Improve dataframe conversion.

Merge branch 'release/0.7'

CITATION.cff

@@ -0,0 +1,18 @@
+cff-version: 1.2.0
+message: "If you use this software, please cite it as below."
+authors:
+  - family-names: "Ramos-Carreño"
+    given-names: "Carlos"
+    orcid: "https://orcid.org/0000-0003-2566-7058"
+    affiliation: "Universidad Autónoma de Madrid"
+    email: [email protected]
+title: "rdata: Read R datasets from Python"
+date-released: 2022-03-24
+url: "https://github.com/vnmabus/rdata"
+license: MIT
+keywords:
+  - rdata
+  - Python
+  - R
+  - parser
+  - conversion

MANIFEST.in

@@ -1,5 +1,5 @@
 include MANIFEST.in
-include VERSION
+include rdata/VERSION
 include LICENSE
 include rdata/py.typed
 include *.txt

rdata/VERSION

@@ -0,0 +1 @@
+0.7

rdata/__init__.py

@@ -1,3 +1,5 @@
+"""rdata: Read R datasets from Python."""
+import errno as _errno
 import os as _os
 import pathlib as _pathlib
 
@@ -13,3 +15,15 @@ def _get_test_data_path() -> _pathlib.Path:
 Path of the test data.
 
 """
+
+try:
+    with open(
+        _pathlib.Path(_os.path.dirname(__file__)) / 'VERSION',
+        'r',
+    ) as version_file:
+        __version__ = version_file.read().strip()
+except IOError as e:
+    if e.errno != _errno.ENOENT:
+        raise
+
+    __version__ = "0.0"

rdata/conversion/_conversion.py

@@ -6,7 +6,6 @@
     Any,
     Callable,
     ChainMap,
-    Hashable,
     List,
     Mapping,
     MutableMapping,
@@ -355,12 +354,16 @@ def convert_array(
 
     dimnames = attrs.get('dimnames')
     if dimnames:
-        dimension_names = ["dim_" + str(i) for i, _ in enumerate(dimnames)]
-        coords: Mapping[Hashable, Any] = {
-            dimension_names[i]: d
-            for i, d in enumerate(dimnames)
-            if d is not None
-        }
+        if isinstance(dimnames, Mapping):
+            dimension_names = list(dimnames.keys())
+            coords = dimnames
+        else:
+            dimension_names = [f"dim_{i}" for i, _ in enumerate(dimnames)]
+            coords = {
+                dimension_names[i]: d
+                for i, d in enumerate(dimnames)
+                if d is not None
+            }
 
         value = xarray.DataArray(value, dims=dimension_names, coords=coords)
 

rdata/tests/test_rdata.py

@@ -8,8 +8,8 @@
 
 import numpy as np
 import pandas as pd
-
 import rdata
+import xarray
 
 TESTDATA_PATH = rdata.TESTDATA_PATH
 
@@ -93,6 +93,74 @@ def test_matrix(self) -> None:
             ]),
         })
 
+    def test_named_matrix(self) -> None:
+        """Test that a named matrix can be parsed."""
+        parsed = rdata.parser.parse_file(
+            TESTDATA_PATH / "test_named_matrix.rda",
+        )
+        converted = rdata.conversion.convert(parsed)
+        reference = xarray.DataArray(
+            [
+                [1.0, 2.0, 3.0],
+                [4.0, 5.0, 6.0],
+            ],
+            dims=["dim_0", "dim_1"],
+            coords={
+                "dim_0": ["dim0_0", "dim0_1"],
+                "dim_1": ["dim1_0", "dim1_1", "dim1_2"],
+            },
+        )
+
+        xarray.testing.assert_identical(
+            converted["test_named_matrix"],
+            reference,
+        )
+
+    def test_half_named_matrix(self) -> None:
+        """Test that a named matrix with no name for a dim can be parsed."""
+        parsed = rdata.parser.parse_file(
+            TESTDATA_PATH / "test_half_named_matrix.rda",
+        )
+        converted = rdata.conversion.convert(parsed)
+        reference = xarray.DataArray(
+            [
+                [1.0, 2.0, 3.0],
+                [4.0, 5.0, 6.0],
+            ],
+            dims=["dim_0", "dim_1"],
+            coords={
+                "dim_0": ["dim0_0", "dim0_1"],
+            },
+        )
+
+        xarray.testing.assert_identical(
+            converted["test_half_named_matrix"],
+            reference,
+        )
+
+    def test_full_named_matrix(self) -> None:
+        """Test that a named matrix with dim names can be parsed."""
+        parsed = rdata.parser.parse_file(
+            TESTDATA_PATH / "test_full_named_matrix.rda",
+        )
+        converted = rdata.conversion.convert(parsed)
+        reference = xarray.DataArray(
+            [
+                [1.0, 2.0, 3.0],
+                [4.0, 5.0, 6.0],
+            ],
+            dims=["my_dim_0", "my_dim_1"],
+            coords={
+                "my_dim_0": ["dim0_0", "dim0_1"],
+                "my_dim_1": ["dim1_0", "dim1_1", "dim1_2"],
+            },
+        )
+
+        xarray.testing.assert_identical(
+            converted["test_full_named_matrix"],
+            reference,
+        )
+
     def test_list(self) -> None:
         """Test that list can be parsed."""
         parsed = rdata.parser.parse_file(TESTDATA_PATH / "test_list.rda")

setup.py

@@ -7,6 +7,7 @@
 language or its libraries, and thus it is released under a MIT license.
 """
 import os
+import pathlib
 import sys
 
 from setuptools import find_packages, setup
@@ -16,44 +17,51 @@
 
 DOCLINES = (__doc__ or '').split("\n")
 
-with open(os.path.join(os.path.dirname(__file__),
-                       'VERSION'), 'r') as version_file:
+with open(
+    pathlib.Path(os.path.dirname(__file__)) / 'rdata' / 'VERSION',
+    'r',
+) as version_file:
     version = version_file.read().strip()
 
-setup(name='rdata',
-      version=version,
-      description=DOCLINES[1],
-      long_description="\n".join(DOCLINES[3:]),
-      url='https://github.com/vnmabus/rdata',
-      author='Carlos Ramos Carreño',
-      author_email='[email protected]',
-      include_package_data=True,
-      platforms=['any'],
-      license='MIT',
-      packages=find_packages(),
-      python_requires='>=3.7, <4',
-      classifiers=[
-          'Development Status :: 4 - Beta',
-          'Intended Audience :: Developers',
-          'Intended Audience :: Science/Research',
-          'License :: OSI Approved :: MIT License',
-          'Natural Language :: English',
-          'Operating System :: OS Independent',
-          'Programming Language :: Python :: 3',
-          'Programming Language :: Python :: 3.6',
-          'Programming Language :: Python :: 3.7',
-          'Programming Language :: Python :: 3.8',
-          'Topic :: Scientific/Engineering :: Mathematics',
-          'Topic :: Software Development :: Libraries :: Python Modules',
-          'Typing :: Typed',
-      ],
-      keywords=['rdata', 'r', 'dataset'],
-      install_requires=['numpy',
-                        'xarray',
-                        'pandas'],
-      setup_requires=pytest_runner,
-      tests_require=['pytest-cov',
-                     'numpy>=1.14'  # The printing format for numpy changes
-                     ],
-      test_suite='rdata.tests',
-      zip_safe=False)
+setup(
+    name='rdata',
+    version=version,
+    description=DOCLINES[1],
+    long_description="\n".join(DOCLINES[3:]),
+    url='https://github.com/vnmabus/rdata',
+    author='Carlos Ramos Carreño',
+    author_email='[email protected]',
+    include_package_data=True,
+    platforms=['any'],
+    license='MIT',
+    packages=find_packages(),
+    python_requires='>=3.7, <4',
+    classifiers=[
+        'Development Status :: 4 - Beta',
+        'Intended Audience :: Developers',
+        'Intended Audience :: Science/Research',
+        'License :: OSI Approved :: MIT License',
+        'Natural Language :: English',
+        'Operating System :: OS Independent',
+        'Programming Language :: Python :: 3',
+        'Programming Language :: Python :: 3.6',
+        'Programming Language :: Python :: 3.7',
+        'Programming Language :: Python :: 3.8',
+        'Topic :: Scientific/Engineering :: Mathematics',
+        'Topic :: Software Development :: Libraries :: Python Modules',
+        'Typing :: Typed',
+    ],
+    keywords=['rdata', 'r', 'dataset'],
+    install_requires=[
+        'numpy',
+        'xarray',
+        'pandas',
+    ],
+    setup_requires=pytest_runner,
+    tests_require=[
+        'pytest-cov',
+        'numpy>=1.14',  # The printing format for numpy changes
+    ],
+    test_suite='rdata.tests',
+    zip_safe=False,
+)