Changed the way parsers are discovered. This also prepares for upcoming changes regarding content types and file types: parsers should declare what they support, and actual file extensions should not be hardcoded everywhere.

src/paperless_text/apps.py

@@ -1,5 +1,7 @@
 from django.apps import AppConfig
 
+from paperless_text.signals import text_consumer_declaration
+
 
 class PaperlessTextConfig(AppConfig):
 
@@ -9,8 +11,6 @@ class PaperlessTextConfig(AppConfig):
 
         from documents.signals import document_consumer_declaration
 
-        from .signals import ConsumerDeclaration
-
-        document_consumer_declaration.connect(ConsumerDeclaration.handle)
+        document_consumer_declaration.connect(text_consumer_declaration)
 
         AppConfig.ready(self)

src/paperless_text/signals.py

@@ -3,21 +3,16 @@ import re
 from .parsers import TextDocumentParser
 
 
-class ConsumerDeclaration:
+def text_consumer_declaration(sender, **kwargs):
+    return {
+        "parser": TextDocumentParser,
+        "weight": 10,
+        "test": text_consumer_test
+    }
 
-    MATCHING_FILES = re.compile(r"^.*\.(te?xt|md|csv)$")
 
-    @classmethod
-    def handle(cls, sender, **kwargs):
-        return cls.test
+MATCHING_FILES = re.compile(r"^.*\.(te?xt|md|csv)$")
 
-    @classmethod
-    def test(cls, doc):
 
-        if cls.MATCHING_FILES.match(doc.lower()):
-            return {
-                "parser": TextDocumentParser,
-                "weight": 10
-            }
-
-        return None
+def text_consumer_test(doc):
+    return MATCHING_FILES.match(doc.lower())