mime type handling

2020-11-20 13:31:03 +01:00
parent cee010cc21
commit f976a0b4ba
19 changed files with 163 additions and 146 deletions
--- a/src/paperless_tesseract/signals.py
+++ b/src/paperless_tesseract/signals.py
@@ -1,5 +1,3 @@
-import re
-
 from .parsers import RasterisedDocumentParser


@@ -7,12 +5,9 @@ def tesseract_consumer_declaration(sender, **kwargs):
    return {
        "parser": RasterisedDocumentParser,
        "weight": 0,
-        "test": tesseract_consumer_test
+        "mime_types": [
+            "application/pdf",
+            "image/jpeg",
+            "image/png"
+        ]
    }
-
-
-MATCHING_FILES = re.compile(r"^.*\.(pdf|jpe?g|gif|png|tiff?|pnm|bmp)$")
-
-
-def tesseract_consumer_test(doc):
-    return MATCHING_FILES.match(doc.lower())