Merge pull request #33 from mdekstrand/tweak/ruff

Format and lint with ruff

.github/workflows/package.yml

@@ -7,6 +7,10 @@ on:
     types: [created,published]
   pull_request:
 
+concurrency:
+  group: test-${{github.ref}}
+  cancel-in-progress: true
+
 jobs:
   lint:
     name: Check Source Style
@@ -17,38 +21,16 @@ jobs:
       with:
         fetch-depth: 0
 
-    - name: Set up Python ${{matrix.python}}
-      uses: actions/setup-python@v2
-      with:
-        python-version: 3.8
-
-    - name: Prep Pip caching
-      id: pip-cache
-      run: |
-        echo "::set-output name=dir::$(pip cache dir)"
-      shell: bash
-
-    - name: Cache Pip wheels
-      uses: actions/cache@v1
-      with:
-        path: ${{ steps.pip-cache.outputs.dir }}
-        key: py38-lint-pip-${{ hashFiles('*.egg-info/requires.txt')}}
-
-    - name: Install environment
-      run: |
-        pip install -U flit
-
-    - name: Install package
-      run: |
-        flit install --pth-file --extras dev
-
-    - name: Run lint
+    - name: Check source code
       run: |
-        # Flake8 problem matcher & transform regex from https://github.com/TrueBrain/actions-flake8
-        echo "::add-matcher::.github/flake8-matcher.json"
-        set -o pipefail
-        flake8 |sed -r 's/: ([^W][0-9][0-9][0-9])/: error: \1/;s/: (W[0-9][0-9][0-9])/: warning: \1/'
-        echo "::remove-matcher owner=flake8::"
+        status=0
+        if ! pipx run ruff check --output-format=github binpickle; then
+          status=1
+        fi
+        pipx run ruff format --diff binpickle
+        if [ $status -ne 0 ]; then
+          exit $status
+        fi
 
   test:
     name: Test with Python ${{matrix.python}} on ${{matrix.platform}}

binpickle/__init__.py

@@ -2,7 +2,7 @@
 Optimized format for pickling binary data.
 """
 
-__version__ = '0.4.0'
+__version__ = "0.4.0"
 
-from .write import dump, BinPickler    # noqa: F401
+from .write import dump, BinPickler  # noqa: F401
 from .read import load, BinPickleFile  # noqa: F401

binpickle/codecs/__init__.py

@@ -78,13 +78,13 @@ def get_codec(name, config):
     if name is None:
         return Null()
     elif name in CODECS:
-        _log.debug('configuring %s: %s', name, config)
+        _log.debug("configuring %s: %s", name, config)
         return CODECS[name](**config)
     else:
-        raise ValueError(f'unknown codec {name}')
+        raise ValueError(f"unknown codec {name}")
 
 
-from .chain import Chain   # noqa: E402
+from .chain import Chain  # noqa: E402
 
 register(Null)
 register(Chain)

binpickle/codecs/blosc.py

@@ -11,7 +11,7 @@
 
 def _split_blocks(buf, blocksize):
     if buf.itemsize > 1:
-        buf = buf.cast('B')
+        buf = buf.cast("B")
     length = buf.nbytes
     chunks = []
     for start in range(0, length, blocksize):
@@ -21,7 +21,7 @@ def _split_blocks(buf, blocksize):
         chunks.append(buf[start:end])
 
     if not chunks:
-        chunks.append(memoryview(b''))
+        chunks.append(memoryview(b""))
     return chunks
 
 
@@ -30,13 +30,12 @@ class Blosc(Codec):
     Blosc codec.
     """
 
-    NAME = 'blosc'
-    AVAILABLE = find_spec('blosc') is not None
+    NAME = "blosc"
+    AVAILABLE = find_spec("blosc") is not None
 
-    def __init__(self, name='blosclz', level=9,
-                 shuffle=1, blocksize=DEFAULT_BLOCKSIZE):
+    def __init__(self, name="blosclz", level=9, shuffle=1, blocksize=DEFAULT_BLOCKSIZE):
         if not self.AVAILABLE:
-            raise ImportError('blosc is not available')
+            raise ImportError("blosc is not available")
         self.name = name
         self.level = level
         self.shuffle = shuffle
@@ -45,22 +44,28 @@ def __init__(self, name='blosclz', level=9,
     def encode_to(self, buf, out):
         # We have to encode by chunks
         import blosc
+
         pack = msgpack.Packer()
         mv = memoryview(buf)
-        _log.debug('encoding %d bytes (itemsize=%d, format=%s)',
-                   mv.nbytes, mv.itemsize, mv.format)
-        _log.debug('splitting with block size %d', self.blocksize)
+        _log.debug("encoding %d bytes (itemsize=%d, format=%s)", mv.nbytes, mv.itemsize, mv.format)
+        _log.debug("splitting with block size %d", self.blocksize)
         blocks = _split_blocks(mv, self.blocksize)
         out.write(pack.pack_array_header(len(blocks)))
         for block in blocks:
             assert block.nbytes <= self.blocksize
-            comp = blosc.compress(block, cname=self.name, clevel=self.level,
-                                  shuffle=self.shuffle, typesize=mv.itemsize)
+            comp = blosc.compress(
+                block,
+                cname=self.name,
+                clevel=self.level,
+                shuffle=self.shuffle,
+                typesize=mv.itemsize,
+            )
             out.write(pack.pack(comp))
             block.release()
 
     def decode_to(self, buf, out):
         import blosc
+
         blocks = msgpack.unpackb(buf, use_list=True)
         pos = 0
         for block in blocks:
@@ -77,8 +82,4 @@ def decode_to(self, buf, out):
             del out[pos:]
 
     def config(self):
-        return {
-            'name': self.name,
-            'level': self.level,
-            'shuffle': self.shuffle
-        }
+        return {"name": self.name, "level": self.level, "shuffle": self.shuffle}

binpickle/codecs/chain.py

@@ -7,7 +7,8 @@ class Chain(Codec):
     Codec that chains together other codecs in sequence.  The codecs are applied
     in the provided order for encoding, and reverse order for decoding.
     """
-    NAME = 'chain'
+
+    NAME = "chain"
 
     def __init__(self, codecs=()):
         self.codecs = [make_codec(c, list_is_tuple=True) for c in codecs]
@@ -31,6 +32,4 @@ def decode_to(self, buf, out):
         out[:] = self.decode(buf)
 
     def config(self):
-        return {
-            'codecs': [(c.NAME, c.config()) for c in self.codecs]
-        }
+        return {"codecs": [(c.NAME, c.config()) for c in self.codecs]}

binpickle/codecs/gz.py

@@ -8,7 +8,7 @@ class GZ(Codec):
     Zlib (gzip-compatible) codec.
     """
 
-    NAME = 'gz'
+    NAME = "gz"
 
     def __init__(self, level=9):
         self.level = level
@@ -27,6 +27,4 @@ def decode_to(self, buf, out):
         out[:] = self.decode(buf)
 
     def config(self):
-        return {
-            'level': self.level
-        }
+        return {"level": self.level}

binpickle/codecs/null.py

@@ -5,7 +5,8 @@ class Null(Codec):
     """
     Null codec (passthrough).
     """
-    NAME = 'null'
+
+    NAME = "null"
 
     def encode(self, buf):
         return buf

binpickle/codecs/numcodecs.py

@@ -6,6 +6,7 @@ def is_numcodec(codec):
     "Test whether a codec is a NumCodecs codec."
     if NC.AVAILABLE:
         import numcodecs
+
         return isinstance(codec, numcodecs.abc.Codec)
     else:
         return False  # if numcodecs aren't available, it can't be one
@@ -15,12 +16,14 @@ class NC(Codec):
     """
     NumCodec wrapper.
     """
-    NAME = 'numcodec'
-    AVAILABLE = find_spec('numcodecs') is not None
+
+    NAME = "numcodec"
+    AVAILABLE = find_spec("numcodecs") is not None
 
     def __init__(self, codec=None, **kwargs):
         if codec is None:
             import numcodecs
+
             self.codec = numcodecs.get_codec(kwargs)
         else:
             self.codec = codec

binpickle/format.py

@@ -5,10 +5,10 @@
 import struct
 from typing import NamedTuple
 
-MAGIC = b'BPCK'
+MAGIC = b"BPCK"
 VERSION = 1
-HEADER_FORMAT = struct.Struct('!4sHHq')
-TRAILER_FORMAT = struct.Struct('!QLL')
+HEADER_FORMAT = struct.Struct("!4sHHq")
+TRAILER_FORMAT = struct.Struct("!QLL")
 
 
 class FileHeader(NamedTuple):
@@ -21,6 +21,7 @@ class FileHeader(NamedTuple):
     3. File length (8 bytes, big-endian).  Length is signed; if the file length is not known,
        this field is set to -1.
     """
+
     version: int = VERSION
     "The NumPy file version."
     length: int = -1
@@ -35,11 +36,11 @@ def decode(cls, buf, *, verify=True):
         "Decode a file header from bytes."
         m, v, pad, off = HEADER_FORMAT.unpack(buf)
         if verify and m != MAGIC:
-            raise ValueError('invalid magic {}'.format(m))
+            raise ValueError("invalid magic {}".format(m))
         if verify and v != VERSION:
-            raise ValueError('invalid version {}'.format(v))
+            raise ValueError("invalid version {}".format(v))
         if verify and pad != 0:
-            raise ValueError('invalid padding')
+            raise ValueError("invalid padding")
         return cls(v, off)
 
     @classmethod
@@ -52,7 +53,7 @@ def trailer_pos(self):
         if self.length >= HEADER_FORMAT.size + TRAILER_FORMAT.size:
             return self.length - TRAILER_FORMAT.size
         elif self.length > 0:
-            raise ValueError('file size {} not enough for BinPickle'.format(self.length))
+            raise ValueError("file size {} not enough for BinPickle".format(self.length))
         else:
             return None  # We do not know the file size
 
@@ -79,14 +80,15 @@ def encode(self):
     @classmethod
     def decode(cls, buf, *, verify=True):
         "Decode a file trailer from bytes."
-        o, l, c = TRAILER_FORMAT.unpack(buf)
-        return cls(o, l, c)
+        off, len, ck = TRAILER_FORMAT.unpack(buf)
+        return cls(off, len, ck)
 
 
 class IndexEntry(NamedTuple):
     """
     Index entry for a buffer in the BinPickle index.
     """
+
     offset: int
     "The position in the file where the buffer begins (bytes)."
     enc_length: int