Don't parse dates with more than 4 digits for the year

The regex was broken before, using `(?!…)` instead of `(?<=…)`.

Author: heinrich5991

src/documents/parsers.py

@@ -21,11 +21,11 @@
 # - MONTH ZZZZ, with ZZZZ being 4 digits
 # - MONTH XX, ZZZZ with XX being 1 or 2 and ZZZZ being 4 digits
 DATE_REGEX = re.compile(
-    r'(\b|(?!=([_-])))([0-9]{1,2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{4}|[0-9]{2})(\b|(?=([_-])))|' +  # NOQA: E501
-    r'(\b|(?!=([_-])))([0-9]{4}|[0-9]{2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{1,2})(\b|(?=([_-])))|' +  # NOQA: E501
-    r'(\b|(?!=([_-])))([0-9]{1,2}[\. ]+[^ ]{3,9} ([0-9]{4}|[0-9]{2}))(\b|(?=([_-])))|' +  # NOQA: E501
-    r'(\b|(?!=([_-])))([^\W\d_]{3,9} [0-9]{1,2}, ([0-9]{4}))(\b|(?=([_-])))|' +
-    r'(\b|(?!=([_-])))([^\W\d_]{3,9} [0-9]{4})(\b|(?=([_-])))'
+    r'(\b|(?<=([_-])))([0-9]{1,2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{4}|[0-9]{2})(\b|(?=([_-])))|' +  # NOQA: E501
+    r'(\b|(?<=([_-])))([0-9]{4}|[0-9]{2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{1,2})(\b|(?=([_-])))|' +  # NOQA: E501
+    r'(\b|(?<=([_-])))([0-9]{1,2}[\. ]+[^ ]{3,9} ([0-9]{4}|[0-9]{2}))(\b|(?=([_-])))|' +  # NOQA: E501
+    r'(\b|(?<=([_-])))([^\W\d_]{3,9} [0-9]{1,2}, ([0-9]{4}))(\b|(?=([_-])))|' +
+    r'(\b|(?<=([_-])))([^\W\d_]{3,9} [0-9]{4})(\b|(?=([_-])))'
 )
 
 

src/paperless_tesseract/tests/test_date.py

@@ -182,3 +182,26 @@ def test_crazy_date_past(self, *args):
         document = RasterisedDocumentParser("/dev/null")
         document.get_text()
         self.assertIsNone(document.get_date())
+
+    EXTRA = {
+        "123/04/2020/3423": None,
+        "-23/04/2020-foo": "2020 04 23",
+        "-23-04-2020-blurb": "2020 04 23",
+        # gets parsed as month: 23, day: 04, which is invalid
+        #"-2020-04-23-bar": "2020 04 23",
+        "12020-04-23-": None,
+        "-2020-04-234": None,
+    }
+
+    @mock.patch(MOCK_SCRATCH, SCRATCH)
+    def test_date_format_bulk(self):
+        timezone = tz.gettz(settings.TIME_ZONE)
+        for input, expected in self.EXTRA.items():
+            if expected is not None:
+                raw = [int(x) for x in expected.split()]
+                expected = datetime.datetime(*raw, tzinfo=timezone)
+
+            input_file = os.path.join(self.SAMPLE_FILES, "")
+            document = RasterisedDocumentParser(input_file)
+            document._text = input
+            self.assertEqual(document.get_date(), expected, msg="Test case {!r}".format(input))