",
+    "cache_target = CacheFSSpecTarget(fs_local, cache_dir.name)\n",
+    "\n",
+    "target_dir = tempfile.TemporaryDirectory()\n",
+    "target = FSSpecTarget(fs_local, target_dir.name)\n",
+    "\n",
+    "recipe.input_cache = cache_target\n",
+    "recipe.target = target\n",
+    "recipe"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we try to load the chunk."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "You are trying to open input https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum-interpolation/v2.1/access/avhrr/198109/oisst-avhrr-v02r01.19810901.nc, but the file is not cached yet. First call `cache_input` or set `require_cache=False`.\n"
+     ]
+    }
+   ],
+   "source": [
+    "try:\n",
+    "    recipe.open_chunk(all_chunks[0])\n",
+    "except Exception as e:\n",
+    "    print(e)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "It still didn't work! That's because we have not cached the inputs yet."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for input_file in recipe.inputs_for_chunk(all_chunks[0]):\n",
+    "    recipe.cache_input(input_file)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 6: Examine some chunks\n",
+    "\n",
+    "Now we can finally open the first chunk!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div><svg style=\"position: absolute; width: 0; height: 0; overflow: hidden\">\n",
+       "<defs>\n",
+       "<symbol id=\"icon-database\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M16 0c-8.837 0-16 2.239-16 5v4c0 2.761 7.163 5 16 5s16-2.239 16-5v-4c0-2.761-7.163-5-16-5z\"></path>\n",
+       "<path d=\"M16 17c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "<path d=\"M16 26c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "</symbol>\n",
+       "<symbol id=\"icon-file-text2\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M28.681 7.159c-0.694-0.947-1.662-2.053-2.724-3.116s-2.169-2.030-3.116-2.724c-1.612-1.182-2.393-1.319-2.841-1.319h-15.5c-1.378 0-2.5 1.121-2.5 2.5v27c0 1.378 1.122 2.5 2.5 2.5h23c1.378 0 2.5-1.122 2.5-2.5v-19.5c0-0.448-0.137-1.23-1.319-2.841zM24.543 5.457c0.959 0.959 1.712 1.825 2.268 2.543h-4.811v-4.811c0.718 0.556 1.584 1.309 2.543 2.268zM28 29.5c0 0.271-0.229 0.5-0.5 0.5h-23c-0.271 0-0.5-0.229-0.5-0.5v-27c0-0.271 0.229-0.5 0.5-0.5 0 0 15.499-0 15.5 0v7c0 0.552 0.448 1 1 1h7v19.5z\"></path>\n",
+       "<path d=\"M23 26h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 22h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 18h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "</symbol>\n",
+       "</defs>\n",
+       "</svg>\n",
+       "<style>/* CSS stylesheet for displaying xarray objects in jupyterlab.\n",
+       " *\n",
+       " */\n",
+       "\n",
+       ":root {\n",
+       "  --xr-font-color0: var(--jp-content-font-color0, rgba(0, 0, 0, 1));\n",
+       "  --xr-font-color2: var(--jp-content-font-color2, rgba(0, 0, 0, 0.54));\n",
+       "  --xr-font-color3: var(--jp-content-font-color3, rgba(0, 0, 0, 0.38));\n",
+       "  --xr-border-color: var(--jp-border-color2, #e0e0e0);\n",
+       "  --xr-disabled-color: var(--jp-layout-color3, #bdbdbd);\n",
+       "  --xr-background-color: var(--jp-layout-color0, white);\n",
+       "  --xr-background-color-row-even: var(--jp-layout-color1, white);\n",
+       "  --xr-background-color-row-odd: var(--jp-layout-color2, #eeeeee);\n",
+       "}\n",
+       "\n",
+       "html[theme=dark],\n",
+       "body.vscode-dark {\n",
+       "  --xr-font-color0: rgba(255, 255, 255, 1);\n",
+       "  --xr-font-color2: rgba(255, 255, 255, 0.54);\n",
+       "  --xr-font-color3: rgba(255, 255, 255, 0.38);\n",
+       "  --xr-border-color: #1F1F1F;\n",
+       "  --xr-disabled-color: #515151;\n",
+       "  --xr-background-color: #111111;\n",
+       "  --xr-background-color-row-even: #111111;\n",
+       "  --xr-background-color-row-odd: #313131;\n",
+       "}\n",
+       "\n",
+       ".xr-wrap {\n",
+       "  display: block;\n",
+       "  min-width: 300px;\n",
+       "  max-width: 700px;\n",
+       "}\n",
+       "\n",
+       ".xr-text-repr-fallback {\n",
+       "  /* fallback to plain text repr when CSS is not injected (untrusted notebook) */\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-header {\n",
+       "  padding-top: 6px;\n",
+       "  padding-bottom: 6px;\n",
+       "  margin-bottom: 4px;\n",
+       "  border-bottom: solid 1px var(--xr-border-color);\n",
+       "}\n",
+       "\n",
+       ".xr-header > div,\n",
+       ".xr-header > ul {\n",
+       "  display: inline;\n",
+       "  margin-top: 0;\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type,\n",
+       ".xr-array-name {\n",
+       "  margin-left: 2px;\n",
+       "  margin-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-sections {\n",
+       "  padding-left: 0 !important;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 150px auto auto 1fr 20px 20px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input + label {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label {\n",
+       "  cursor: pointer;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label:hover {\n",
+       "  color: var(--xr-font-color0);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary {\n",
+       "  grid-column: 1;\n",
+       "  color: var(--xr-font-color2);\n",
+       "  font-weight: 500;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary > span {\n",
+       "  display: inline-block;\n",
+       "  padding-left: 0.5em;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in + label:before {\n",
+       "  display: inline-block;\n",
+       "  content: '►';\n",
+       "  font-size: 11px;\n",
+       "  width: 15px;\n",
+       "  text-align: center;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label:before {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label:before {\n",
+       "  content: '▼';\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label > span {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary,\n",
+       ".xr-section-inline-details {\n",
+       "  padding-top: 4px;\n",
+       "  padding-bottom: 4px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-inline-details {\n",
+       "  grid-column: 2 / -1;\n",
+       "}\n",
+       "\n",
+       ".xr-section-details {\n",
+       "  display: none;\n",
+       "  grid-column: 1 / -1;\n",
+       "  margin-bottom: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked ~ .xr-section-details {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap {\n",
+       "  grid-column: 1 / -1;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 20px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap > label {\n",
+       "  grid-column: 1;\n",
+       "  vertical-align: top;\n",
+       "}\n",
+       "\n",
+       ".xr-preview {\n",
+       "  color: var(--xr-font-color3);\n",
+       "}\n",
+       "\n",
+       ".xr-array-preview,\n",
+       ".xr-array-data {\n",
+       "  padding: 0 5px !important;\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-array-data,\n",
+       ".xr-array-in:checked ~ .xr-array-preview {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-array-in:checked ~ .xr-array-data,\n",
+       ".xr-array-preview {\n",
+       "  display: inline-block;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list {\n",
+       "  display: inline-block !important;\n",
+       "  list-style: none;\n",
+       "  padding: 0 !important;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li {\n",
+       "  display: inline-block;\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:before {\n",
+       "  content: '(';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:after {\n",
+       "  content: ')';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li:not(:last-child):after {\n",
+       "  content: ',';\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-has-index {\n",
+       "  font-weight: bold;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list,\n",
+       ".xr-var-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > div,\n",
+       ".xr-var-item label,\n",
+       ".xr-var-item > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-even);\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > .xr-var-name:hover span {\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list > li:nth-child(odd) > div,\n",
+       ".xr-var-list > li:nth-child(odd) > label,\n",
+       ".xr-var-list > li:nth-child(odd) > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-odd);\n",
+       "}\n",
+       "\n",
+       ".xr-var-name {\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dims {\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dtype {\n",
+       "  grid-column: 3;\n",
+       "  text-align: right;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-var-preview {\n",
+       "  grid-column: 4;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name,\n",
+       ".xr-var-dims,\n",
+       ".xr-var-dtype,\n",
+       ".xr-preview,\n",
+       ".xr-attrs dt {\n",
+       "  white-space: nowrap;\n",
+       "  overflow: hidden;\n",
+       "  text-overflow: ellipsis;\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name:hover,\n",
+       ".xr-var-dims:hover,\n",
+       ".xr-var-dtype:hover,\n",
+       ".xr-attrs dt:hover {\n",
+       "  overflow: visible;\n",
+       "  width: auto;\n",
+       "  z-index: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  display: none;\n",
+       "  background-color: var(--xr-background-color) !important;\n",
+       "  padding-bottom: 5px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs-in:checked ~ .xr-var-attrs,\n",
+       ".xr-var-data-in:checked ~ .xr-var-data {\n",
+       "  display: block;\n",
+       "}\n",
+       "\n",
+       ".xr-var-data > table {\n",
+       "  float: right;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name span,\n",
+       ".xr-var-data,\n",
+       ".xr-attrs {\n",
+       "  padding-left: 25px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs,\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  grid-column: 1 / -1;\n",
+       "}\n",
+       "\n",
+       "dl.xr-attrs {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 125px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt,\n",
+       ".xr-attrs dd {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  float: left;\n",
+       "  padding-right: 10px;\n",
+       "  width: auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt {\n",
+       "  font-weight: normal;\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt:hover span {\n",
+       "  display: inline-block;\n",
+       "  background: var(--xr-background-color);\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dd {\n",
+       "  grid-column: 2;\n",
+       "  white-space: pre-wrap;\n",
+       "  word-break: break-all;\n",
+       "}\n",
+       "\n",
+       ".xr-icon-database,\n",
+       ".xr-icon-file-text2 {\n",
+       "  display: inline-block;\n",
+       "  vertical-align: middle;\n",
+       "  width: 1em;\n",
+       "  height: 1.5em !important;\n",
+       "  stroke-width: 0;\n",
+       "  stroke: currentColor;\n",
+       "  fill: currentColor;\n",
+       "}\n",
+       "</style><pre class='xr-text-repr-fallback'>&lt;xarray.Dataset&gt;\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 20, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 1981-09-01T12:00:00 ... 1981-09-20T12:00:00\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 nan nan nan nan ... 0.41 0.41 0.41\n",
+       "    err      (time, zlev, lat, lon) float32 nan nan nan nan ... 0.3 0.3 0.3 0.3\n",
+       "    ice      (time, zlev, lat, lon) float32 nan nan nan nan ... 0.9 0.9 0.9 0.9\n",
+       "    sst      (time, zlev, lat, lon) float32 nan nan nan ... -1.39 -1.39 -1.39\n",
+       "Attributes:\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    id:                         oisst-avhrr-v02r01.19810901.nc\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    cdm_data_type:              Grid\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    product_version:            Version v02r01\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    keywords:                   Earth Science &gt; Oceans &gt; Ocean Temperature &gt; ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments &gt; Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    time_coverage_start:        1981-09-01T00:00:00Z\n",
+       "    time_coverage_end:          1981-09-01T23:59:59Z\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-...</pre><div class='xr-wrap' hidden><div class='xr-header'><div class='xr-obj-type'>xarray.Dataset</div></div><ul class='xr-sections'><li class='xr-section-item'><input id='section-696f68f1-1a15-44e5-a401-3de89cfc5e31' class='xr-section-summary-in' type='checkbox' disabled ><label for='section-696f68f1-1a15-44e5-a401-3de89cfc5e31' class='xr-section-summary'  title='Expand/collapse section'>Dimensions:</label><div class='xr-section-inline-details'><ul class='xr-dim-list'><li><span class='xr-has-index'>lat</span>: 720</li><li><span class='xr-has-index'>lon</span>: 1440</li><li><span class='xr-has-index'>time</span>: 20</li><li><span class='xr-has-index'>zlev</span>: 1</li></ul></div><div class='xr-section-details'></div></li><li class='xr-section-item'><input id='section-e9966ab1-dd51-4a96-801b-4756781bc00c' class='xr-section-summary-in' type='checkbox'  checked><label for='section-e9966ab1-dd51-4a96-801b-4756781bc00c' class='xr-section-summary' >Coordinates: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lat</span></div><div class='xr-var-dims'>(lat)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>-89.88 -89.62 ... 89.62 89.88</div><input id='attrs-eca21f1f-f3a9-4743-b43e-5f5238cd0aa4' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-eca21f1f-f3a9-4743-b43e-5f5238cd0aa4' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-54d9f9b4-7b08-4826-b406-e65e6a899023' class='xr-var-data-in' type='checkbox'><label for='data-54d9f9b4-7b08-4826-b406-e65e6a899023' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Latitude</dd><dt><span>units :</span></dt><dd>degrees_north</dd><dt><span>grids :</span></dt><dd>Uniform grid from -89.875 to 89.875 by 0.25</dd></dl></div><div class='xr-var-data'><pre>array([-89.875, -89.625, -89.375, ...,  89.375,  89.625,  89.875],\n",
+       "      dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lon</span></div><div class='xr-var-dims'>(lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.125 0.375 0.625 ... 359.6 359.9</div><input id='attrs-87bbdacb-8dff-4ffc-b665-4ca46b292eda' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-87bbdacb-8dff-4ffc-b665-4ca46b292eda' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-4a15b971-93a2-45ff-b051-840c8b8cd453' class='xr-var-data-in' type='checkbox'><label for='data-4a15b971-93a2-45ff-b051-840c8b8cd453' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Longitude</dd><dt><span>units :</span></dt><dd>degrees_east</dd><dt><span>grids :</span></dt><dd>Uniform grid from 0.125 to 359.875 by 0.25</dd></dl></div><div class='xr-var-data'><pre>array([1.25000e-01, 3.75000e-01, 6.25000e-01, ..., 3.59375e+02, 3.59625e+02,\n",
+       "       3.59875e+02], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>time</span></div><div class='xr-var-dims'>(time)</div><div class='xr-var-dtype'>datetime64[ns]</div><div class='xr-var-preview xr-preview'>1981-09-01T12:00:00 ... 1981-09-...</div><input id='attrs-11f73010-d572-4464-89ad-9dd3994deb4f' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-11f73010-d572-4464-89ad-9dd3994deb4f' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-2daab819-0a9a-48a9-90dd-45f74c4c7f43' class='xr-var-data-in' type='checkbox'><label for='data-2daab819-0a9a-48a9-90dd-45f74c4c7f43' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Center time of the day</dd></dl></div><div class='xr-var-data'><pre>array([&#x27;1981-09-01T12:00:00.000000000&#x27;, &#x27;1981-09-02T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-03T12:00:00.000000000&#x27;, &#x27;1981-09-04T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-05T12:00:00.000000000&#x27;, &#x27;1981-09-06T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-07T12:00:00.000000000&#x27;, &#x27;1981-09-08T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-09T12:00:00.000000000&#x27;, &#x27;1981-09-10T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-11T12:00:00.000000000&#x27;, &#x27;1981-09-12T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-13T12:00:00.000000000&#x27;, &#x27;1981-09-14T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-15T12:00:00.000000000&#x27;, &#x27;1981-09-16T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-17T12:00:00.000000000&#x27;, &#x27;1981-09-18T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-19T12:00:00.000000000&#x27;, &#x27;1981-09-20T12:00:00.000000000&#x27;],\n",
+       "      dtype=&#x27;datetime64[ns]&#x27;)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>zlev</span></div><div class='xr-var-dims'>(zlev)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.0</div><input id='attrs-9ffc7cef-bd2e-4ca0-8622-c518f4f4ccbd' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-9ffc7cef-bd2e-4ca0-8622-c518f4f4ccbd' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-93e40272-308d-4067-9240-29110e0a8588' class='xr-var-data-in' type='checkbox'><label for='data-93e40272-308d-4067-9240-29110e0a8588' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea surface height</dd><dt><span>units :</span></dt><dd>meters</dd><dt><span>actual_range :</span></dt><dd>0, 0</dd><dt><span>positive :</span></dt><dd>down</dd></dl></div><div class='xr-var-data'><pre>array([0.], dtype=float32)</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-f1af56de-aaee-4362-b546-a2a030e042df' class='xr-section-summary-in' type='checkbox'  checked><label for='section-f1af56de-aaee-4362-b546-a2a030e042df' class='xr-section-summary' >Data variables: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span>anom</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan nan ... 0.41 0.41 0.41</div><input id='attrs-6261b599-b4d9-4f33-a209-6c96f12d79d0' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-6261b599-b4d9-4f33-a209-6c96f12d79d0' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-d7fe4a10-9959-4194-9a58-d43d76976693' class='xr-var-data-in' type='checkbox'><label for='data-d7fe4a10-9959-4194-9a58-d43d76976693' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature anomalies</dd><dt><span>valid_min :</span></dt><dd>-1200</dd><dt><span>valid_max :</span></dt><dd>1200</dd><dt><span>units :</span></dt><dd>Celsius</dd></dl></div><div class='xr-var-data'><pre>array([[[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.71, 0.71, 0.71, ..., 0.71, 0.71, 0.71],\n",
+       "         [0.71, 0.71, 0.71, ..., 0.71, 0.71, 0.71],\n",
+       "         [0.7 , 0.7 , 0.7 , ..., 0.7 , 0.7 , 0.7 ]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.7 , 0.7 , 0.7 , ..., 0.7 , 0.7 , 0.7 ],\n",
+       "         [0.7 , 0.7 , 0.7 , ..., 0.7 , 0.7 , 0.7 ],\n",
+       "         [0.69, 0.69, 0.69, ..., 0.69, 0.69, 0.69]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "...\n",
+       "         [0.41, 0.41, 0.41, ..., 0.41, 0.41, 0.41],\n",
+       "         [0.41, 0.41, 0.41, ..., 0.41, 0.41, 0.41]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.42, 0.42, 0.42, ..., 0.42, 0.42, 0.42],\n",
+       "         [0.41, 0.41, 0.41, ..., 0.41, 0.41, 0.41],\n",
+       "         [0.41, 0.41, 0.41, ..., 0.41, 0.41, 0.41]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.41, 0.41, 0.41, ..., 0.42, 0.42, 0.42],\n",
+       "         [0.41, 0.41, 0.41, ..., 0.41, 0.41, 0.41],\n",
+       "         [0.41, 0.41, 0.41, ..., 0.41, 0.41, 0.41]]]], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>err</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan nan ... 0.3 0.3 0.3 0.3</div><input id='attrs-b2b1bf52-7937-462e-846c-8ddca8981cf7' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-b2b1bf52-7937-462e-846c-8ddca8981cf7' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-6dac2ff5-0e99-478c-97f2-9ed83a84cae0' class='xr-var-data-in' type='checkbox'><label for='data-6dac2ff5-0e99-478c-97f2-9ed83a84cae0' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Estimated error standard deviation of analysed_sst</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_min :</span></dt><dd>0</dd><dt><span>valid_max :</span></dt><dd>1000</dd></dl></div><div class='xr-var-data'><pre>array([[[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "...\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998]]]], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>ice</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan nan ... 0.9 0.9 0.9 0.9</div><input id='attrs-59a20e4a-d5b4-48b7-9e5d-1f610aac8d40' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-59a20e4a-d5b4-48b7-9e5d-1f610aac8d40' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-773ed873-733e-4f9b-a83a-fd5842839aef' class='xr-var-data-in' type='checkbox'><label for='data-773ed873-733e-4f9b-a83a-fd5842839aef' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea ice concentration</dd><dt><span>units :</span></dt><dd>%</dd><dt><span>valid_min :</span></dt><dd>0</dd><dt><span>valid_max :</span></dt><dd>100</dd></dl></div><div class='xr-var-data'><pre>array([[[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.82, 0.82, 0.82, ..., 0.82, 0.82, 0.82],\n",
+       "         [0.83, 0.83, 0.83, ..., 0.83, 0.83, 0.83],\n",
+       "         [0.83, 0.83, 0.83, ..., 0.83, 0.83, 0.83]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.82, 0.82, 0.82, ..., 0.82, 0.82, 0.82],\n",
+       "         [0.82, 0.82, 0.82, ..., 0.82, 0.82, 0.82],\n",
+       "         [0.83, 0.83, 0.83, ..., 0.83, 0.83, 0.83]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "...\n",
+       "         [0.9 , 0.9 , 0.9 , ..., 0.9 , 0.9 , 0.9 ],\n",
+       "         [0.9 , 0.9 , 0.9 , ..., 0.9 , 0.9 , 0.9 ]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.89, 0.89, 0.89, ..., 0.89, 0.89, 0.89],\n",
+       "         [0.89, 0.89, 0.89, ..., 0.89, 0.89, 0.89],\n",
+       "         [0.89, 0.89, 0.89, ..., 0.89, 0.89, 0.89]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.89, 0.89, 0.89, ..., 0.89, 0.89, 0.89],\n",
+       "         [0.9 , 0.9 , 0.9 , ..., 0.9 , 0.9 , 0.9 ],\n",
+       "         [0.9 , 0.9 , 0.9 , ..., 0.9 , 0.9 , 0.9 ]]]], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>sst</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan ... -1.39 -1.39 -1.39</div><input id='attrs-54a18600-ecb4-46df-b8cd-badee82a6fde' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-54a18600-ecb4-46df-b8cd-badee82a6fde' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-cccdbd13-bc84-4e5c-a280-f6c1d41b8646' class='xr-var-data-in' type='checkbox'><label for='data-cccdbd13-bc84-4e5c-a280-f6c1d41b8646' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_min :</span></dt><dd>-300</dd><dt><span>valid_max :</span></dt><dd>4500</dd></dl></div><div class='xr-var-data'><pre>array([[[[  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         ...,\n",
+       "         [-1.09, -1.09, -1.09, ..., -1.09, -1.09, -1.09],\n",
+       "         [-1.09, -1.09, -1.09, ..., -1.09, -1.09, -1.09],\n",
+       "         [-1.1 , -1.1 , -1.1 , ..., -1.1 , -1.1 , -1.1 ]]],\n",
+       "\n",
+       "\n",
+       "       [[[  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         ...,\n",
+       "         [-1.1 , -1.1 , -1.1 , ..., -1.1 , -1.1 , -1.1 ],\n",
+       "         [-1.1 , -1.1 , -1.1 , ..., -1.1 , -1.1 , -1.1 ],\n",
+       "         [-1.11, -1.11, -1.11, ..., -1.11, -1.11, -1.11]]],\n",
+       "\n",
+       "\n",
+       "       [[[  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "...\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.39, -1.39, -1.39],\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.39, -1.39, -1.39]]],\n",
+       "\n",
+       "\n",
+       "       [[[  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         ...,\n",
+       "         [-1.38, -1.38, -1.38, ..., -1.38, -1.38, -1.38],\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.39, -1.39, -1.39],\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.39, -1.39, -1.39]]],\n",
+       "\n",
+       "\n",
+       "       [[[  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         [  nan,   nan,   nan, ...,   nan,   nan,   nan],\n",
+       "         ...,\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.38, -1.38, -1.38],\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.39, -1.39, -1.39],\n",
+       "         [-1.39, -1.39, -1.39, ..., -1.39, -1.39, -1.39]]]], dtype=float32)</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-675a73ab-482c-4f1c-bf4f-1746959c9760' class='xr-section-summary-in' type='checkbox'  ><label for='section-675a73ab-482c-4f1c-bf4f-1746959c9760' class='xr-section-summary' >Attributes: <span>(37)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><dl class='xr-attrs'><dt><span>title :</span></dt><dd>NOAA/NCEI 1/4 Degree Daily Optimum Interpolation Sea Surface Temperature (OISST) Analysis, Version 2.1 - Final</dd><dt><span>source :</span></dt><dd>ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfinder_AVHRR, Navy_AVHRR</dd><dt><span>id :</span></dt><dd>oisst-avhrr-v02r01.19810901.nc</dd><dt><span>naming_authority :</span></dt><dd>gov.noaa.ncei</dd><dt><span>summary :</span></dt><dd>NOAAs 1/4-degree Daily Optimum Interpolation Sea Surface Temperature (OISST) (sometimes referred to as Reynolds SST, which however also refers to earlier products at different resolution), currently available as version v02r01, is created by interpolating and extrapolating SST observations from different sources, resulting in a smoothed complete field. The sources of data are satellite (AVHRR) and in situ platforms (i.e., ships and buoys), and the specific datasets employed may change over time. At the marginal ice zone, sea ice concentrations are used to generate proxy SSTs.  A preliminary version of this file is produced in near-real time (1-day latency), and then replaced with a final version after 2 weeks. Note that this is the AVHRR-ONLY DOISST, available from Oct 1981, but there is a companion DOISST product that includes microwave satellite data, available from June 2002</dd><dt><span>cdm_data_type :</span></dt><dd>Grid</dd><dt><span>history :</span></dt><dd>Final file created using preliminary as first guess, and 3 days of AVHRR data. Preliminary uses only 1 day of AVHRR data.</dd><dt><span>date_modified :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>date_created :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>product_version :</span></dt><dd>Version v02r01</dd><dt><span>processing_level :</span></dt><dd>NOAA Level 4</dd><dt><span>institution :</span></dt><dd>NOAA/National Centers for Environmental Information</dd><dt><span>creator_url :</span></dt><dd>https://www.ncei.noaa.gov/</dd><dt><span>creator_email :</span></dt><dd>oisst-help@noaa.gov</dd><dt><span>keywords :</span></dt><dd>Earth Science &gt; Oceans &gt; Ocean Temperature &gt; Sea Surface Temperature</dd><dt><span>keywords_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Earth Science Keywords</dd><dt><span>platform :</span></dt><dd>Ships, buoys, Argo floats, MetOp-A, MetOp-B</dd><dt><span>platform_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Platform Keywords</dd><dt><span>instrument :</span></dt><dd>Earth Remote Sensing Instruments &gt; Passive Remote Sensing &gt; Spectrometers/Radiometers &gt; Imaging Spectrometers/Radiometers &gt; AVHRR &gt; Advanced Very High Resolution Radiometer</dd><dt><span>instrument_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Instrument Keywords</dd><dt><span>standard_name_vocabulary :</span></dt><dd>CF Standard Name Table (v40, 25 January 2017)</dd><dt><span>geospatial_lat_min :</span></dt><dd>-90.0</dd><dt><span>geospatial_lat_max :</span></dt><dd>90.0</dd><dt><span>geospatial_lon_min :</span></dt><dd>0.0</dd><dt><span>geospatial_lon_max :</span></dt><dd>360.0</dd><dt><span>geospatial_lat_units :</span></dt><dd>degrees_north</dd><dt><span>geospatial_lat_resolution :</span></dt><dd>0.25</dd><dt><span>geospatial_lon_units :</span></dt><dd>degrees_east</dd><dt><span>geospatial_lon_resolution :</span></dt><dd>0.25</dd><dt><span>time_coverage_start :</span></dt><dd>1981-09-01T00:00:00Z</dd><dt><span>time_coverage_end :</span></dt><dd>1981-09-01T23:59:59Z</dd><dt><span>metadata_link :</span></dt><dd>https://doi.org/10.25921/RE9P-PT57</dd><dt><span>ncei_template_version :</span></dt><dd>NCEI_NetCDF_Grid_Template_v2.0</dd><dt><span>comment :</span></dt><dd>Data was converted from NetCDF-3 to NetCDF-4 format with metadata updates in November 2017.</dd><dt><span>sensor :</span></dt><dd>Thermometer, AVHRR</dd><dt><span>Conventions :</span></dt><dd>CF-1.6, ACDD-1.3</dd><dt><span>references :</span></dt><dd>Reynolds, et al.(2007) Daily High-Resolution-Blended Analyses for Sea Surface Temperature (available at https://doi.org/10.1175/2007JCLI1824.1). Banzon, et al.(2016) A long-term record of blended satellite and in situ sea-surface temperature for climate monitoring, modeling and environmental studies (available at https://doi.org/10.5194/essd-8-165-2016). Huang et al. (2020) Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version v02r01, submitted.Climatology is based on 1971-2000 OI.v2 SST. Satellite data: Pathfinder AVHRR SST and Navy AVHRR SST. Ice data: NCEP Ice and GSFC Ice.</dd></dl></div></li></ul></div></div>"
+      ],
+      "text/plain": [
+       "<xarray.Dataset>\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 20, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 1981-09-01T12:00:00 ... 1981-09-20T12:00:00\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 nan nan nan nan ... 0.41 0.41 0.41\n",
+       "    err      (time, zlev, lat, lon) float32 nan nan nan nan ... 0.3 0.3 0.3 0.3\n",
+       "    ice      (time, zlev, lat, lon) float32 nan nan nan nan ... 0.9 0.9 0.9 0.9\n",
+       "    sst      (time, zlev, lat, lon) float32 nan nan nan ... -1.39 -1.39 -1.39\n",
+       "Attributes:\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    id:                         oisst-avhrr-v02r01.19810901.nc\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    cdm_data_type:              Grid\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    product_version:            Version v02r01\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    keywords:                   Earth Science > Oceans > Ocean Temperature > ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments > Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    time_coverage_start:        1981-09-01T00:00:00Z\n",
+       "    time_coverage_end:          1981-09-01T23:59:59Z\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-..."
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ds_chunk = recipe.open_chunk(all_chunks[0])\n",
+    "ds_chunk"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Total chunk size: 331.784804 MB\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(f'Total chunk size: {ds_chunk.nbytes / 1e6} MB')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "👀 Inspect the Xarray HTML repr carefully by clicking on the buttons to expand the different sections.\n",
+    "- ✅ Is the shape of the variable what we expect\n",
+    "- ✅ Is `time` in going the right order\n",
+    "- ✅ Do the variable attributes make sense?\n",
+    "\n",
+    "\n",
+    "Now let's visualize some data make sure things look"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.collections.QuadMesh at 0x7ff8a047c100>"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ds_chunk.sst[0].plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.collections.QuadMesh at 0x7ff882c18520>"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ds_chunk.ice[-1].plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The data look good!\n",
+    "Now let's try a random chunk from the middle."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "chunk_number = 500\n",
+    "for input_file in recipe.inputs_for_chunk(chunk_number):\n",
+    "    recipe.cache_input(input_file)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div><svg style=\"position: absolute; width: 0; height: 0; overflow: hidden\">\n",
+       "<defs>\n",
+       "<symbol id=\"icon-database\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M16 0c-8.837 0-16 2.239-16 5v4c0 2.761 7.163 5 16 5s16-2.239 16-5v-4c0-2.761-7.163-5-16-5z\"></path>\n",
+       "<path d=\"M16 17c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "<path d=\"M16 26c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "</symbol>\n",
+       "<symbol id=\"icon-file-text2\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M28.681 7.159c-0.694-0.947-1.662-2.053-2.724-3.116s-2.169-2.030-3.116-2.724c-1.612-1.182-2.393-1.319-2.841-1.319h-15.5c-1.378 0-2.5 1.121-2.5 2.5v27c0 1.378 1.122 2.5 2.5 2.5h23c1.378 0 2.5-1.122 2.5-2.5v-19.5c0-0.448-0.137-1.23-1.319-2.841zM24.543 5.457c0.959 0.959 1.712 1.825 2.268 2.543h-4.811v-4.811c0.718 0.556 1.584 1.309 2.543 2.268zM28 29.5c0 0.271-0.229 0.5-0.5 0.5h-23c-0.271 0-0.5-0.229-0.5-0.5v-27c0-0.271 0.229-0.5 0.5-0.5 0 0 15.499-0 15.5 0v7c0 0.552 0.448 1 1 1h7v19.5z\"></path>\n",
+       "<path d=\"M23 26h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 22h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 18h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "</symbol>\n",
+       "</defs>\n",
+       "</svg>\n",
+       "<style>/* CSS stylesheet for displaying xarray objects in jupyterlab.\n",
+       " *\n",
+       " */\n",
+       "\n",
+       ":root {\n",
+       "  --xr-font-color0: var(--jp-content-font-color0, rgba(0, 0, 0, 1));\n",
+       "  --xr-font-color2: var(--jp-content-font-color2, rgba(0, 0, 0, 0.54));\n",
+       "  --xr-font-color3: var(--jp-content-font-color3, rgba(0, 0, 0, 0.38));\n",
+       "  --xr-border-color: var(--jp-border-color2, #e0e0e0);\n",
+       "  --xr-disabled-color: var(--jp-layout-color3, #bdbdbd);\n",
+       "  --xr-background-color: var(--jp-layout-color0, white);\n",
+       "  --xr-background-color-row-even: var(--jp-layout-color1, white);\n",
+       "  --xr-background-color-row-odd: var(--jp-layout-color2, #eeeeee);\n",
+       "}\n",
+       "\n",
+       "html[theme=dark],\n",
+       "body.vscode-dark {\n",
+       "  --xr-font-color0: rgba(255, 255, 255, 1);\n",
+       "  --xr-font-color2: rgba(255, 255, 255, 0.54);\n",
+       "  --xr-font-color3: rgba(255, 255, 255, 0.38);\n",
+       "  --xr-border-color: #1F1F1F;\n",
+       "  --xr-disabled-color: #515151;\n",
+       "  --xr-background-color: #111111;\n",
+       "  --xr-background-color-row-even: #111111;\n",
+       "  --xr-background-color-row-odd: #313131;\n",
+       "}\n",
+       "\n",
+       ".xr-wrap {\n",
+       "  display: block;\n",
+       "  min-width: 300px;\n",
+       "  max-width: 700px;\n",
+       "}\n",
+       "\n",
+       ".xr-text-repr-fallback {\n",
+       "  /* fallback to plain text repr when CSS is not injected (untrusted notebook) */\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-header {\n",
+       "  padding-top: 6px;\n",
+       "  padding-bottom: 6px;\n",
+       "  margin-bottom: 4px;\n",
+       "  border-bottom: solid 1px var(--xr-border-color);\n",
+       "}\n",
+       "\n",
+       ".xr-header > div,\n",
+       ".xr-header > ul {\n",
+       "  display: inline;\n",
+       "  margin-top: 0;\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type,\n",
+       ".xr-array-name {\n",
+       "  margin-left: 2px;\n",
+       "  margin-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-sections {\n",
+       "  padding-left: 0 !important;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 150px auto auto 1fr 20px 20px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input + label {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label {\n",
+       "  cursor: pointer;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label:hover {\n",
+       "  color: var(--xr-font-color0);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary {\n",
+       "  grid-column: 1;\n",
+       "  color: var(--xr-font-color2);\n",
+       "  font-weight: 500;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary > span {\n",
+       "  display: inline-block;\n",
+       "  padding-left: 0.5em;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in + label:before {\n",
+       "  display: inline-block;\n",
+       "  content: '►';\n",
+       "  font-size: 11px;\n",
+       "  width: 15px;\n",
+       "  text-align: center;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label:before {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label:before {\n",
+       "  content: '▼';\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label > span {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary,\n",
+       ".xr-section-inline-details {\n",
+       "  padding-top: 4px;\n",
+       "  padding-bottom: 4px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-inline-details {\n",
+       "  grid-column: 2 / -1;\n",
+       "}\n",
+       "\n",
+       ".xr-section-details {\n",
+       "  display: none;\n",
+       "  grid-column: 1 / -1;\n",
+       "  margin-bottom: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked ~ .xr-section-details {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap {\n",
+       "  grid-column: 1 / -1;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 20px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap > label {\n",
+       "  grid-column: 1;\n",
+       "  vertical-align: top;\n",
+       "}\n",
+       "\n",
+       ".xr-preview {\n",
+       "  color: var(--xr-font-color3);\n",
+       "}\n",
+       "\n",
+       ".xr-array-preview,\n",
+       ".xr-array-data {\n",
+       "  padding: 0 5px !important;\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-array-data,\n",
+       ".xr-array-in:checked ~ .xr-array-preview {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-array-in:checked ~ .xr-array-data,\n",
+       ".xr-array-preview {\n",
+       "  display: inline-block;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list {\n",
+       "  display: inline-block !important;\n",
+       "  list-style: none;\n",
+       "  padding: 0 !important;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li {\n",
+       "  display: inline-block;\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:before {\n",
+       "  content: '(';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:after {\n",
+       "  content: ')';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li:not(:last-child):after {\n",
+       "  content: ',';\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-has-index {\n",
+       "  font-weight: bold;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list,\n",
+       ".xr-var-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > div,\n",
+       ".xr-var-item label,\n",
+       ".xr-var-item > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-even);\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > .xr-var-name:hover span {\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list > li:nth-child(odd) > div,\n",
+       ".xr-var-list > li:nth-child(odd) > label,\n",
+       ".xr-var-list > li:nth-child(odd) > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-odd);\n",
+       "}\n",
+       "\n",
+       ".xr-var-name {\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dims {\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dtype {\n",
+       "  grid-column: 3;\n",
+       "  text-align: right;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-var-preview {\n",
+       "  grid-column: 4;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name,\n",
+       ".xr-var-dims,\n",
+       ".xr-var-dtype,\n",
+       ".xr-preview,\n",
+       ".xr-attrs dt {\n",
+       "  white-space: nowrap;\n",
+       "  overflow: hidden;\n",
+       "  text-overflow: ellipsis;\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name:hover,\n",
+       ".xr-var-dims:hover,\n",
+       ".xr-var-dtype:hover,\n",
+       ".xr-attrs dt:hover {\n",
+       "  overflow: visible;\n",
+       "  width: auto;\n",
+       "  z-index: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  display: none;\n",
+       "  background-color: var(--xr-background-color) !important;\n",
+       "  padding-bottom: 5px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs-in:checked ~ .xr-var-attrs,\n",
+       ".xr-var-data-in:checked ~ .xr-var-data {\n",
+       "  display: block;\n",
+       "}\n",
+       "\n",
+       ".xr-var-data > table {\n",
+       "  float: right;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name span,\n",
+       ".xr-var-data,\n",
+       ".xr-attrs {\n",
+       "  padding-left: 25px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs,\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  grid-column: 1 / -1;\n",
+       "}\n",
+       "\n",
+       "dl.xr-attrs {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 125px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt,\n",
+       ".xr-attrs dd {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  float: left;\n",
+       "  padding-right: 10px;\n",
+       "  width: auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt {\n",
+       "  font-weight: normal;\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt:hover span {\n",
+       "  display: inline-block;\n",
+       "  background: var(--xr-background-color);\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dd {\n",
+       "  grid-column: 2;\n",
+       "  white-space: pre-wrap;\n",
+       "  word-break: break-all;\n",
+       "}\n",
+       "\n",
+       ".xr-icon-database,\n",
+       ".xr-icon-file-text2 {\n",
+       "  display: inline-block;\n",
+       "  vertical-align: middle;\n",
+       "  width: 1em;\n",
+       "  height: 1.5em !important;\n",
+       "  stroke-width: 0;\n",
+       "  stroke: currentColor;\n",
+       "  fill: currentColor;\n",
+       "}\n",
+       "</style><pre class='xr-text-repr-fallback'>&lt;xarray.Dataset&gt;\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 20, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 2009-01-17T12:00:00 ... 2009-02-05T12:00:00\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 nan nan nan nan ... 0.08 0.08 0.08\n",
+       "    err      (time, zlev, lat, lon) float32 nan nan nan nan ... 0.3 0.3 0.3 0.3\n",
+       "    ice      (time, zlev, lat, lon) float32 nan nan nan nan ... 1.0 1.0 1.0 1.0\n",
+       "    sst      (time, zlev, lat, lon) float32 nan nan nan ... -1.72 -1.72 -1.72\n",
+       "Attributes:\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...\n",
+       "    Description:                Reynolds, et al.(2007) Daily High-resolution ...\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    id:                         oisst-avhrr-v02r01.20090117.nc\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    cdm_data_type:              Grid\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    product_version:            Version v02r01\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    keywords:                   Earth Science &gt; Oceans &gt; Ocean Temperature &gt; ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments &gt; Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    time_coverage_start:        2009-01-17T00:00:00Z\n",
+       "    time_coverage_end:          2009-01-17T23:59:59Z\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-...</pre><div class='xr-wrap' hidden><div class='xr-header'><div class='xr-obj-type'>xarray.Dataset</div></div><ul class='xr-sections'><li class='xr-section-item'><input id='section-47ab2480-1754-47fd-b594-eff995c32ad5' class='xr-section-summary-in' type='checkbox' disabled ><label for='section-47ab2480-1754-47fd-b594-eff995c32ad5' class='xr-section-summary'  title='Expand/collapse section'>Dimensions:</label><div class='xr-section-inline-details'><ul class='xr-dim-list'><li><span class='xr-has-index'>lat</span>: 720</li><li><span class='xr-has-index'>lon</span>: 1440</li><li><span class='xr-has-index'>time</span>: 20</li><li><span class='xr-has-index'>zlev</span>: 1</li></ul></div><div class='xr-section-details'></div></li><li class='xr-section-item'><input id='section-ae3170c5-e4ed-4292-92e8-0974cb7cd815' class='xr-section-summary-in' type='checkbox'  checked><label for='section-ae3170c5-e4ed-4292-92e8-0974cb7cd815' class='xr-section-summary' >Coordinates: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lat</span></div><div class='xr-var-dims'>(lat)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>-89.88 -89.62 ... 89.62 89.88</div><input id='attrs-d2aec2da-dd2f-4b00-8343-3e3384050938' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-d2aec2da-dd2f-4b00-8343-3e3384050938' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-41a305b1-2f17-40ab-b7f3-0c2197d1212c' class='xr-var-data-in' type='checkbox'><label for='data-41a305b1-2f17-40ab-b7f3-0c2197d1212c' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Latitude</dd><dt><span>units :</span></dt><dd>degrees_north</dd><dt><span>grids :</span></dt><dd>Uniform grid from -89.875 to 89.875 by 0.25</dd></dl></div><div class='xr-var-data'><pre>array([-89.875, -89.625, -89.375, ...,  89.375,  89.625,  89.875],\n",
+       "      dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lon</span></div><div class='xr-var-dims'>(lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.125 0.375 0.625 ... 359.6 359.9</div><input id='attrs-519fc94e-45e3-4e75-a6e1-434298330c18' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-519fc94e-45e3-4e75-a6e1-434298330c18' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-48ce7876-f86a-4958-a296-6c89edc0dfa5' class='xr-var-data-in' type='checkbox'><label for='data-48ce7876-f86a-4958-a296-6c89edc0dfa5' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Longitude</dd><dt><span>units :</span></dt><dd>degrees_east</dd><dt><span>grids :</span></dt><dd>Uniform grid from 0.125 to 359.875 by 0.25</dd></dl></div><div class='xr-var-data'><pre>array([1.25000e-01, 3.75000e-01, 6.25000e-01, ..., 3.59375e+02, 3.59625e+02,\n",
+       "       3.59875e+02], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>time</span></div><div class='xr-var-dims'>(time)</div><div class='xr-var-dtype'>datetime64[ns]</div><div class='xr-var-preview xr-preview'>2009-01-17T12:00:00 ... 2009-02-...</div><input id='attrs-dd725f60-e310-4e04-9541-c5242b828144' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-dd725f60-e310-4e04-9541-c5242b828144' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-59efa1d2-cb36-410e-b97b-43628ec71811' class='xr-var-data-in' type='checkbox'><label for='data-59efa1d2-cb36-410e-b97b-43628ec71811' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Center time of the day</dd></dl></div><div class='xr-var-data'><pre>array([&#x27;2009-01-17T12:00:00.000000000&#x27;, &#x27;2009-01-18T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-19T12:00:00.000000000&#x27;, &#x27;2009-01-20T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-21T12:00:00.000000000&#x27;, &#x27;2009-01-22T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-23T12:00:00.000000000&#x27;, &#x27;2009-01-24T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-25T12:00:00.000000000&#x27;, &#x27;2009-01-26T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-27T12:00:00.000000000&#x27;, &#x27;2009-01-28T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-29T12:00:00.000000000&#x27;, &#x27;2009-01-30T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-01-31T12:00:00.000000000&#x27;, &#x27;2009-02-01T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-02-02T12:00:00.000000000&#x27;, &#x27;2009-02-03T12:00:00.000000000&#x27;,\n",
+       "       &#x27;2009-02-04T12:00:00.000000000&#x27;, &#x27;2009-02-05T12:00:00.000000000&#x27;],\n",
+       "      dtype=&#x27;datetime64[ns]&#x27;)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>zlev</span></div><div class='xr-var-dims'>(zlev)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.0</div><input id='attrs-d53d68ab-2f89-40d1-bca7-20e23a760eb8' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-d53d68ab-2f89-40d1-bca7-20e23a760eb8' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-2165c3bd-8cee-402f-8efb-f4da0f1208a1' class='xr-var-data-in' type='checkbox'><label for='data-2165c3bd-8cee-402f-8efb-f4da0f1208a1' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea surface height</dd><dt><span>units :</span></dt><dd>meters</dd><dt><span>actual_range :</span></dt><dd>0, 0</dd><dt><span>positive :</span></dt><dd>down</dd></dl></div><div class='xr-var-data'><pre>array([0.], dtype=float32)</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-0a5ad931-5ecc-4e48-9a3c-9ec5a25d0908' class='xr-section-summary-in' type='checkbox'  checked><label for='section-0a5ad931-5ecc-4e48-9a3c-9ec5a25d0908' class='xr-section-summary' >Data variables: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span>anom</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan nan ... 0.08 0.08 0.08</div><input id='attrs-863c3f8b-4735-4983-ac68-3e036661ce5a' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-863c3f8b-4735-4983-ac68-3e036661ce5a' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-181a408d-4c5c-4940-a0a6-4e610c377bcc' class='xr-var-data-in' type='checkbox'><label for='data-181a408d-4c5c-4940-a0a6-4e610c377bcc' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature anomalies</dd><dt><span>valid_min :</span></dt><dd>-1200</dd><dt><span>valid_max :</span></dt><dd>1200</dd><dt><span>units :</span></dt><dd>Celsius</dd></dl></div><div class='xr-var-data'><pre>array([[[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.06, 0.06, 0.06, ..., 0.06, 0.06, 0.06],\n",
+       "         [0.06, 0.06, 0.06, ..., 0.06, 0.06, 0.06],\n",
+       "         [0.06, 0.06, 0.06, ..., 0.06, 0.06, 0.06]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.06, 0.06, 0.06, ..., 0.06, 0.06, 0.06],\n",
+       "         [0.06, 0.06, 0.06, ..., 0.06, 0.06, 0.06],\n",
+       "         [0.06, 0.06, 0.06, ..., 0.06, 0.06, 0.06]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "...\n",
+       "         [0.07, 0.07, 0.07, ..., 0.07, 0.07, 0.07],\n",
+       "         [0.07, 0.07, 0.07, ..., 0.07, 0.07, 0.07]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.08, 0.08, 0.08, ..., 0.08, 0.08, 0.08],\n",
+       "         [0.08, 0.08, 0.08, ..., 0.08, 0.08, 0.08],\n",
+       "         [0.08, 0.08, 0.08, ..., 0.08, 0.08, 0.08]]],\n",
+       "\n",
+       "\n",
+       "       [[[ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         [ nan,  nan,  nan, ...,  nan,  nan,  nan],\n",
+       "         ...,\n",
+       "         [0.08, 0.08, 0.08, ..., 0.08, 0.08, 0.08],\n",
+       "         [0.08, 0.08, 0.08, ..., 0.08, 0.08, 0.08],\n",
+       "         [0.08, 0.08, 0.08, ..., 0.08, 0.08, 0.08]]]], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>err</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan nan ... 0.3 0.3 0.3 0.3</div><input id='attrs-ca40e1f8-fce6-4312-bc07-05c0eaa5092c' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-ca40e1f8-fce6-4312-bc07-05c0eaa5092c' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-5a8a2983-2b83-45b3-ba58-7c11f211a257' class='xr-var-data-in' type='checkbox'><label for='data-5a8a2983-2b83-45b3-ba58-7c11f211a257' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Estimated error standard deviation of analysed_sst</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_min :</span></dt><dd>0</dd><dt><span>valid_max :</span></dt><dd>1000</dd></dl></div><div class='xr-var-data'><pre>array([[[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "...\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998],\n",
+       "         [0.29999998, 0.29999998, 0.29999998, ..., 0.29999998,\n",
+       "          0.29999998, 0.29999998]]]], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>ice</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan nan ... 1.0 1.0 1.0 1.0</div><input id='attrs-3aab3988-0599-4d5f-b735-600a3795a918' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-3aab3988-0599-4d5f-b735-600a3795a918' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-6cc1b0fe-a70c-4a11-84ac-ed16ddc05689' class='xr-var-data-in' type='checkbox'><label for='data-6cc1b0fe-a70c-4a11-84ac-ed16ddc05689' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea ice concentration</dd><dt><span>units :</span></dt><dd>%</dd><dt><span>valid_min :</span></dt><dd>0</dd><dt><span>valid_max :</span></dt><dd>100</dd></dl></div><div class='xr-var-data'><pre>array([[[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [1.        , 1.        , 1.        , ..., 1.        ,\n",
+       "          1.        , 1.        ],\n",
+       "         [0.98999995, 0.98999995, 0.98999995, ..., 0.98999995,\n",
+       "          0.98999995, 0.98999995],\n",
+       "         [0.98999995, 0.98999995, 0.98999995, ..., 0.98999995,\n",
+       "          0.98999995, 0.98999995]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "...\n",
+       "          1.        , 1.        ],\n",
+       "         [1.        , 1.        , 1.        , ..., 1.        ,\n",
+       "          1.        , 1.        ],\n",
+       "         [1.        , 1.        , 1.        , ..., 1.        ,\n",
+       "          1.        , 1.        ]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [1.        , 1.        , 1.        , ..., 1.        ,\n",
+       "          1.        , 1.        ],\n",
+       "         [1.        , 1.        , 1.        , ..., 1.        ,\n",
+       "          1.        , 1.        ],\n",
+       "         [1.        , 1.        , 1.        , ..., 1.        ,\n",
+       "          1.        , 1.        ]]]], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span>sst</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>nan nan nan ... -1.72 -1.72 -1.72</div><input id='attrs-a2f2bf91-c9f0-4dde-89a0-902d451c8e21' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-a2f2bf91-c9f0-4dde-89a0-902d451c8e21' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-e5c430eb-35ab-4b4d-a411-dcd0bb942603' class='xr-var-data-in' type='checkbox'><label for='data-e5c430eb-35ab-4b4d-a411-dcd0bb942603' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_min :</span></dt><dd>-300</dd><dt><span>valid_max :</span></dt><dd>4500</dd></dl></div><div class='xr-var-data'><pre>array([[[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [-1.74     , -1.74     , -1.74     , ..., -1.74     ,\n",
+       "          -1.74     , -1.74     ],\n",
+       "         [-1.74     , -1.74     , -1.74     , ..., -1.74     ,\n",
+       "          -1.74     , -1.74     ],\n",
+       "         [-1.74     , -1.74     , -1.74     , ..., -1.74     ,\n",
+       "          -1.74     , -1.74     ]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "...\n",
+       "          -1.7199999, -1.7199999],\n",
+       "         [-1.7199999, -1.7199999, -1.7199999, ..., -1.7199999,\n",
+       "          -1.7199999, -1.7199999],\n",
+       "         [-1.7199999, -1.7199999, -1.7199999, ..., -1.7199999,\n",
+       "          -1.7199999, -1.7199999]]],\n",
+       "\n",
+       "\n",
+       "       [[[       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         [       nan,        nan,        nan, ...,        nan,\n",
+       "                 nan,        nan],\n",
+       "         ...,\n",
+       "         [-1.7199999, -1.7199999, -1.7199999, ..., -1.7199999,\n",
+       "          -1.7199999, -1.7199999],\n",
+       "         [-1.7199999, -1.7199999, -1.7199999, ..., -1.7199999,\n",
+       "          -1.7199999, -1.7199999],\n",
+       "         [-1.7199999, -1.7199999, -1.7199999, ..., -1.7199999,\n",
+       "          -1.7199999, -1.7199999]]]], dtype=float32)</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-a49d746f-6e85-4921-a0d7-23ba40d6d7b6' class='xr-section-summary-in' type='checkbox'  ><label for='section-a49d746f-6e85-4921-a0d7-23ba40d6d7b6' class='xr-section-summary' >Attributes: <span>(38)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><dl class='xr-attrs'><dt><span>title :</span></dt><dd>NOAA/NCEI 1/4 Degree Daily Optimum Interpolation Sea Surface Temperature (OISST) Analysis, Version 2.1 - Final</dd><dt><span>Description :</span></dt><dd>Reynolds, et al.(2007) Daily High-resolution Blended Analyses. Available at ftp://eclipse.ncdc.noaa.gov/pub/OI-daily/daily-sst.pdf  Climatology is based on 1971-2000 OI.v2 SST, Satellite data: Navy  NOAA18 METOP AVHRR, Ice data: NCEP ice</dd><dt><span>source :</span></dt><dd>ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfinder_AVHRR, Navy_AVHRR</dd><dt><span>id :</span></dt><dd>oisst-avhrr-v02r01.20090117.nc</dd><dt><span>naming_authority :</span></dt><dd>gov.noaa.ncei</dd><dt><span>summary :</span></dt><dd>NOAAs 1/4-degree Daily Optimum Interpolation Sea Surface Temperature (OISST) (sometimes referred to as Reynolds SST, which however also refers to earlier products at different resolution), currently available as version v02r01, is created by interpolating and extrapolating SST observations from different sources, resulting in a smoothed complete field. The sources of data are satellite (AVHRR) and in situ platforms (i.e., ships and buoys), and the specific datasets employed may change over time. At the marginal ice zone, sea ice concentrations are used to generate proxy SSTs.  A preliminary version of this file is produced in near-real time (1-day latency), and then replaced with a final version after 2 weeks. Note that this is the AVHRR-ONLY DOISST, available from Oct 1981, but there is a companion DOISST product that includes microwave satellite data, available from June 2002</dd><dt><span>cdm_data_type :</span></dt><dd>Grid</dd><dt><span>history :</span></dt><dd>Final file created using preliminary as first guess, and 3 days of AVHRR data. Preliminary uses only 1 day of AVHRR data.</dd><dt><span>date_modified :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>date_created :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>product_version :</span></dt><dd>Version v02r01</dd><dt><span>processing_level :</span></dt><dd>NOAA Level 4</dd><dt><span>institution :</span></dt><dd>NOAA/National Centers for Environmental Information</dd><dt><span>creator_url :</span></dt><dd>https://www.ncei.noaa.gov/</dd><dt><span>creator_email :</span></dt><dd>oisst-help@noaa.gov</dd><dt><span>keywords :</span></dt><dd>Earth Science &gt; Oceans &gt; Ocean Temperature &gt; Sea Surface Temperature</dd><dt><span>keywords_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Earth Science Keywords</dd><dt><span>platform :</span></dt><dd>Ships, buoys, Argo floats, MetOp-A, MetOp-B</dd><dt><span>platform_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Platform Keywords</dd><dt><span>instrument :</span></dt><dd>Earth Remote Sensing Instruments &gt; Passive Remote Sensing &gt; Spectrometers/Radiometers &gt; Imaging Spectrometers/Radiometers &gt; AVHRR &gt; Advanced Very High Resolution Radiometer</dd><dt><span>instrument_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Instrument Keywords</dd><dt><span>standard_name_vocabulary :</span></dt><dd>CF Standard Name Table (v40, 25 January 2017)</dd><dt><span>geospatial_lat_min :</span></dt><dd>-90.0</dd><dt><span>geospatial_lat_max :</span></dt><dd>90.0</dd><dt><span>geospatial_lon_min :</span></dt><dd>0.0</dd><dt><span>geospatial_lon_max :</span></dt><dd>360.0</dd><dt><span>geospatial_lat_units :</span></dt><dd>degrees_north</dd><dt><span>geospatial_lat_resolution :</span></dt><dd>0.25</dd><dt><span>geospatial_lon_units :</span></dt><dd>degrees_east</dd><dt><span>geospatial_lon_resolution :</span></dt><dd>0.25</dd><dt><span>time_coverage_start :</span></dt><dd>2009-01-17T00:00:00Z</dd><dt><span>time_coverage_end :</span></dt><dd>2009-01-17T23:59:59Z</dd><dt><span>metadata_link :</span></dt><dd>https://doi.org/10.25921/RE9P-PT57</dd><dt><span>ncei_template_version :</span></dt><dd>NCEI_NetCDF_Grid_Template_v2.0</dd><dt><span>comment :</span></dt><dd>Data was converted from NetCDF-3 to NetCDF-4 format with metadata updates in November 2017.</dd><dt><span>sensor :</span></dt><dd>Thermometer, AVHRR</dd><dt><span>Conventions :</span></dt><dd>CF-1.6, ACDD-1.3</dd><dt><span>references :</span></dt><dd>Reynolds, et al.(2007) Daily High-Resolution-Blended Analyses for Sea Surface Temperature (available at https://doi.org/10.1175/2007JCLI1824.1). Banzon, et al.(2016) A long-term record of blended satellite and in situ sea-surface temperature for climate monitoring, modeling and environmental studies (available at https://doi.org/10.5194/essd-8-165-2016). Huang et al. (2020) Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version v02r01, submitted.Climatology is based on 1971-2000 OI.v2 SST. Satellite data: Pathfinder AVHRR SST and Navy AVHRR SST. Ice data: NCEP Ice and GSFC Ice.</dd></dl></div></li></ul></div></div>"
+      ],
+      "text/plain": [
+       "<xarray.Dataset>\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 20, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 2009-01-17T12:00:00 ... 2009-02-05T12:00:00\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 nan nan nan nan ... 0.08 0.08 0.08\n",
+       "    err      (time, zlev, lat, lon) float32 nan nan nan nan ... 0.3 0.3 0.3 0.3\n",
+       "    ice      (time, zlev, lat, lon) float32 nan nan nan nan ... 1.0 1.0 1.0 1.0\n",
+       "    sst      (time, zlev, lat, lon) float32 nan nan nan ... -1.72 -1.72 -1.72\n",
+       "Attributes:\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...\n",
+       "    Description:                Reynolds, et al.(2007) Daily High-resolution ...\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    id:                         oisst-avhrr-v02r01.20090117.nc\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    cdm_data_type:              Grid\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    product_version:            Version v02r01\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    keywords:                   Earth Science > Oceans > Ocean Temperature > ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments > Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    time_coverage_start:        2009-01-17T00:00:00Z\n",
+       "    time_coverage_end:          2009-01-17T23:59:59Z\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-..."
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ds_chunk = recipe.open_chunk(chunk_number)\n",
+    "ds_chunk"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Step 7: Try writing data\n",
+    "\n",
+    "Now that we can see our chunks opening correctly, we are ready to try writing data to our target.\n",
+    "\n",
+    "First we need to prepare the target."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "recipe.prepare_target()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We should now see a Zarr group at the target location.\n",
+    "Only the coordinates have been written, not the data variables."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "/\n",
+      " ├── anom (14372, 1, 720, 1440) float32\n",
+      " ├── err (14372, 1, 720, 1440) float32\n",
+      " ├── ice (14372, 1, 720, 1440) float32\n",
+      " ├── lat (720,) float32\n",
+      " ├── lon (1440,) float32\n",
+      " ├── sst (14372, 1, 720, 1440) float32\n",
+      " ├── time (14372,) int64\n",
+      " └── zlev (1,) float32\n"
+     ]
+    }
+   ],
+   "source": [
+    "import zarr\n",
+    "zgroup = zarr.open(target_dir.name)\n",
+    "print(zgroup.tree())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let's examine one of the data variables."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"zarr-info\"><tbody><tr><th style=\"text-align: left\">Name</th><td style=\"text-align: left\">/sst</td></tr><tr><th style=\"text-align: left\">Type</th><td style=\"text-align: left\">zarr.core.Array</td></tr><tr><th style=\"text-align: left\">Data type</th><td style=\"text-align: left\">float32</td></tr><tr><th style=\"text-align: left\">Shape</th><td style=\"text-align: left\">(14372, 1, 720, 1440)</td></tr><tr><th style=\"text-align: left\">Chunk shape</th><td style=\"text-align: left\">(20, 1, 720, 1440)</td></tr><tr><th style=\"text-align: left\">Order</th><td style=\"text-align: left\">C</td></tr><tr><th style=\"text-align: left\">Read-only</th><td style=\"text-align: left\">False</td></tr><tr><th style=\"text-align: left\">Compressor</th><td style=\"text-align: left\">Blosc(cname='lz4', clevel=5, shuffle=SHUFFLE, blocksize=0)</td></tr><tr><th style=\"text-align: left\">Store type</th><td style=\"text-align: left\">zarr.storage.DirectoryStore</td></tr><tr><th style=\"text-align: left\">No. bytes</th><td style=\"text-align: left\">59603558400 (55.5G)</td></tr><tr><th style=\"text-align: left\">No. bytes stored</th><td style=\"text-align: left\">611</td></tr><tr><th style=\"text-align: left\">Storage ratio</th><td style=\"text-align: left\">97550832.1</td></tr><tr><th style=\"text-align: left\">Chunks initialized</th><td style=\"text-align: left\">0/719</td></tr></tbody></table>"
+      ],
+      "text/plain": [
+       "Name               : /sst\n",
+       "Type               : zarr.core.Array\n",
+       "Data type          : float32\n",
+       "Shape              : (14372, 1, 720, 1440)\n",
+       "Chunk shape        : (20, 1, 720, 1440)\n",
+       "Order              : C\n",
+       "Read-only          : False\n",
+       "Compressor         : Blosc(cname='lz4', clevel=5, shuffle=SHUFFLE, blocksize=0)\n",
+       "Store type         : zarr.storage.DirectoryStore\n",
+       "No. bytes          : 59603558400 (55.5G)\n",
+       "No. bytes stored   : 611\n",
+       "Storage ratio      : 97550832.1\n",
+       "Chunks initialized : 0/719"
+      ]
+     },
+     "execution_count": 25,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "zgroup['sst'].info"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now let's write the first chunk."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"zarr-info\"><tbody><tr><th style=\"text-align: left\">Name</th><td style=\"text-align: left\">/sst</td></tr><tr><th style=\"text-align: left\">Type</th><td style=\"text-align: left\">zarr.core.Array</td></tr><tr><th style=\"text-align: left\">Data type</th><td style=\"text-align: left\">float32</td></tr><tr><th style=\"text-align: left\">Shape</th><td style=\"text-align: left\">(14372, 1, 720, 1440)</td></tr><tr><th style=\"text-align: left\">Chunk shape</th><td style=\"text-align: left\">(20, 1, 720, 1440)</td></tr><tr><th style=\"text-align: left\">Order</th><td style=\"text-align: left\">C</td></tr><tr><th style=\"text-align: left\">Read-only</th><td style=\"text-align: left\">False</td></tr><tr><th style=\"text-align: left\">Compressor</th><td style=\"text-align: left\">Blosc(cname='lz4', clevel=5, shuffle=SHUFFLE, blocksize=0)</td></tr><tr><th style=\"text-align: left\">Store type</th><td style=\"text-align: left\">zarr.storage.DirectoryStore</td></tr><tr><th style=\"text-align: left\">No. bytes</th><td style=\"text-align: left\">59603558400 (55.5G)</td></tr><tr><th style=\"text-align: left\">No. bytes stored</th><td style=\"text-align: left\">39402334 (37.6M)</td></tr><tr><th style=\"text-align: left\">Storage ratio</th><td style=\"text-align: left\">1512.7</td></tr><tr><th style=\"text-align: left\">Chunks initialized</th><td style=\"text-align: left\">1/719</td></tr></tbody></table>"
+      ],
+      "text/plain": [
+       "Name               : /sst\n",
+       "Type               : zarr.core.Array\n",
+       "Data type          : float32\n",
+       "Shape              : (14372, 1, 720, 1440)\n",
+       "Chunk shape        : (20, 1, 720, 1440)\n",
+       "Order              : C\n",
+       "Read-only          : False\n",
+       "Compressor         : Blosc(cname='lz4', clevel=5, shuffle=SHUFFLE, blocksize=0)\n",
+       "Store type         : zarr.storage.DirectoryStore\n",
+       "No. bytes          : 59603558400 (55.5G)\n",
+       "No. bytes stored   : 39402334 (37.6M)\n",
+       "Storage ratio      : 1512.7\n",
+       "Chunks initialized : 1/719"
+      ]
+     },
+     "execution_count": 26,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "recipe.store_chunk(all_chunks[0])\n",
+    "zgroup['sst'].info"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can see that one of the chunks has been written! 🎉"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can also open the dataset with xarray"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div><svg style=\"position: absolute; width: 0; height: 0; overflow: hidden\">\n",
+       "<defs>\n",
+       "<symbol id=\"icon-database\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M16 0c-8.837 0-16 2.239-16 5v4c0 2.761 7.163 5 16 5s16-2.239 16-5v-4c0-2.761-7.163-5-16-5z\"></path>\n",
+       "<path d=\"M16 17c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "<path d=\"M16 26c-8.837 0-16-2.239-16-5v6c0 2.761 7.163 5 16 5s16-2.239 16-5v-6c0 2.761-7.163 5-16 5z\"></path>\n",
+       "</symbol>\n",
+       "<symbol id=\"icon-file-text2\" viewBox=\"0 0 32 32\">\n",
+       "<path d=\"M28.681 7.159c-0.694-0.947-1.662-2.053-2.724-3.116s-2.169-2.030-3.116-2.724c-1.612-1.182-2.393-1.319-2.841-1.319h-15.5c-1.378 0-2.5 1.121-2.5 2.5v27c0 1.378 1.122 2.5 2.5 2.5h23c1.378 0 2.5-1.122 2.5-2.5v-19.5c0-0.448-0.137-1.23-1.319-2.841zM24.543 5.457c0.959 0.959 1.712 1.825 2.268 2.543h-4.811v-4.811c0.718 0.556 1.584 1.309 2.543 2.268zM28 29.5c0 0.271-0.229 0.5-0.5 0.5h-23c-0.271 0-0.5-0.229-0.5-0.5v-27c0-0.271 0.229-0.5 0.5-0.5 0 0 15.499-0 15.5 0v7c0 0.552 0.448 1 1 1h7v19.5z\"></path>\n",
+       "<path d=\"M23 26h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 22h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "<path d=\"M23 18h-14c-0.552 0-1-0.448-1-1s0.448-1 1-1h14c0.552 0 1 0.448 1 1s-0.448 1-1 1z\"></path>\n",
+       "</symbol>\n",
+       "</defs>\n",
+       "</svg>\n",
+       "<style>/* CSS stylesheet for displaying xarray objects in jupyterlab.\n",
+       " *\n",
+       " */\n",
+       "\n",
+       ":root {\n",
+       "  --xr-font-color0: var(--jp-content-font-color0, rgba(0, 0, 0, 1));\n",
+       "  --xr-font-color2: var(--jp-content-font-color2, rgba(0, 0, 0, 0.54));\n",
+       "  --xr-font-color3: var(--jp-content-font-color3, rgba(0, 0, 0, 0.38));\n",
+       "  --xr-border-color: var(--jp-border-color2, #e0e0e0);\n",
+       "  --xr-disabled-color: var(--jp-layout-color3, #bdbdbd);\n",
+       "  --xr-background-color: var(--jp-layout-color0, white);\n",
+       "  --xr-background-color-row-even: var(--jp-layout-color1, white);\n",
+       "  --xr-background-color-row-odd: var(--jp-layout-color2, #eeeeee);\n",
+       "}\n",
+       "\n",
+       "html[theme=dark],\n",
+       "body.vscode-dark {\n",
+       "  --xr-font-color0: rgba(255, 255, 255, 1);\n",
+       "  --xr-font-color2: rgba(255, 255, 255, 0.54);\n",
+       "  --xr-font-color3: rgba(255, 255, 255, 0.38);\n",
+       "  --xr-border-color: #1F1F1F;\n",
+       "  --xr-disabled-color: #515151;\n",
+       "  --xr-background-color: #111111;\n",
+       "  --xr-background-color-row-even: #111111;\n",
+       "  --xr-background-color-row-odd: #313131;\n",
+       "}\n",
+       "\n",
+       ".xr-wrap {\n",
+       "  display: block;\n",
+       "  min-width: 300px;\n",
+       "  max-width: 700px;\n",
+       "}\n",
+       "\n",
+       ".xr-text-repr-fallback {\n",
+       "  /* fallback to plain text repr when CSS is not injected (untrusted notebook) */\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-header {\n",
+       "  padding-top: 6px;\n",
+       "  padding-bottom: 6px;\n",
+       "  margin-bottom: 4px;\n",
+       "  border-bottom: solid 1px var(--xr-border-color);\n",
+       "}\n",
+       "\n",
+       ".xr-header > div,\n",
+       ".xr-header > ul {\n",
+       "  display: inline;\n",
+       "  margin-top: 0;\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type,\n",
+       ".xr-array-name {\n",
+       "  margin-left: 2px;\n",
+       "  margin-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-obj-type {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-sections {\n",
+       "  padding-left: 0 !important;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 150px auto auto 1fr 20px 20px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input + label {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label {\n",
+       "  cursor: pointer;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-item input:enabled + label:hover {\n",
+       "  color: var(--xr-font-color0);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary {\n",
+       "  grid-column: 1;\n",
+       "  color: var(--xr-font-color2);\n",
+       "  font-weight: 500;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary > span {\n",
+       "  display: inline-block;\n",
+       "  padding-left: 0.5em;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label {\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in + label:before {\n",
+       "  display: inline-block;\n",
+       "  content: '►';\n",
+       "  font-size: 11px;\n",
+       "  width: 15px;\n",
+       "  text-align: center;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:disabled + label:before {\n",
+       "  color: var(--xr-disabled-color);\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label:before {\n",
+       "  content: '▼';\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked + label > span {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary,\n",
+       ".xr-section-inline-details {\n",
+       "  padding-top: 4px;\n",
+       "  padding-bottom: 4px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-inline-details {\n",
+       "  grid-column: 2 / -1;\n",
+       "}\n",
+       "\n",
+       ".xr-section-details {\n",
+       "  display: none;\n",
+       "  grid-column: 1 / -1;\n",
+       "  margin-bottom: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-section-summary-in:checked ~ .xr-section-details {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap {\n",
+       "  grid-column: 1 / -1;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 20px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-array-wrap > label {\n",
+       "  grid-column: 1;\n",
+       "  vertical-align: top;\n",
+       "}\n",
+       "\n",
+       ".xr-preview {\n",
+       "  color: var(--xr-font-color3);\n",
+       "}\n",
+       "\n",
+       ".xr-array-preview,\n",
+       ".xr-array-data {\n",
+       "  padding: 0 5px !important;\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-array-data,\n",
+       ".xr-array-in:checked ~ .xr-array-preview {\n",
+       "  display: none;\n",
+       "}\n",
+       "\n",
+       ".xr-array-in:checked ~ .xr-array-data,\n",
+       ".xr-array-preview {\n",
+       "  display: inline-block;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list {\n",
+       "  display: inline-block !important;\n",
+       "  list-style: none;\n",
+       "  padding: 0 !important;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li {\n",
+       "  display: inline-block;\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:before {\n",
+       "  content: '(';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list:after {\n",
+       "  content: ')';\n",
+       "}\n",
+       "\n",
+       ".xr-dim-list li:not(:last-child):after {\n",
+       "  content: ',';\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-has-index {\n",
+       "  font-weight: bold;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list,\n",
+       ".xr-var-item {\n",
+       "  display: contents;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > div,\n",
+       ".xr-var-item label,\n",
+       ".xr-var-item > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-even);\n",
+       "  margin-bottom: 0;\n",
+       "}\n",
+       "\n",
+       ".xr-var-item > .xr-var-name:hover span {\n",
+       "  padding-right: 5px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-list > li:nth-child(odd) > div,\n",
+       ".xr-var-list > li:nth-child(odd) > label,\n",
+       ".xr-var-list > li:nth-child(odd) > .xr-var-name span {\n",
+       "  background-color: var(--xr-background-color-row-odd);\n",
+       "}\n",
+       "\n",
+       ".xr-var-name {\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dims {\n",
+       "  grid-column: 2;\n",
+       "}\n",
+       "\n",
+       ".xr-var-dtype {\n",
+       "  grid-column: 3;\n",
+       "  text-align: right;\n",
+       "  color: var(--xr-font-color2);\n",
+       "}\n",
+       "\n",
+       ".xr-var-preview {\n",
+       "  grid-column: 4;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name,\n",
+       ".xr-var-dims,\n",
+       ".xr-var-dtype,\n",
+       ".xr-preview,\n",
+       ".xr-attrs dt {\n",
+       "  white-space: nowrap;\n",
+       "  overflow: hidden;\n",
+       "  text-overflow: ellipsis;\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name:hover,\n",
+       ".xr-var-dims:hover,\n",
+       ".xr-var-dtype:hover,\n",
+       ".xr-attrs dt:hover {\n",
+       "  overflow: visible;\n",
+       "  width: auto;\n",
+       "  z-index: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  display: none;\n",
+       "  background-color: var(--xr-background-color) !important;\n",
+       "  padding-bottom: 5px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-var-attrs-in:checked ~ .xr-var-attrs,\n",
+       ".xr-var-data-in:checked ~ .xr-var-data {\n",
+       "  display: block;\n",
+       "}\n",
+       "\n",
+       ".xr-var-data > table {\n",
+       "  float: right;\n",
+       "}\n",
+       "\n",
+       ".xr-var-name span,\n",
+       ".xr-var-data,\n",
+       ".xr-attrs {\n",
+       "  padding-left: 25px !important;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs,\n",
+       ".xr-var-attrs,\n",
+       ".xr-var-data {\n",
+       "  grid-column: 1 / -1;\n",
+       "}\n",
+       "\n",
+       "dl.xr-attrs {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  display: grid;\n",
+       "  grid-template-columns: 125px auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt,\n",
+       ".xr-attrs dd {\n",
+       "  padding: 0;\n",
+       "  margin: 0;\n",
+       "  float: left;\n",
+       "  padding-right: 10px;\n",
+       "  width: auto;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt {\n",
+       "  font-weight: normal;\n",
+       "  grid-column: 1;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dt:hover span {\n",
+       "  display: inline-block;\n",
+       "  background: var(--xr-background-color);\n",
+       "  padding-right: 10px;\n",
+       "}\n",
+       "\n",
+       ".xr-attrs dd {\n",
+       "  grid-column: 2;\n",
+       "  white-space: pre-wrap;\n",
+       "  word-break: break-all;\n",
+       "}\n",
+       "\n",
+       ".xr-icon-database,\n",
+       ".xr-icon-file-text2 {\n",
+       "  display: inline-block;\n",
+       "  vertical-align: middle;\n",
+       "  width: 1em;\n",
+       "  height: 1.5em !important;\n",
+       "  stroke-width: 0;\n",
+       "  stroke: currentColor;\n",
+       "  fill: currentColor;\n",
+       "}\n",
+       "</style><pre class='xr-text-repr-fallback'>&lt;xarray.Dataset&gt;\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 14372, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 1981-09-01T12:00:00 ... NaT\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;\n",
+       "    err      (time, zlev, lat, lon) float32 dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;\n",
+       "    ice      (time, zlev, lat, lon) float32 dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;\n",
+       "    sst      (time, zlev, lat, lon) float32 dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;\n",
+       "Attributes:\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    cdm_data_type:              Grid\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    id:                         oisst-avhrr-v02r01.19810901.nc\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments &gt; Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    keywords:                   Earth Science &gt; Oceans &gt; Ocean Temperature &gt; ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    product_version:            Version v02r01\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    time_coverage_end:          1981-09-01T23:59:59Z\n",
+       "    time_coverage_start:        1981-09-01T00:00:00Z\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat...</pre><div class='xr-wrap' hidden><div class='xr-header'><div class='xr-obj-type'>xarray.Dataset</div></div><ul class='xr-sections'><li class='xr-section-item'><input id='section-9a9d0647-4354-45fe-becc-30ab2e423018' class='xr-section-summary-in' type='checkbox' disabled ><label for='section-9a9d0647-4354-45fe-becc-30ab2e423018' class='xr-section-summary'  title='Expand/collapse section'>Dimensions:</label><div class='xr-section-inline-details'><ul class='xr-dim-list'><li><span class='xr-has-index'>lat</span>: 720</li><li><span class='xr-has-index'>lon</span>: 1440</li><li><span class='xr-has-index'>time</span>: 14372</li><li><span class='xr-has-index'>zlev</span>: 1</li></ul></div><div class='xr-section-details'></div></li><li class='xr-section-item'><input id='section-fe692aff-7e8c-4dbf-be21-5143638139e3' class='xr-section-summary-in' type='checkbox'  checked><label for='section-fe692aff-7e8c-4dbf-be21-5143638139e3' class='xr-section-summary' >Coordinates: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lat</span></div><div class='xr-var-dims'>(lat)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>-89.88 -89.62 ... 89.62 89.88</div><input id='attrs-986dca20-e34e-4d2b-918f-ab32d9560996' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-986dca20-e34e-4d2b-918f-ab32d9560996' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-b554f5cb-b3e9-42e9-8bcf-e2862ff03b33' class='xr-var-data-in' type='checkbox'><label for='data-b554f5cb-b3e9-42e9-8bcf-e2862ff03b33' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>grids :</span></dt><dd>Uniform grid from -89.875 to 89.875 by 0.25</dd><dt><span>long_name :</span></dt><dd>Latitude</dd><dt><span>units :</span></dt><dd>degrees_north</dd></dl></div><div class='xr-var-data'><pre>array([-89.875, -89.625, -89.375, ...,  89.375,  89.625,  89.875],\n",
+       "      dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>lon</span></div><div class='xr-var-dims'>(lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.125 0.375 0.625 ... 359.6 359.9</div><input id='attrs-bb45b0cf-30fa-451d-abef-7cc24b1d3c97' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-bb45b0cf-30fa-451d-abef-7cc24b1d3c97' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-e19ac340-ad81-4a17-ba7f-596f2a71fc66' class='xr-var-data-in' type='checkbox'><label for='data-e19ac340-ad81-4a17-ba7f-596f2a71fc66' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>grids :</span></dt><dd>Uniform grid from 0.125 to 359.875 by 0.25</dd><dt><span>long_name :</span></dt><dd>Longitude</dd><dt><span>units :</span></dt><dd>degrees_east</dd></dl></div><div class='xr-var-data'><pre>array([1.25000e-01, 3.75000e-01, 6.25000e-01, ..., 3.59375e+02, 3.59625e+02,\n",
+       "       3.59875e+02], dtype=float32)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>time</span></div><div class='xr-var-dims'>(time)</div><div class='xr-var-dtype'>datetime64[ns]</div><div class='xr-var-preview xr-preview'>1981-09-01T12:00:00 ... NaT</div><input id='attrs-60ce66c9-a552-48b4-80ee-0f11f0cf0347' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-60ce66c9-a552-48b4-80ee-0f11f0cf0347' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-4c75a7fa-c8f6-4ca4-8834-291a2af20734' class='xr-var-data-in' type='checkbox'><label for='data-4c75a7fa-c8f6-4ca4-8834-291a2af20734' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Center time of the day</dd></dl></div><div class='xr-var-data'><pre>array([&#x27;1981-09-01T12:00:00.000000000&#x27;, &#x27;1981-09-02T12:00:00.000000000&#x27;,\n",
+       "       &#x27;1981-09-03T12:00:00.000000000&#x27;, ...,                           &#x27;NaT&#x27;,\n",
+       "                                 &#x27;NaT&#x27;,                           &#x27;NaT&#x27;],\n",
+       "      dtype=&#x27;datetime64[ns]&#x27;)</pre></div></li><li class='xr-var-item'><div class='xr-var-name'><span class='xr-has-index'>zlev</span></div><div class='xr-var-dims'>(zlev)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>0.0</div><input id='attrs-5f396e5e-c47a-49b9-a5a9-1b60aebdd5d2' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-5f396e5e-c47a-49b9-a5a9-1b60aebdd5d2' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-b82925ea-7ac8-40c2-8f0b-85ab7623c309' class='xr-var-data-in' type='checkbox'><label for='data-b82925ea-7ac8-40c2-8f0b-85ab7623c309' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>actual_range :</span></dt><dd>0, 0</dd><dt><span>long_name :</span></dt><dd>Sea surface height</dd><dt><span>positive :</span></dt><dd>down</dd><dt><span>units :</span></dt><dd>meters</dd></dl></div><div class='xr-var-data'><pre>array([0.], dtype=float32)</pre></div></li></ul></div></li><li class='xr-section-item'><input id='section-548b87fe-fca3-4373-9bbd-f26bd98991e9' class='xr-section-summary-in' type='checkbox'  checked><label for='section-548b87fe-fca3-4373-9bbd-f26bd98991e9' class='xr-section-summary' >Data variables: <span>(4)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><ul class='xr-var-list'><li class='xr-var-item'><div class='xr-var-name'><span>anom</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;</div><input id='attrs-edbbe642-5847-4be7-adcd-b8092a1abbc0' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-edbbe642-5847-4be7-adcd-b8092a1abbc0' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-8223e778-9f30-49cc-af1f-51aa1d881f48' class='xr-var-data-in' type='checkbox'><label for='data-8223e778-9f30-49cc-af1f-51aa1d881f48' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature anomalies</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_max :</span></dt><dd>1200</dd><dt><span>valid_min :</span></dt><dd>-1200</dd></dl></div><div class='xr-var-data'><table>\n",
+       "<tr>\n",
+       "<td>\n",
+       "<table>\n",
+       "  <thead>\n",
+       "    <tr><td> </td><th> Array </th><th> Chunk </th></tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr><th> Bytes </th><td> 59.60 GB </td> <td> 82.94 MB </td></tr>\n",
+       "    <tr><th> Shape </th><td> (14372, 1, 720, 1440) </td> <td> (20, 1, 720, 1440) </td></tr>\n",
+       "    <tr><th> Count </th><td> 720 Tasks </td><td> 719 Chunks </td></tr>\n",
+       "    <tr><th> Type </th><td> float32 </td><td> numpy.ndarray </td></tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</td>\n",
+       "<td>\n",
+       "<svg width=\"483\" height=\"99\" style=\"stroke:rgb(0,0,0);stroke-width:1\" >\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"120\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"0\" y1=\"25\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"0\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"3\" y1=\"0\" x2=\"3\" y2=\"25\" />\n",
+       "  <line x1=\"7\" y1=\"0\" x2=\"7\" y2=\"25\" />\n",
+       "  <line x1=\"11\" y1=\"0\" x2=\"11\" y2=\"25\" />\n",
+       "  <line x1=\"14\" y1=\"0\" x2=\"14\" y2=\"25\" />\n",
+       "  <line x1=\"18\" y1=\"0\" x2=\"18\" y2=\"25\" />\n",
+       "  <line x1=\"22\" y1=\"0\" x2=\"22\" y2=\"25\" />\n",
+       "  <line x1=\"26\" y1=\"0\" x2=\"26\" y2=\"25\" />\n",
+       "  <line x1=\"29\" y1=\"0\" x2=\"29\" y2=\"25\" />\n",
+       "  <line x1=\"33\" y1=\"0\" x2=\"33\" y2=\"25\" />\n",
+       "  <line x1=\"37\" y1=\"0\" x2=\"37\" y2=\"25\" />\n",
+       "  <line x1=\"41\" y1=\"0\" x2=\"41\" y2=\"25\" />\n",
+       "  <line x1=\"44\" y1=\"0\" x2=\"44\" y2=\"25\" />\n",
+       "  <line x1=\"48\" y1=\"0\" x2=\"48\" y2=\"25\" />\n",
+       "  <line x1=\"52\" y1=\"0\" x2=\"52\" y2=\"25\" />\n",
+       "  <line x1=\"56\" y1=\"0\" x2=\"56\" y2=\"25\" />\n",
+       "  <line x1=\"59\" y1=\"0\" x2=\"59\" y2=\"25\" />\n",
+       "  <line x1=\"63\" y1=\"0\" x2=\"63\" y2=\"25\" />\n",
+       "  <line x1=\"67\" y1=\"0\" x2=\"67\" y2=\"25\" />\n",
+       "  <line x1=\"71\" y1=\"0\" x2=\"71\" y2=\"25\" />\n",
+       "  <line x1=\"74\" y1=\"0\" x2=\"74\" y2=\"25\" />\n",
+       "  <line x1=\"78\" y1=\"0\" x2=\"78\" y2=\"25\" />\n",
+       "  <line x1=\"82\" y1=\"0\" x2=\"82\" y2=\"25\" />\n",
+       "  <line x1=\"86\" y1=\"0\" x2=\"86\" y2=\"25\" />\n",
+       "  <line x1=\"90\" y1=\"0\" x2=\"90\" y2=\"25\" />\n",
+       "  <line x1=\"93\" y1=\"0\" x2=\"93\" y2=\"25\" />\n",
+       "  <line x1=\"97\" y1=\"0\" x2=\"97\" y2=\"25\" />\n",
+       "  <line x1=\"101\" y1=\"0\" x2=\"101\" y2=\"25\" />\n",
+       "  <line x1=\"105\" y1=\"0\" x2=\"105\" y2=\"25\" />\n",
+       "  <line x1=\"108\" y1=\"0\" x2=\"108\" y2=\"25\" />\n",
+       "  <line x1=\"112\" y1=\"0\" x2=\"112\" y2=\"25\" />\n",
+       "  <line x1=\"116\" y1=\"0\" x2=\"116\" y2=\"25\" />\n",
+       "  <line x1=\"120\" y1=\"0\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"0.0,0.0 120.0,0.0 120.0,25.412616514582485 0.0,25.412616514582485\" style=\"fill:#8B4903A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"60.000000\" y=\"45.412617\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >14372</text>\n",
+       "  <text x=\"140.000000\" y=\"12.706308\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(0,140.000000,12.706308)\">1</text>\n",
+       "\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"190\" y1=\"34\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"190\" y2=\"34\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 204.9485979497544,14.948597949754403 204.9485979497544,49.46153131194814 190.0,34.51293336219374\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"228\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"228\" y1=\"0\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 228.6076470436033,0.0 243.5562449933577,14.948597949754403 204.9485979497544,14.948597949754403\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"49\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"243\" y1=\"14\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"204.9485979497544,14.948597949754403 243.5562449933577,14.948597949754403 243.5562449933577,49.46153131194814 204.9485979497544,49.46153131194814\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"224.252421\" y=\"69.461531\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >1440</text>\n",
+       "  <text x=\"263.556245\" y=\"32.205065\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(-90,263.556245,32.205065)\">720</text>\n",
+       "  <text x=\"187.474299\" y=\"61.987232\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(45,187.474299,61.987232)\">1</text>\n",
+       "</svg>\n",
+       "</td>\n",
+       "</tr>\n",
+       "</table></div></li><li class='xr-var-item'><div class='xr-var-name'><span>err</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;</div><input id='attrs-bf3a525c-6d8d-470f-9c34-390c238e585c' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-bf3a525c-6d8d-470f-9c34-390c238e585c' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-f85cb2c3-467c-4f6b-8ecb-85926f1d7b3d' class='xr-var-data-in' type='checkbox'><label for='data-f85cb2c3-467c-4f6b-8ecb-85926f1d7b3d' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Estimated error standard deviation of analysed_sst</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_max :</span></dt><dd>1000</dd><dt><span>valid_min :</span></dt><dd>0</dd></dl></div><div class='xr-var-data'><table>\n",
+       "<tr>\n",
+       "<td>\n",
+       "<table>\n",
+       "  <thead>\n",
+       "    <tr><td> </td><th> Array </th><th> Chunk </th></tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr><th> Bytes </th><td> 59.60 GB </td> <td> 82.94 MB </td></tr>\n",
+       "    <tr><th> Shape </th><td> (14372, 1, 720, 1440) </td> <td> (20, 1, 720, 1440) </td></tr>\n",
+       "    <tr><th> Count </th><td> 720 Tasks </td><td> 719 Chunks </td></tr>\n",
+       "    <tr><th> Type </th><td> float32 </td><td> numpy.ndarray </td></tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</td>\n",
+       "<td>\n",
+       "<svg width=\"483\" height=\"99\" style=\"stroke:rgb(0,0,0);stroke-width:1\" >\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"120\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"0\" y1=\"25\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"0\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"3\" y1=\"0\" x2=\"3\" y2=\"25\" />\n",
+       "  <line x1=\"7\" y1=\"0\" x2=\"7\" y2=\"25\" />\n",
+       "  <line x1=\"11\" y1=\"0\" x2=\"11\" y2=\"25\" />\n",
+       "  <line x1=\"14\" y1=\"0\" x2=\"14\" y2=\"25\" />\n",
+       "  <line x1=\"18\" y1=\"0\" x2=\"18\" y2=\"25\" />\n",
+       "  <line x1=\"22\" y1=\"0\" x2=\"22\" y2=\"25\" />\n",
+       "  <line x1=\"26\" y1=\"0\" x2=\"26\" y2=\"25\" />\n",
+       "  <line x1=\"29\" y1=\"0\" x2=\"29\" y2=\"25\" />\n",
+       "  <line x1=\"33\" y1=\"0\" x2=\"33\" y2=\"25\" />\n",
+       "  <line x1=\"37\" y1=\"0\" x2=\"37\" y2=\"25\" />\n",
+       "  <line x1=\"41\" y1=\"0\" x2=\"41\" y2=\"25\" />\n",
+       "  <line x1=\"44\" y1=\"0\" x2=\"44\" y2=\"25\" />\n",
+       "  <line x1=\"48\" y1=\"0\" x2=\"48\" y2=\"25\" />\n",
+       "  <line x1=\"52\" y1=\"0\" x2=\"52\" y2=\"25\" />\n",
+       "  <line x1=\"56\" y1=\"0\" x2=\"56\" y2=\"25\" />\n",
+       "  <line x1=\"59\" y1=\"0\" x2=\"59\" y2=\"25\" />\n",
+       "  <line x1=\"63\" y1=\"0\" x2=\"63\" y2=\"25\" />\n",
+       "  <line x1=\"67\" y1=\"0\" x2=\"67\" y2=\"25\" />\n",
+       "  <line x1=\"71\" y1=\"0\" x2=\"71\" y2=\"25\" />\n",
+       "  <line x1=\"74\" y1=\"0\" x2=\"74\" y2=\"25\" />\n",
+       "  <line x1=\"78\" y1=\"0\" x2=\"78\" y2=\"25\" />\n",
+       "  <line x1=\"82\" y1=\"0\" x2=\"82\" y2=\"25\" />\n",
+       "  <line x1=\"86\" y1=\"0\" x2=\"86\" y2=\"25\" />\n",
+       "  <line x1=\"90\" y1=\"0\" x2=\"90\" y2=\"25\" />\n",
+       "  <line x1=\"93\" y1=\"0\" x2=\"93\" y2=\"25\" />\n",
+       "  <line x1=\"97\" y1=\"0\" x2=\"97\" y2=\"25\" />\n",
+       "  <line x1=\"101\" y1=\"0\" x2=\"101\" y2=\"25\" />\n",
+       "  <line x1=\"105\" y1=\"0\" x2=\"105\" y2=\"25\" />\n",
+       "  <line x1=\"108\" y1=\"0\" x2=\"108\" y2=\"25\" />\n",
+       "  <line x1=\"112\" y1=\"0\" x2=\"112\" y2=\"25\" />\n",
+       "  <line x1=\"116\" y1=\"0\" x2=\"116\" y2=\"25\" />\n",
+       "  <line x1=\"120\" y1=\"0\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"0.0,0.0 120.0,0.0 120.0,25.412616514582485 0.0,25.412616514582485\" style=\"fill:#8B4903A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"60.000000\" y=\"45.412617\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >14372</text>\n",
+       "  <text x=\"140.000000\" y=\"12.706308\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(0,140.000000,12.706308)\">1</text>\n",
+       "\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"190\" y1=\"34\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"190\" y2=\"34\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 204.9485979497544,14.948597949754403 204.9485979497544,49.46153131194814 190.0,34.51293336219374\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"228\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"228\" y1=\"0\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 228.6076470436033,0.0 243.5562449933577,14.948597949754403 204.9485979497544,14.948597949754403\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"49\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"243\" y1=\"14\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"204.9485979497544,14.948597949754403 243.5562449933577,14.948597949754403 243.5562449933577,49.46153131194814 204.9485979497544,49.46153131194814\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"224.252421\" y=\"69.461531\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >1440</text>\n",
+       "  <text x=\"263.556245\" y=\"32.205065\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(-90,263.556245,32.205065)\">720</text>\n",
+       "  <text x=\"187.474299\" y=\"61.987232\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(45,187.474299,61.987232)\">1</text>\n",
+       "</svg>\n",
+       "</td>\n",
+       "</tr>\n",
+       "</table></div></li><li class='xr-var-item'><div class='xr-var-name'><span>ice</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;</div><input id='attrs-05791218-d5bc-45e2-97b8-813dc8764967' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-05791218-d5bc-45e2-97b8-813dc8764967' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-061f7e09-37d1-44d9-a699-afb1ffc6cae2' class='xr-var-data-in' type='checkbox'><label for='data-061f7e09-37d1-44d9-a699-afb1ffc6cae2' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Sea ice concentration</dd><dt><span>units :</span></dt><dd>%</dd><dt><span>valid_max :</span></dt><dd>100</dd><dt><span>valid_min :</span></dt><dd>0</dd></dl></div><div class='xr-var-data'><table>\n",
+       "<tr>\n",
+       "<td>\n",
+       "<table>\n",
+       "  <thead>\n",
+       "    <tr><td> </td><th> Array </th><th> Chunk </th></tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr><th> Bytes </th><td> 59.60 GB </td> <td> 82.94 MB </td></tr>\n",
+       "    <tr><th> Shape </th><td> (14372, 1, 720, 1440) </td> <td> (20, 1, 720, 1440) </td></tr>\n",
+       "    <tr><th> Count </th><td> 720 Tasks </td><td> 719 Chunks </td></tr>\n",
+       "    <tr><th> Type </th><td> float32 </td><td> numpy.ndarray </td></tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</td>\n",
+       "<td>\n",
+       "<svg width=\"483\" height=\"99\" style=\"stroke:rgb(0,0,0);stroke-width:1\" >\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"120\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"0\" y1=\"25\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"0\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"3\" y1=\"0\" x2=\"3\" y2=\"25\" />\n",
+       "  <line x1=\"7\" y1=\"0\" x2=\"7\" y2=\"25\" />\n",
+       "  <line x1=\"11\" y1=\"0\" x2=\"11\" y2=\"25\" />\n",
+       "  <line x1=\"14\" y1=\"0\" x2=\"14\" y2=\"25\" />\n",
+       "  <line x1=\"18\" y1=\"0\" x2=\"18\" y2=\"25\" />\n",
+       "  <line x1=\"22\" y1=\"0\" x2=\"22\" y2=\"25\" />\n",
+       "  <line x1=\"26\" y1=\"0\" x2=\"26\" y2=\"25\" />\n",
+       "  <line x1=\"29\" y1=\"0\" x2=\"29\" y2=\"25\" />\n",
+       "  <line x1=\"33\" y1=\"0\" x2=\"33\" y2=\"25\" />\n",
+       "  <line x1=\"37\" y1=\"0\" x2=\"37\" y2=\"25\" />\n",
+       "  <line x1=\"41\" y1=\"0\" x2=\"41\" y2=\"25\" />\n",
+       "  <line x1=\"44\" y1=\"0\" x2=\"44\" y2=\"25\" />\n",
+       "  <line x1=\"48\" y1=\"0\" x2=\"48\" y2=\"25\" />\n",
+       "  <line x1=\"52\" y1=\"0\" x2=\"52\" y2=\"25\" />\n",
+       "  <line x1=\"56\" y1=\"0\" x2=\"56\" y2=\"25\" />\n",
+       "  <line x1=\"59\" y1=\"0\" x2=\"59\" y2=\"25\" />\n",
+       "  <line x1=\"63\" y1=\"0\" x2=\"63\" y2=\"25\" />\n",
+       "  <line x1=\"67\" y1=\"0\" x2=\"67\" y2=\"25\" />\n",
+       "  <line x1=\"71\" y1=\"0\" x2=\"71\" y2=\"25\" />\n",
+       "  <line x1=\"74\" y1=\"0\" x2=\"74\" y2=\"25\" />\n",
+       "  <line x1=\"78\" y1=\"0\" x2=\"78\" y2=\"25\" />\n",
+       "  <line x1=\"82\" y1=\"0\" x2=\"82\" y2=\"25\" />\n",
+       "  <line x1=\"86\" y1=\"0\" x2=\"86\" y2=\"25\" />\n",
+       "  <line x1=\"90\" y1=\"0\" x2=\"90\" y2=\"25\" />\n",
+       "  <line x1=\"93\" y1=\"0\" x2=\"93\" y2=\"25\" />\n",
+       "  <line x1=\"97\" y1=\"0\" x2=\"97\" y2=\"25\" />\n",
+       "  <line x1=\"101\" y1=\"0\" x2=\"101\" y2=\"25\" />\n",
+       "  <line x1=\"105\" y1=\"0\" x2=\"105\" y2=\"25\" />\n",
+       "  <line x1=\"108\" y1=\"0\" x2=\"108\" y2=\"25\" />\n",
+       "  <line x1=\"112\" y1=\"0\" x2=\"112\" y2=\"25\" />\n",
+       "  <line x1=\"116\" y1=\"0\" x2=\"116\" y2=\"25\" />\n",
+       "  <line x1=\"120\" y1=\"0\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"0.0,0.0 120.0,0.0 120.0,25.412616514582485 0.0,25.412616514582485\" style=\"fill:#8B4903A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"60.000000\" y=\"45.412617\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >14372</text>\n",
+       "  <text x=\"140.000000\" y=\"12.706308\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(0,140.000000,12.706308)\">1</text>\n",
+       "\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"190\" y1=\"34\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"190\" y2=\"34\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 204.9485979497544,14.948597949754403 204.9485979497544,49.46153131194814 190.0,34.51293336219374\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"228\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"228\" y1=\"0\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 228.6076470436033,0.0 243.5562449933577,14.948597949754403 204.9485979497544,14.948597949754403\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"49\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"243\" y1=\"14\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"204.9485979497544,14.948597949754403 243.5562449933577,14.948597949754403 243.5562449933577,49.46153131194814 204.9485979497544,49.46153131194814\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"224.252421\" y=\"69.461531\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >1440</text>\n",
+       "  <text x=\"263.556245\" y=\"32.205065\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(-90,263.556245,32.205065)\">720</text>\n",
+       "  <text x=\"187.474299\" y=\"61.987232\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(45,187.474299,61.987232)\">1</text>\n",
+       "</svg>\n",
+       "</td>\n",
+       "</tr>\n",
+       "</table></div></li><li class='xr-var-item'><div class='xr-var-name'><span>sst</span></div><div class='xr-var-dims'>(time, zlev, lat, lon)</div><div class='xr-var-dtype'>float32</div><div class='xr-var-preview xr-preview'>dask.array&lt;chunksize=(20, 1, 720, 1440), meta=np.ndarray&gt;</div><input id='attrs-86de51d9-cb03-45e1-b60b-10fda6c0872d' class='xr-var-attrs-in' type='checkbox' ><label for='attrs-86de51d9-cb03-45e1-b60b-10fda6c0872d' title='Show/Hide attributes'><svg class='icon xr-icon-file-text2'><use xlink:href='#icon-file-text2'></use></svg></label><input id='data-ec099a3e-1b7a-4897-8eec-6c0e9917bf5d' class='xr-var-data-in' type='checkbox'><label for='data-ec099a3e-1b7a-4897-8eec-6c0e9917bf5d' title='Show/Hide data repr'><svg class='icon xr-icon-database'><use xlink:href='#icon-database'></use></svg></label><div class='xr-var-attrs'><dl class='xr-attrs'><dt><span>long_name :</span></dt><dd>Daily sea surface temperature</dd><dt><span>units :</span></dt><dd>Celsius</dd><dt><span>valid_max :</span></dt><dd>4500</dd><dt><span>valid_min :</span></dt><dd>-300</dd></dl></div><div class='xr-var-data'><table>\n",
+       "<tr>\n",
+       "<td>\n",
+       "<table>\n",
+       "  <thead>\n",
+       "    <tr><td> </td><th> Array </th><th> Chunk </th></tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr><th> Bytes </th><td> 59.60 GB </td> <td> 82.94 MB </td></tr>\n",
+       "    <tr><th> Shape </th><td> (14372, 1, 720, 1440) </td> <td> (20, 1, 720, 1440) </td></tr>\n",
+       "    <tr><th> Count </th><td> 720 Tasks </td><td> 719 Chunks </td></tr>\n",
+       "    <tr><th> Type </th><td> float32 </td><td> numpy.ndarray </td></tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</td>\n",
+       "<td>\n",
+       "<svg width=\"483\" height=\"99\" style=\"stroke:rgb(0,0,0);stroke-width:1\" >\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"120\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"0\" y1=\"25\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"0\" y1=\"0\" x2=\"0\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"3\" y1=\"0\" x2=\"3\" y2=\"25\" />\n",
+       "  <line x1=\"7\" y1=\"0\" x2=\"7\" y2=\"25\" />\n",
+       "  <line x1=\"11\" y1=\"0\" x2=\"11\" y2=\"25\" />\n",
+       "  <line x1=\"14\" y1=\"0\" x2=\"14\" y2=\"25\" />\n",
+       "  <line x1=\"18\" y1=\"0\" x2=\"18\" y2=\"25\" />\n",
+       "  <line x1=\"22\" y1=\"0\" x2=\"22\" y2=\"25\" />\n",
+       "  <line x1=\"26\" y1=\"0\" x2=\"26\" y2=\"25\" />\n",
+       "  <line x1=\"29\" y1=\"0\" x2=\"29\" y2=\"25\" />\n",
+       "  <line x1=\"33\" y1=\"0\" x2=\"33\" y2=\"25\" />\n",
+       "  <line x1=\"37\" y1=\"0\" x2=\"37\" y2=\"25\" />\n",
+       "  <line x1=\"41\" y1=\"0\" x2=\"41\" y2=\"25\" />\n",
+       "  <line x1=\"44\" y1=\"0\" x2=\"44\" y2=\"25\" />\n",
+       "  <line x1=\"48\" y1=\"0\" x2=\"48\" y2=\"25\" />\n",
+       "  <line x1=\"52\" y1=\"0\" x2=\"52\" y2=\"25\" />\n",
+       "  <line x1=\"56\" y1=\"0\" x2=\"56\" y2=\"25\" />\n",
+       "  <line x1=\"59\" y1=\"0\" x2=\"59\" y2=\"25\" />\n",
+       "  <line x1=\"63\" y1=\"0\" x2=\"63\" y2=\"25\" />\n",
+       "  <line x1=\"67\" y1=\"0\" x2=\"67\" y2=\"25\" />\n",
+       "  <line x1=\"71\" y1=\"0\" x2=\"71\" y2=\"25\" />\n",
+       "  <line x1=\"74\" y1=\"0\" x2=\"74\" y2=\"25\" />\n",
+       "  <line x1=\"78\" y1=\"0\" x2=\"78\" y2=\"25\" />\n",
+       "  <line x1=\"82\" y1=\"0\" x2=\"82\" y2=\"25\" />\n",
+       "  <line x1=\"86\" y1=\"0\" x2=\"86\" y2=\"25\" />\n",
+       "  <line x1=\"90\" y1=\"0\" x2=\"90\" y2=\"25\" />\n",
+       "  <line x1=\"93\" y1=\"0\" x2=\"93\" y2=\"25\" />\n",
+       "  <line x1=\"97\" y1=\"0\" x2=\"97\" y2=\"25\" />\n",
+       "  <line x1=\"101\" y1=\"0\" x2=\"101\" y2=\"25\" />\n",
+       "  <line x1=\"105\" y1=\"0\" x2=\"105\" y2=\"25\" />\n",
+       "  <line x1=\"108\" y1=\"0\" x2=\"108\" y2=\"25\" />\n",
+       "  <line x1=\"112\" y1=\"0\" x2=\"112\" y2=\"25\" />\n",
+       "  <line x1=\"116\" y1=\"0\" x2=\"116\" y2=\"25\" />\n",
+       "  <line x1=\"120\" y1=\"0\" x2=\"120\" y2=\"25\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"0.0,0.0 120.0,0.0 120.0,25.412616514582485 0.0,25.412616514582485\" style=\"fill:#8B4903A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"60.000000\" y=\"45.412617\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >14372</text>\n",
+       "  <text x=\"140.000000\" y=\"12.706308\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(0,140.000000,12.706308)\">1</text>\n",
+       "\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"190\" y1=\"34\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"190\" y2=\"34\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 204.9485979497544,14.948597949754403 204.9485979497544,49.46153131194814 190.0,34.51293336219374\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"228\" y2=\"0\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"190\" y1=\"0\" x2=\"204\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"228\" y1=\"0\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"190.0,0.0 228.6076470436033,0.0 243.5562449933577,14.948597949754403 204.9485979497544,14.948597949754403\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Horizontal lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"243\" y2=\"14\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"204\" y1=\"49\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Vertical lines -->\n",
+       "  <line x1=\"204\" y1=\"14\" x2=\"204\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "  <line x1=\"243\" y1=\"14\" x2=\"243\" y2=\"49\" style=\"stroke-width:2\" />\n",
+       "\n",
+       "  <!-- Colored Rectangle -->\n",
+       "  <polygon points=\"204.9485979497544,14.948597949754403 243.5562449933577,14.948597949754403 243.5562449933577,49.46153131194814 204.9485979497544,49.46153131194814\" style=\"fill:#ECB172A0;stroke-width:0\"/>\n",
+       "\n",
+       "  <!-- Text -->\n",
+       "  <text x=\"224.252421\" y=\"69.461531\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" >1440</text>\n",
+       "  <text x=\"263.556245\" y=\"32.205065\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(-90,263.556245,32.205065)\">720</text>\n",
+       "  <text x=\"187.474299\" y=\"61.987232\" font-size=\"1.0rem\" font-weight=\"100\" text-anchor=\"middle\" transform=\"rotate(45,187.474299,61.987232)\">1</text>\n",
+       "</svg>\n",
+       "</td>\n",
+       "</tr>\n",
+       "</table></div></li></ul></div></li><li class='xr-section-item'><input id='section-898b18b5-8612-460a-a044-f30c720728e9' class='xr-section-summary-in' type='checkbox'  ><label for='section-898b18b5-8612-460a-a044-f30c720728e9' class='xr-section-summary' >Attributes: <span>(37)</span></label><div class='xr-section-inline-details'></div><div class='xr-section-details'><dl class='xr-attrs'><dt><span>Conventions :</span></dt><dd>CF-1.6, ACDD-1.3</dd><dt><span>cdm_data_type :</span></dt><dd>Grid</dd><dt><span>comment :</span></dt><dd>Data was converted from NetCDF-3 to NetCDF-4 format with metadata updates in November 2017.</dd><dt><span>creator_email :</span></dt><dd>oisst-help@noaa.gov</dd><dt><span>creator_url :</span></dt><dd>https://www.ncei.noaa.gov/</dd><dt><span>date_created :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>date_modified :</span></dt><dd>2020-05-08T19:05:13Z</dd><dt><span>geospatial_lat_max :</span></dt><dd>90.0</dd><dt><span>geospatial_lat_min :</span></dt><dd>-90.0</dd><dt><span>geospatial_lat_resolution :</span></dt><dd>0.25</dd><dt><span>geospatial_lat_units :</span></dt><dd>degrees_north</dd><dt><span>geospatial_lon_max :</span></dt><dd>360.0</dd><dt><span>geospatial_lon_min :</span></dt><dd>0.0</dd><dt><span>geospatial_lon_resolution :</span></dt><dd>0.25</dd><dt><span>geospatial_lon_units :</span></dt><dd>degrees_east</dd><dt><span>history :</span></dt><dd>Final file created using preliminary as first guess, and 3 days of AVHRR data. Preliminary uses only 1 day of AVHRR data.</dd><dt><span>id :</span></dt><dd>oisst-avhrr-v02r01.19810901.nc</dd><dt><span>institution :</span></dt><dd>NOAA/National Centers for Environmental Information</dd><dt><span>instrument :</span></dt><dd>Earth Remote Sensing Instruments &gt; Passive Remote Sensing &gt; Spectrometers/Radiometers &gt; Imaging Spectrometers/Radiometers &gt; AVHRR &gt; Advanced Very High Resolution Radiometer</dd><dt><span>instrument_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Instrument Keywords</dd><dt><span>keywords :</span></dt><dd>Earth Science &gt; Oceans &gt; Ocean Temperature &gt; Sea Surface Temperature</dd><dt><span>keywords_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Earth Science Keywords</dd><dt><span>metadata_link :</span></dt><dd>https://doi.org/10.25921/RE9P-PT57</dd><dt><span>naming_authority :</span></dt><dd>gov.noaa.ncei</dd><dt><span>ncei_template_version :</span></dt><dd>NCEI_NetCDF_Grid_Template_v2.0</dd><dt><span>platform :</span></dt><dd>Ships, buoys, Argo floats, MetOp-A, MetOp-B</dd><dt><span>platform_vocabulary :</span></dt><dd>Global Change Master Directory (GCMD) Platform Keywords</dd><dt><span>processing_level :</span></dt><dd>NOAA Level 4</dd><dt><span>product_version :</span></dt><dd>Version v02r01</dd><dt><span>references :</span></dt><dd>Reynolds, et al.(2007) Daily High-Resolution-Blended Analyses for Sea Surface Temperature (available at https://doi.org/10.1175/2007JCLI1824.1). Banzon, et al.(2016) A long-term record of blended satellite and in situ sea-surface temperature for climate monitoring, modeling and environmental studies (available at https://doi.org/10.5194/essd-8-165-2016). Huang et al. (2020) Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version v02r01, submitted.Climatology is based on 1971-2000 OI.v2 SST. Satellite data: Pathfinder AVHRR SST and Navy AVHRR SST. Ice data: NCEP Ice and GSFC Ice.</dd><dt><span>sensor :</span></dt><dd>Thermometer, AVHRR</dd><dt><span>source :</span></dt><dd>ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfinder_AVHRR, Navy_AVHRR</dd><dt><span>standard_name_vocabulary :</span></dt><dd>CF Standard Name Table (v40, 25 January 2017)</dd><dt><span>summary :</span></dt><dd>NOAAs 1/4-degree Daily Optimum Interpolation Sea Surface Temperature (OISST) (sometimes referred to as Reynolds SST, which however also refers to earlier products at different resolution), currently available as version v02r01, is created by interpolating and extrapolating SST observations from different sources, resulting in a smoothed complete field. The sources of data are satellite (AVHRR) and in situ platforms (i.e., ships and buoys), and the specific datasets employed may change over time. At the marginal ice zone, sea ice concentrations are used to generate proxy SSTs.  A preliminary version of this file is produced in near-real time (1-day latency), and then replaced with a final version after 2 weeks. Note that this is the AVHRR-ONLY DOISST, available from Oct 1981, but there is a companion DOISST product that includes microwave satellite data, available from June 2002</dd><dt><span>time_coverage_end :</span></dt><dd>1981-09-01T23:59:59Z</dd><dt><span>time_coverage_start :</span></dt><dd>1981-09-01T00:00:00Z</dd><dt><span>title :</span></dt><dd>NOAA/NCEI 1/4 Degree Daily Optimum Interpolation Sea Surface Temperature (OISST) Analysis, Version 2.1 - Final</dd></dl></div></li></ul></div></div>"
+      ],
+      "text/plain": [
+       "<xarray.Dataset>\n",
+       "Dimensions:  (lat: 720, lon: 1440, time: 14372, zlev: 1)\n",
+       "Coordinates:\n",
+       "  * lat      (lat) float32 -89.88 -89.62 -89.38 -89.12 ... 89.38 89.62 89.88\n",
+       "  * lon      (lon) float32 0.125 0.375 0.625 0.875 ... 359.1 359.4 359.6 359.9\n",
+       "  * time     (time) datetime64[ns] 1981-09-01T12:00:00 ... NaT\n",
+       "  * zlev     (zlev) float32 0.0\n",
+       "Data variables:\n",
+       "    anom     (time, zlev, lat, lon) float32 dask.array<chunksize=(20, 1, 720, 1440), meta=np.ndarray>\n",
+       "    err      (time, zlev, lat, lon) float32 dask.array<chunksize=(20, 1, 720, 1440), meta=np.ndarray>\n",
+       "    ice      (time, zlev, lat, lon) float32 dask.array<chunksize=(20, 1, 720, 1440), meta=np.ndarray>\n",
+       "    sst      (time, zlev, lat, lon) float32 dask.array<chunksize=(20, 1, 720, 1440), meta=np.ndarray>\n",
+       "Attributes:\n",
+       "    Conventions:                CF-1.6, ACDD-1.3\n",
+       "    cdm_data_type:              Grid\n",
+       "    comment:                    Data was converted from NetCDF-3 to NetCDF-4 ...\n",
+       "    creator_email:              oisst-help@noaa.gov\n",
+       "    creator_url:                https://www.ncei.noaa.gov/\n",
+       "    date_created:               2020-05-08T19:05:13Z\n",
+       "    date_modified:              2020-05-08T19:05:13Z\n",
+       "    geospatial_lat_max:         90.0\n",
+       "    geospatial_lat_min:         -90.0\n",
+       "    geospatial_lat_resolution:  0.25\n",
+       "    geospatial_lat_units:       degrees_north\n",
+       "    geospatial_lon_max:         360.0\n",
+       "    geospatial_lon_min:         0.0\n",
+       "    geospatial_lon_resolution:  0.25\n",
+       "    geospatial_lon_units:       degrees_east\n",
+       "    history:                    Final file created using preliminary as first...\n",
+       "    id:                         oisst-avhrr-v02r01.19810901.nc\n",
+       "    institution:                NOAA/National Centers for Environmental Infor...\n",
+       "    instrument:                 Earth Remote Sensing Instruments > Passive Re...\n",
+       "    instrument_vocabulary:      Global Change Master Directory (GCMD) Instrum...\n",
+       "    keywords:                   Earth Science > Oceans > Ocean Temperature > ...\n",
+       "    keywords_vocabulary:        Global Change Master Directory (GCMD) Earth S...\n",
+       "    metadata_link:              https://doi.org/10.25921/RE9P-PT57\n",
+       "    naming_authority:           gov.noaa.ncei\n",
+       "    ncei_template_version:      NCEI_NetCDF_Grid_Template_v2.0\n",
+       "    platform:                   Ships, buoys, Argo floats, MetOp-A, MetOp-B\n",
+       "    platform_vocabulary:        Global Change Master Directory (GCMD) Platfor...\n",
+       "    processing_level:           NOAA Level 4\n",
+       "    product_version:            Version v02r01\n",
+       "    references:                 Reynolds, et al.(2007) Daily High-Resolution-...\n",
+       "    sensor:                     Thermometer, AVHRR\n",
+       "    source:                     ICOADS, NCEP_GTS, GSFC_ICE, NCEP_ICE, Pathfin...\n",
+       "    standard_name_vocabulary:   CF Standard Name Table (v40, 25 January 2017)\n",
+       "    summary:                    NOAAs 1/4-degree Daily Optimum Interpolation ...\n",
+       "    time_coverage_end:          1981-09-01T23:59:59Z\n",
+       "    time_coverage_start:        1981-09-01T00:00:00Z\n",
+       "    title:                      NOAA/NCEI 1/4 Degree Daily Optimum Interpolat..."
+      ]
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ds = xr.open_zarr(target_dir.name)\n",
+    "ds"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "There should be data at the beginning..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.collections.QuadMesh at 0x7ff8b0dcf100>"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ds.sst[0].plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "But not the end..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.collections.QuadMesh at 0x7ff8b0e53e80>"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ds.sst[-1].plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Postscript: Execute the full recipe\n",
+    "\n",
+    "We are now confident that our recipe works as we expect.\n",
+    "At this point we could either:\n",
+    "- Execute it all ourselves (see {doc}`execution`)\n",
+    "- Create a new recipe feedstock on Pangeo Forge\n",
+    "\n",
+    "If we wanted to execute it ourselves, one way would be to simply run the following code\n",
+    "```python\n",
+    "for input_name in recipe.iter_inputs():\n",
+    "    recipe.cache_input(input_name)\n",
+    "recipe.prepare_target()\n",
+    "for chunk in recipe.iter_chunks():\n",
+    "    recipe.store_chunk(chunk)\n",
+    "recipe.finalize_target()\n",
+    "```\n",
+    "\n",
+    "We aren't going to do this in this notebook because it would take too long.\n",
+    "\n",
+    "But hopefully now you have a better understanding of how Pangeo Forge recipes work."
+   ]
   }
  ],
  "metadata": {
diff --git a/docs/recipes.md b/docs/recipes.md
index a45789fc..4328f13b 100644
--- a/docs/recipes.md
+++ b/docs/recipes.md
@@ -12,7 +12,7 @@ you are reading matches your installed version of pangeo_forge.
 
 Recipes need a place to store data.
 The location where the final dataset produced by the recipe is stored is called the
-``Target``. Pangeo forge has a special class for this: {class}`pangeo_forge.storage.Target`
+``Target``. Pangeo forge has a special class for this: {class}`pangeo_forge.storage.FSSpecTarget`
 
 Creating a Target requires two arguments:
 - The ``fs`` argument is an [fsspec](https://filesystem-spec.readthedocs.io/en/latest/)
@@ -20,20 +20,17 @@ Creating a Target requires two arguments:
   [built in](https://filesystem-spec.readthedocs.io/en/latest/api.html#built-in-implementations)
   and [third party](https://filesystem-spec.readthedocs.io/en/latest/api.html#other-known-implementations)
   implementations.
-- The `path` argument specifies the specific path where the data should be stored.
+- The `root_path` argument specifies the specific path where the data should be stored.
 
 For example, creating a storage target for AWS S3 might look like this:
 ```{code-block} python
 import s3fs
 fs = s3fs.S3FileSystem(key="MY_AWS_KEY", secret="MY_AWS_SECRET")
 target_path = "pangeo-forge-bucket/my-dataset-v1.zarr"
-target = Target(fs=fs, path=target_path)
+target = FSSpecTarget(fs=fs, root_path=target_path)
 ```
 
-Temporary data can be stored in an {class}`pangeo_forge.storage.InputCache` object.
-``InputCache`` is similar to ``Target``, but instead of specifying a ``path``,
-you specify ``prefix``.
-
+Temporary data is recommended to use a {class}`pangeo_forge.storage.CacheFSSpecTarget` object.
 
 ## The Base Recipe Class
 
@@ -45,9 +42,6 @@ recipe = Recipe(option1='foo', option2=)
 All recipes follow the same basic steps.
 
 
-
-
-
 ## Specific Recipe Classes
 
 ```{eval-rst}
diff --git a/pangeo_forge/recipe.py b/pangeo_forge/recipe.py
index ebe82dda..55cb88cf 100644
--- a/pangeo_forge/recipe.py
+++ b/pangeo_forge/recipe.py
@@ -138,7 +138,9 @@ class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     :param consolidate_zarr: Whether to consolidate the resulting Zarr dataset.
     :param xarray_open_kwargs: Extra options for opening the inputs with Xarray.
     :param xarray_concat_kwargs: Extra options to pass to Xarray when concatenating
-       the inputs to form a chunk.
+      the inputs to form a chunk.
+    :param delete_input_encoding: Whether to remove Xarray encoding from variables
+      in the input dataset
     """
 
     input_urls: Iterable[str] = field(repr=False)
@@ -151,6 +153,7 @@ class NetCDFtoZarrSequentialRecipe(BaseRecipe):
     consolidate_zarr: bool = True
     xarray_open_kwargs: dict = field(default_factory=dict)
     xarray_concat_kwargs: dict = field(default_factory=dict)
+    delete_input_encoding: bool = True
 
     def __post_init__(self):
         self._chunks_inputs = {
@@ -245,6 +248,11 @@ def open_input(self, fname: str):
             # explicitly load into memory
             ds = ds.load()
         ds = fix_scalar_attr_encoding(ds)
+
+        if self.delete_input_encoding:
+            for var in ds.variables:
+                ds[var].encoding = {}
+
         logger.debug(f"{ds}")
         return ds
 
@@ -255,8 +263,6 @@ def open_chunk(self, chunk_key):
         # CONCAT DELETES ENCODING!!!
         # OR NO IT DOESN'T! Not in the latest version of xarray?
         ds = xr.concat(dsets, self.sequence_dim, **self.xarray_concat_kwargs)
-        for var in ds.variables:
-            ds[var].encoding = {}
         logger.debug(f"{ds}")
 
         # TODO: maybe do some chunking here?

From 63e2297366584702853368fbe8918a2e3c7890ee Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 23:46:39 -0500
Subject: [PATCH 31/34] last commit of the night

---
 docs/_static/custom.css                       | 27 +++++++
 docs/api.md                                   | 24 ++++++
 docs/execution.md                             | 81 +++++++++++++++++++
 docs/index.md                                 |  5 +-
 docs/recipes.md                               | 41 +++++++---
 docs/tutorials/index.md                       |  9 +++
 .../netcdf_zarr_sequential.ipynb}             | 17 +++-
 pangeo_forge/executors.py                     |  3 +
 tests/test_executors.py                       |  3 +-
 9 files changed, 194 insertions(+), 16 deletions(-)
 create mode 100644 docs/_static/custom.css
 create mode 100644 docs/tutorials/index.md
 rename docs/{recipe_tutorial.ipynb => tutorials/netcdf_zarr_sequential.ipynb} (99%)
 create mode 100644 pangeo_forge/executors.py

diff --git a/docs/_static/custom.css b/docs/_static/custom.css
new file mode 100644
index 00000000..9b2ac1b6
--- /dev/null
+++ b/docs/_static/custom.css
@@ -0,0 +1,27 @@
+/* Put your custom CSS here */
+@import url('http://fonts.cdnfonts.com/css/panton-black-caps');
+
+h1 {
+    font-family: "Panton Black Caps", sans-serif;
+    color: #003B71 !important;
+}
+
+h2 {
+    font-family: "Panton Light Caps", sans-serif;
+    color: #003B71 !important;
+}
+
+a {
+    color: #5eb130 !important;
+}
+
+
+
+/* Fixing up some pygments and code-styling CSS for accessibility */
+code { font-size: 100%; color: #e50051; }
+pre { font-family: monospace; }
+
+/* .highlight { font-size: 125%; } */
+.highlight .c1 { color: #e50051; }
+.highlight .si { color: #e50051; }
+.highlight .nn { color: #e50051; }
diff --git a/docs/api.md b/docs/api.md
index 667283b2..1ba199ae 100644
--- a/docs/api.md
+++ b/docs/api.md
@@ -8,9 +8,16 @@
     :members:
 ```
 
+```{eval-rst}
+.. autoclass:: pangeo_forge.storage.FlatFSSpecTarget
+    :members:
+    :show-inheritance:
+```
+
 ```{eval-rst}
 .. autoclass:: pangeo_forge.storage.CacheFSSpecTarget
     :members:
+    :show-inheritance:
 ```
 
 ## Recipes
@@ -24,3 +31,20 @@
 .. autoclass:: pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe
     :show-inheritance:
 ```
+
+## Excutors
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.executors.PythonPipelineExecutor
+    :members:
+```
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.executors.DaskPipelineExecutor
+    :members:
+```
+
+```{eval-rst}
+.. autoclass:: pangeo_forge.executors.PrefectPipelineExecutor
+    :members:
+```
diff --git a/docs/execution.md b/docs/execution.md
index 7bb5b258..0ec71286 100644
--- a/docs/execution.md
+++ b/docs/execution.md
@@ -1 +1,82 @@
 # Recipe Execution
+
+There are many different types of Pangeo Forge recipes.
+However, **all recipes are executed the same way**!
+This is a key part of the Pangeo Forge design.
+
+Once you have created a recipe object (see {doc}`recipes`) you have two
+options for executing it. In the subsequent code, we will assume that a
+recipe has already been initialized in the variable `recipe`.
+
+## Manual Execution
+
+A recipe can be executed manually, step by step, in serial, from a notebook
+or interactive interpreter. The ability to manually step through a recipe
+is very important for developing and debugging complex recipes.
+There are four stages of recipe execution.
+
+### Stage 1: Cache Inputs
+
+Recipes may define files that have to be cached locally before the subsequent
+steps can proceed. The common use case here is for files that have to be
+extracted from a slow FTP server. Here is how to cache the inputs.
+
+```{code-block} python
+for input_name in recipe.iter_inputs():
+    recipe.cache_input(input_name)
+```
+
+If the recipe doesn't do input caching, nothing will happen here.
+
+### Stage 2: Prepare Target
+
+Once the inputs have been cached, we can get the target ready.
+Preparing the target for writing is done as follows:
+
+```{code-block} python
+recipe.prepare_target()
+```
+
+For example, for Zarr targets, this sets up the Zarr group with the necessary
+arrays and metadata.
+
+### Stage 3: Store Chunks
+
+This is the step where the bulk of the work happens.
+
+```{code-block} python
+for chunk in recipe.iter_chunks():
+    recipe.store_chunk(chunk)
+```
+
+### Stage 4: Finalize Target
+
+If there is any cleanup or consolidation to be done, it happens here.
+
+```{code-block} python
+recipe.finalize_target()
+```
+
+For example, consolidating Zarr metadta happens in the finalize step.
+
+## Execution by Executors
+
+Very large recipes cannot feasibly be executed this way.
+To support distributed parallel execution, Pangeo Forge borrows the
+[Executors framework from Rechunker](https://rechunker.readthedocs.io/en/latest/executors.html).
+
+There are currently three executors implemented.
+- {class}`pangeo_forge.executors.PythonPipelineExecutor`: a reference executor
+  using simple python
+- {class}`pangeo_forge.executors.DaskPipelineExecutor`: distributed executor using Dask
+- {class}`pangeo_forge.executors.PrefectPipelineExecutor`: distributed executor using Prefect
+
+To use an executor, the recipe must first be transformed into a `Pipeline` object.
+The full process looks like this:
+
+```{code-block} python
+pipeline = recipe.to_pipelines()
+executor = PrefectPipelineExecutor()
+plan = executor.pipelines_to_plan(pipeline)
+executor.execute_plan(plan)  # actually runs the recipe
+```
diff --git a/docs/index.md b/docs/index.md
index 319199e4..7e475018 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -14,7 +14,8 @@ The most important concept in Pangeo Forge is a ``recipe``.
 A recipe defines how to transform data in one format / location into another format / location.
 The primary way people contribute to Pangeo Forge is by writing / maintaining recipes.
 Recipes developed by the community are stored in GitHub repositories.
-For information about how to write a recipe, see {doc}`recipes`.
+For information about how recipes work see {doc}`recipes`.
+The {doc}`tutorials/index` provide deep dives into how to develop and debug Pangeo Forge recipes.
 
 ## Recipe Execution
 
@@ -35,7 +36,7 @@ For more information, see {doc}`bakeries`.
 :caption: Contents
 
 recipes
-recipe_tutorial
+tutorials/index
 execution
 bakeries
 contribute
diff --git a/docs/recipes.md b/docs/recipes.md
index 4328f13b..527df8b8 100644
--- a/docs/recipes.md
+++ b/docs/recipes.md
@@ -32,20 +32,43 @@ target = FSSpecTarget(fs=fs, root_path=target_path)
 
 Temporary data is recommended to use a {class}`pangeo_forge.storage.CacheFSSpecTarget` object.
 
-## The Base Recipe Class
+## The Recipe Object
+
+You define a recipe by instantiating a class that inherits from {class}`pangeo_forge.recipe.BaseRecipe`.
+The `pangeo_forge` package includes several pre-defined Recipe classes which
+cover common scenarios. You can also define your own Recipe class.
+
+For a the common scenario of assembling many NetCDF files into a single Zarr
+group, we use {class}`pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe`.
+Initializing a recipe looks something like this.
 
-A recipe is initialized from a recipe class.
 ```{code-block} python
-recipe = Recipe(option1='foo', option2=)
+from pangeo_forge.recipes import NetCDFtoZarrSequentialRecipe
+input_urls = [...]  # build a list of inputs
+recipe = NetCDFtoZarrSequentialRecipe(
+    input_urls=input_urls,
+    sequence_dim="time"
+)
 ```
 
-All recipes follow the same basic steps.
+There are many other options we can pass, all covered in the [API documentation](api).
+For a deeper dive on how to pick these options and what they mean, check out the
+tutorial: {doc}`tutorials/netcdf_zarr_sequential`.
 
+Your recipe will also need storage.
+If you have already defined a `Target` object (as in the the [Storage section](#storage)),
+then you can either assign it when you initialize the recipe or later, e.g.
 
-## Specific Recipe Classes
+```{code-block} python
+recipe.target = FSSpecTarget(fs=fs, root_path=target_path)
+```
 
-```{eval-rst}
-.. autoclass:: pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe
-    :show-inheritance:
-    :noindex:
+This particular class of recipe also requires a cache, a place to store temporary
+files. We can create one as follows.
+
+```{code-block} python
+recipe.input_cache = CacheFSSpecTarget(fs=fs, root_path=cache_path)
 ```
+
+Once your recipe is defined and has its targets assigned, you're ready to
+move on to {doc}`execution`.
diff --git a/docs/tutorials/index.md b/docs/tutorials/index.md
new file mode 100644
index 00000000..ef34442c
--- /dev/null
+++ b/docs/tutorials/index.md
@@ -0,0 +1,9 @@
+# Recipe Tutorials
+
+These tutorials are deep dives into how to develop and debug Pangeo Forge recipes.
+
+```{toctree}
+:maxdepth: 1
+
+netcdf_zarr_sequential
+```
diff --git a/docs/recipe_tutorial.ipynb b/docs/tutorials/netcdf_zarr_sequential.ipynb
similarity index 99%
rename from docs/recipe_tutorial.ipynb
rename to docs/tutorials/netcdf_zarr_sequential.ipynb
index 3e7b2cfe..4d501d87 100644
--- a/docs/recipe_tutorial.ipynb
+++ b/docs/tutorials/netcdf_zarr_sequential.ipynb
@@ -4,9 +4,11 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Recipe Tutorial\n",
+    "# NetCDF Zarr Sequential Recipe\n",
     "\n",
-    "This tutorial describes how to create a recipe from scratch.\n"
+    "This tutorial describes how to create a recipe from scratch.\n",
+    "The source data is a sequence of NetCDF files accessed via HTTP.\n",
+    "The target is a Zarr store."
    ]
   },
   {
@@ -567,7 +569,7 @@
     "For our first recipe, we will want to use a pre-defined Recipe class from Pangeo\n",
     "Forge.\n",
     "\n",
-    "By examining the {doc}`recipes` documentation page, we see that our scenario is\n",
+    "By examining the {doc}`../recipes` documentation page, we see that our scenario is\n",
     "a good case for the {class}`pangeo_forge.recipe.NetCDFtoZarrSequentialRecipe`\n",
     "class. Let's examine its documentation string in our notebook.\n"
    ]
@@ -3602,7 +3604,7 @@
     "\n",
     "We are now confident that our recipe works as we expect.\n",
     "At this point we could either:\n",
-    "- Execute it all ourselves (see {doc}`execution`)\n",
+    "- Execute it all ourselves (see {doc}`../execution`)\n",
     "- Create a new recipe feedstock on Pangeo Forge\n",
     "\n",
     "If we wanted to execute it ourselves, one way would be to simply run the following code\n",
@@ -3619,6 +3621,13 @@
     "\n",
     "But hopefully now you have a better understanding of how Pangeo Forge recipes work."
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
diff --git a/pangeo_forge/executors.py b/pangeo_forge/executors.py
new file mode 100644
index 00000000..bca10b1b
--- /dev/null
+++ b/pangeo_forge/executors.py
@@ -0,0 +1,3 @@
+from rechunker.executors import DaskPipelineExecutor  # noqa: F401
+from rechunker.executors import PrefectPipelineExecutor  # noqa: F401
+from rechunker.executors import PythonPipelineExecutor  # noqa: F401
diff --git a/tests/test_executors.py b/tests/test_executors.py
index b21ad56a..6294809a 100644
--- a/tests/test_executors.py
+++ b/tests/test_executors.py
@@ -1,6 +1,7 @@
 import pytest
 import xarray as xr
-from rechunker.executors import (
+
+from pangeo_forge.executors import (
     DaskPipelineExecutor,
     PrefectPipelineExecutor,
     PythonPipelineExecutor,

From e0c97b009555d9d8fcdeeef5264acee78b993256 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 23:48:27 -0500
Subject: [PATCH 32/34] update doc requirements

---
 docs/requirements.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/docs/requirements.txt b/docs/requirements.txt
index 8a7fbe2f..88595554 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -1,3 +1,5 @@
 sphinx
-sphinx-pangeo-theme
+sphinx_book_theme
 myst-parser
+myst-nb
+sphinx-copybutton

From 382663d4f6b963fd02553f00c94956f87aa35be8 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Thu, 21 Jan 2021 23:57:40 -0500
Subject: [PATCH 33/34] use rechunker from github

---
 requirements.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements.txt b/requirements.txt
index 83520897..b8770e9b 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,7 +2,7 @@ setuptools
 click
 dask
 distributed
-rechunker
+-e git://github.com/rabernat/rechunker.git@refactor-executors#egg=rechunker
 xarray >= 0.16.2
 zarr >= 2.6.0
 fsspec[http]

From 57304e86ff31e5e4c56cc8dd5f27f79a8e9b47f4 Mon Sep 17 00:00:00 2001
From: Ryan Abernathey <ryan.abernathey@gmail.com>
Date: Fri, 22 Jan 2021 07:34:20 -0500
Subject: [PATCH 34/34] fix requirements

---
 docs/requirements.txt | 1 +
 requirements.txt      | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/docs/requirements.txt b/docs/requirements.txt
index 88595554..51875dbb 100644
--- a/docs/requirements.txt
+++ b/docs/requirements.txt
@@ -1,4 +1,5 @@
 sphinx
+pydata-sphinx-theme>=0.4.2
 sphinx_book_theme
 myst-parser
 myst-nb
diff --git a/requirements.txt b/requirements.txt
index b8770e9b..83520897 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -2,7 +2,7 @@ setuptools
 click
 dask
 distributed
--e git://github.com/rabernat/rechunker.git@refactor-executors#egg=rechunker
+rechunker
 xarray >= 0.16.2
 zarr >= 2.6.0
 fsspec[http]