moved metadata extraction to the parsers

src/documents/parsers.py

@@ -210,6 +210,7 @@ class DocumentParser(LoggingMixin):
     def __init__(self, logging_group):
         super().__init__()
         self.logging_group = logging_group
+        os.makedirs(settings.SCRATCH_DIR, exist_ok=True)
         self.tempdir = tempfile.mkdtemp(
             prefix="paperless-", dir=settings.SCRATCH_DIR)
 
@@ -217,6 +218,9 @@ class DocumentParser(LoggingMixin):
         self.text = None
         self.date = None
 
+    def extract_metadata(self, document_path, mime_type):
+        return []
+
     def parse(self, document_path, mime_type):
         raise NotImplementedError()