Adds better error handling/checking around getting content of a document via Tika

Signed-off-by: Trenton Holmes <797416+stumpylog@users.noreply.github.com>

src/paperless_tika/parsers.py

@@ -4,6 +4,7 @@ from pathlib import Path
 import httpx
 from django.conf import settings
 from tika_client import TikaClient
+from tika_client.data_models import TikaKey
 
 from documents.parsers import DocumentParser
 from documents.parsers import ParseError
@@ -58,8 +59,15 @@ class TikaDocumentParser(DocumentParser):
                 f"{settings.TIKA_ENDPOINT}: {err}",
             ) from err
 
-        self.text = parsed.content.strip()
-        self.date = parsed.metadata.created
+        self.text = None
+        if hasattr(parsed, "content") and parsed.content is not None:
+            self.text = parsed.content.strip()
+        elif TikaKey.Content in parsed.data:
+            # May not be a completely handled type, but
+            # the Tika response may still include content
+            self.text = parsed.data[TikaKey.Content].strip()
+
+        self.date = parsed.created
         self.archive_path = self.convert_to_pdf(document_path, file_name)
 
     def convert_to_pdf(self, document_path, file_name):

src/paperless_tika/tests/test_live_tika.py

@@ -118,3 +118,28 @@ class TestTikaParserAgainstServer(TestCase):
             self.assertTrue(b"PDF-" in f.read()[:10])
 
         # self.assertEqual(self.parser.date, datetime.datetime(2022, 9, 14))
+
+    def test_basic_parse_doc(self):
+        """
+        GIVEN:
+            - An input DOC format document
+        WHEN:
+            - The document is parsed
+        THEN:
+            - Document content is correct
+            - Document date is correct
+        """
+        test_file = self.SAMPLE_DIR / "sample.doc"
+
+        self.try_parse_with_wait(
+            test_file,
+            "application/msword",
+        )
+
+        self.assertIn(
+            "his is a test document, saved in the older .doc format",
+            self.parser.text,
+        )
+        self.assertIsNotNone(self.parser.archive_path)
+        with open(self.parser.archive_path, "rb") as f:
+            self.assertTrue(b"PDF-" in f.read()[:10])