Merge pull request #426 from NaturalHistoryMuseum/feature/425-warnings

[#425] Warnings

CHANGELOG.md

@@ -2,6 +2,7 @@ This is an overview of major changes. Refer to the git repository for a full log
 
 Version 0.1.35
 -------------
+- #425 Warnings
 - #420 zbar decoder on Mac OS X
 - #418 Latest gouda and pylibdmtx
 - #416 User testing / Python 3 - error message when opening document on 32-bit Windows

build.sh

@@ -32,7 +32,7 @@ for module in cv2 numpy libdmtx scipy sklearn zbar; do
 done
 
 echo Tests
-nosetests --with-coverage --cover-html --cover-inclusive --cover-erase --cover-tests --cover-package=inselect inselect
+PYTHONWARNINGS=module nosetests --with-coverage --cover-html --cover-inclusive --cover-erase --cover-tests --cover-package=inselect inselect
 
 echo Wheel build
 ./setup.py bdist_wheel --universal

inselect/lib/document.py

@@ -220,43 +220,46 @@ def load(cls, path):
         path = Path(path)
 
         # Sniff the first few bytes - file must look like a json document
-        if not re.match(b'^{[ (\n)|(\r\n)]*"', path.open('rb').read(20)):
+        with path.open('rb') as infile:
+            if not re.match(b'^{[ (\n)|(\r\n)]*"', infile.read(20)):
+                raise InselectError('Not an inselect document')
+
+        with path.open(encoding='utf8') as infile:
+            doc = json.load(infile)
+
+        v = doc.get('inselect version')
+
+        if not v:
             raise InselectError('Not an inselect document')
+        elif v not in cls.FILE_VERSIONS:
+            raise InselectError('Unsupported version [{0}]'.format(v))
         else:
-            doc = json.load(path.open(encoding='utf8'))
-            v = doc.get('inselect version')
-
-            if not v:
-                raise InselectError('Not an inselect document')
-            elif v not in cls.FILE_VERSIONS:
-                raise InselectError('Unsupported version [{0}]'.format(v))
-            else:
-                if 1 == v:
-                    # Version 1 contained just three illustrative fields -
-                    # convert these to Darwin Core fields
-                    for item in doc['items']:
-                        fields = item['fields']
-                        if fields.get('Taxonomic group'):
-                            fields['scientificName'] = fields.pop('Taxonomic group')
-                        if fields.get('Location'):
-                            fields['otherCatalogNumbers'] = fields.pop('Location')
-                        if fields.get('Specimen number'):
-                            fields['catalogNumber'] = fields.pop('Specimen number')
-                        item['fields'] = fields
-
-                scanned = path.with_suffix(doc['scanned extension'])
-
-                properties = doc.get('properties', {})
-
-                # Parse datetimes
-                for dt in {'Saved on', 'Created on'}.intersection(properties.keys()):
-                    properties[dt] = cls._parse_datetime(properties[dt])
-
-                msg = 'Loaded [{0}] items from [{1}]'
-                debug_print(msg.format(len(doc['items']), path))
-
-                return cls(scanned_path=scanned, items=doc['items'],
-                           properties=properties)
+            if 1 == v:
+                # Version 1 contained just three illustrative fields -
+                # convert these to Darwin Core fields
+                for item in doc['items']:
+                    fields = item['fields']
+                    if fields.get('Taxonomic group'):
+                        fields['scientificName'] = fields.pop('Taxonomic group')
+                    if fields.get('Location'):
+                        fields['otherCatalogNumbers'] = fields.pop('Location')
+                    if fields.get('Specimen number'):
+                        fields['catalogNumber'] = fields.pop('Specimen number')
+                    item['fields'] = fields
+
+            scanned = path.with_suffix(doc['scanned extension'])
+
+            properties = doc.get('properties', {})
+
+            # Parse datetimes
+            for dt in {'Saved on', 'Created on'}.intersection(properties.keys()):
+                properties[dt] = cls._parse_datetime(properties[dt])
+
+            msg = 'Loaded [{0}] items from [{1}]'
+            debug_print(msg.format(len(doc['items']), path))
+
+            return cls(scanned_path=scanned, items=doc['items'],
+                       properties=properties)
 
     def save(self):
         "Saves to self.document_path"

inselect/lib/parse.py

@@ -289,8 +289,12 @@ def parse_in_choices(choices, value):
     else:
         return value
 
-# Populate dict {name: parse function}.
+# Populate dict {name: parse function}. See comment alongside the first
+# declaration of the PARSERS global, towards the top of this file.
 PARSERS = inspect.getmembers(sys.modules[__name__], inspect.isfunction)
 PARSERS = filter(lambda v: re.match(r'^parse_.+$', v[0]), PARSERS)
-PARSERS = filter(lambda v: ['value'] == inspect.getargspec(v[1]).args, PARSERS)
+PARSERS = filter(
+    lambda v: ['value'] == list(inspect.signature(v[1]).parameters.keys()),
+    PARSERS
+)
 PARSERS = dict(PARSERS)

inselect/tests/gui/test_export_csv.py

@@ -25,8 +25,8 @@ def _test_csv(self):
         csv = doc.document_path.with_suffix('.csv')
 
         # Check CSV contents
-        with csv.open('rb') as f:
-            res = unicodecsv.DictReader(f, encoding='utf-8')
+        with csv.open('rb') as infile:
+            res = unicodecsv.DictReader(infile, encoding='utf-8')
             for index, item, row in zip(count(), doc.items, res):
                 expected = item['fields']
                 expected.update({
@@ -73,7 +73,7 @@ def test_export_overwite_existing_csv(self, mock_information, mock_question):
                                        TESTDATA / 'shapes.png') as tempdir:
 
             # Create a CSV file to force the GUI to prompt for over-write
-            (tempdir / 'shapes.csv').open('w')
+            (tempdir / 'shapes.csv').touch()
 
             # Load document and export CSV file
             w.open_document(path=tempdir / 'shapes.inselect')
@@ -102,14 +102,15 @@ def test_export_do_not_overwite_existing_csv(self, mock_information, mock_questi
                                        TESTDATA / 'shapes.png') as tempdir:
 
             # Create a CSV file to force the GUI to prompt for over-write
-            (tempdir / 'shapes.csv').open('w')
+            (tempdir / 'shapes.csv').touch()
 
             # Load document and export CSV file
             w.open_document(tempdir / 'shapes.inselect')
             w.export_csv(user_template=DWC)
 
             # File should not have been altered
-            self.assertEqual('', (tempdir / 'shapes.csv').open().read())
+            with (tempdir / 'shapes.csv').open() as infile:
+                self.assertEqual('', infile.read())
 
             # User should not have been told about the export
             self.assertFalse(mock_information.called)

inselect/tests/gui/test_utils.py

@@ -31,7 +31,7 @@ def test_reveal_path_windows(self, mock_subprocess):
         "User reveals a path"
         with temp_directory_with_files() as tempdir:
             path = tempdir / 'xyz'
-            path.open('w')
+            path.touch()
             reveal_path(path)
             expected = "explorer.exe /select,{0}".format(path.resolve())
             mock_subprocess.assert_called_once_with(expected)
@@ -42,7 +42,7 @@ def test_reveal_path_os_x(self, mock_subprocess):
         "User reveals a path"
         with temp_directory_with_files() as tempdir:
             path = tempdir / 'xyz'
-            path.open('w')
+            path.touch()
             reveal_path(path)
             expected = [
                 '/usr/bin/osascript',

inselect/tests/lib/test_document.py

@@ -123,21 +123,22 @@ def test_load_images(self):
         source = TESTDATA / 'shapes.inselect'
         with temp_directory_with_files(TESTDATA / 'shapes.inselect') as tempdir:
             doc_temp = tempdir / 'shapes.inselect'
-            doc_temp.open('w').write(source.open().read())
+            with doc_temp.open('w') as outfile, source.open() as infile:
+                outfile.write(infile.read())
 
             # Document load with neither scanned image file nor thumbnail
             self.assertRaises(InselectError, InselectDocument.load, doc_temp)
 
             # Document load with thumbnail but no scanned image file
             thumbnail_temp = tempdir / 'shapes_thumbnail.jpg'
-            thumbnail_temp.open('w')       # File only needs to exist
+            thumbnail_temp.touch()       # File only needs to exist
             doc = InselectDocument.load(doc_temp)
             self.assertFalse(doc.scanned.available)
             self.assertTrue(doc.thumbnail.available)
 
             # Document load with both scanned and thumbnail files
             scanned_temp = tempdir / 'shapes.png'
-            scanned_temp.open('w')       # File only needs to exist
+            scanned_temp.touch()       # File only needs to exist
             actual = InselectDocument.load(doc_temp)
             self.assertEqual(InselectDocument.load(source).items, actual.items)
             self.assertTrue(actual.scanned.available)
@@ -278,13 +279,13 @@ def test_thumbnail_path_of_scanned(self):
     def test_path_is_thumbnail_file(self):
         with temp_directory_with_files() as tempdir:
             thumbnail = tempdir / 'xx_thumbnail.jpg'
-            thumbnail.open('w')       # File only needs to exist
+            thumbnail.touch()       # File only needs to exist
 
             # Thumbnail file exists but there is no corresponding .inselect doc
             self.assertFalse(InselectDocument.path_is_thumbnail_file(thumbnail))
 
             doc = tempdir / 'xx.inselect'
-            doc.open('w')       # File only needs to exist
+            doc.touch()       # File only needs to exist
 
             # Thumbnail file and corresponding .inselect file both exist
             self.assertTrue(InselectDocument.path_is_thumbnail_file(thumbnail))

inselect/tests/lib/test_utils.py

@@ -50,9 +50,9 @@ def test_rmtree_readonly(self):
         d = tempfile.mkdtemp()
         path = Path(d)
         try:
-            (path / 'a file').open('w')
+            (path / 'a file').touch()
             (path / 'a directory').mkdir()
-            (path / 'a directory' / 'another file').open('w')
+            (path / 'a directory' / 'another file').touch()
             make_readonly(path)
             self.assertTrue(path.is_dir())
         finally:

inselect/tests/scripts/test_ingest.py

@@ -30,9 +30,9 @@ def test_images_suffixes_re(self):
         self.assertRegex('x.TIFF', re)
         self.assertRegex('x.Tiff', re)
 
-        self.assertNotRegexpMatches('x.jpgx', re)
-        self.assertNotRegexpMatches('x.jpg ', re)
-        self.assertNotRegexpMatches('x.txt', re)
+        self.assertNotRegex('x.jpgx', re)
+        self.assertNotRegex('x.jpg ', re)
+        self.assertNotRegex('x.txt', re)
 
 
 class TestIngest(unittest.TestCase):