python-dwca-reader 0.16.0__tar.gz → 0.16.1__tar.gz

{python-dwca-reader-0.16.0 → python-dwca-reader-0.16.1}/CHANGES.txt

@@ -1,3 +1,10 @@
+v0.16.1 (2024-07-08)
+--------------------
+
+- Added official support for Python 3.12
+- Proper error message when trying to use an unsupported combination of Pandas option
+and archives with default values (issue #106).
+
 v0.16.0 (2023-11-13)
 --------------------
 

{python-dwca-reader-0.16.0 → python-dwca-reader-0.16.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: python-dwca-reader
-Version: 0.16.0
+Version: 0.16.1
 Summary: A simple Python package to read Darwin Core Archive (DwC-A) files.
 Home-page: https://github.com/BelgianBiodiversityPlatform/python-dwca-reader
 Author: Nicolas Noé - Belgian Biodiversity Platform

{python-dwca-reader-0.16.0 → python-dwca-reader-0.16.1}/dwca/helpers.py

@@ -6,7 +6,7 @@ def remove_tree(path, retries=3, sleep=0.1):
     for i in range(retries):
         try:
             rmtree(path, False)
-        except WindowsError:
+        except OSError:
             time.sleep(sleep)
         else:
             break

{python-dwca-reader-0.16.0 → python-dwca-reader-0.16.1}/dwca/read.py

@@ -10,6 +10,8 @@ from tempfile import mkdtemp
 from typing import List, Optional, Dict, Any, IO, Tuple
 from xml.etree.ElementTree import Element
 
+from pandas.io.parsers import TextFileReader
+
 import dwca.vendor
 from dwca.descriptors import ArchiveDescriptor, DataFileDescriptor, shorten_term
 from dwca.exceptions import RowNotFound, InvalidArchive, InvalidSimpleArchive, NotADataFile
@@ -186,7 +188,9 @@ class DwCAReader(object):
         .. note::
 
             Default values of Darwin Core Archive are supported: A column will be added to the DataFrame if a term has
-            a default value in the Metafile (but no corresponding column in the CSV Data File).
+            a default value in the Metafile (but no corresponding column in the CSV Data File). This is unfortunately
+            not supported in case the value returned by `pandas.read_csv()` is a `TextFileReader` (e.g. when using
+            `chunksize` or `iterator=True`).
         """
         datafile_descriptor = self.get_descriptor_for(relative_path)  # type: DataFileDescriptor
 
@@ -200,15 +204,25 @@ class DwCAReader(object):
         kwargs['header'] = None
         kwargs['names'] = datafile_descriptor.short_headers
 
-        df = read_csv(self.absolute_temporary_path(relative_path), **kwargs)
+        df_or_textreader = read_csv(self.absolute_temporary_path(relative_path), **kwargs)
 
         # Add a column for default values, if present in the file descriptor
         for field in datafile_descriptor.fields:
             field_default_value = field['default']
             if field_default_value is not None:
-                df[shorten_term(field['term'])] = field_default_value
+                if isinstance(df_or_textreader, TextFileReader):
+                    # I don't see how to assign default values to a TextFileReader, so
+                    # this is currently unsupported
+                    raise ValueError(
+                        "Pandas read_csv() was called with a chunksize or iterator=True, "
+                        "and therefore returns a TextFileReader instead of a DataFrame "
+                        "which is not supported in combination with default values of "
+                        "the archive."
+                    )
+
+                df_or_textreader[shorten_term(field['term'])] = field_default_value
 
-        return df
+        return df_or_textreader
 
     def orphaned_extension_rows(self) -> Dict[str, Dict[str, List[int]]]:
         """Return a dict of the orphaned extension rows.
@@ -434,6 +448,7 @@ class DwCAReader(object):
         except zipfile.BadZipfile:
             # Doesn't look like a valid zip, let's see if it's a tar archive (possibly compressed)
             try:
+                # TODO: Once we only support Python 3.12+, we should pass the filter="data" argument to extractall()
                 tarfile.open(self.archive_path, 'r:*').extractall(tmp_dir)
             except tarfile.ReadError:
                 raise InvalidArchive("The archive cannot be read. Is it a .zip or .tgz file?")

{python-dwca-reader-0.16.0 → python-dwca-reader-0.16.1}/dwca/test/test_dwcareader.py

@@ -48,6 +48,26 @@ class TestPandasIntegration(unittest.TestCase):
             assert df["scientificName"].values.tolist() == \
                 ["tetraodon fluviatilis", "betta splendens"]
 
+    def test_pd_read_chunked_default_value(self):
+        """Pandas chuncksize should not be used with default values.
+
+        See: https://github.com/BelgianBiodiversityPlatform/python-dwca-reader/issues/106
+        """
+        with DwCAReader(sample_data_path("dwca-test-default.zip")) as dwca:
+            with pytest.raises(ValueError):
+                for chunk in dwca.pd_read("occurrence.txt", chunksize=1):
+                    pass
+
+    def test_pd_read_chunked(self):
+        """If no default values are available in the archive, chunksize should work.
+
+        See: https://github.com/BelgianBiodiversityPlatform/python-dwca-reader/issues/106
+        """
+        with DwCAReader(sample_data_path("dwca-simple-test-archive.zip")) as dwca:
+            for chunk in dwca.pd_read("occurrence.txt", chunksize=2):
+                assert isinstance(chunk, pd.DataFrame)
+
+
     def test_pd_read_no_data_files(self):
         with DwCAReader(sample_data_path("dwca-simple-test-archive.zip")) as dwca:
             with pytest.raises(NotADataFile):

python-dwca-reader-0.16.1/dwca/version.py

@@ -0,0 +1 @@
+__version__ = '0.16.1'

{python-dwca-reader-0.16.0 → python-dwca-reader-0.16.1}/python_dwca_reader.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: python-dwca-reader
-Version: 0.16.0
+Version: 0.16.1
 Summary: A simple Python package to read Darwin Core Archive (DwC-A) files.
 Home-page: https://github.com/BelgianBiodiversityPlatform/python-dwca-reader
 Author: Nicolas Noé - Belgian Biodiversity Platform

python-dwca-reader-0.16.0/dwca/version.py

@@ -1 +0,0 @@
-__version__ = '0.16.0'