add support for archive files.

src/documents/consumer.py

@@ -134,6 +134,7 @@ class Consumer(LoggingMixin):
             self.log("debug", "Parsing {}...".format(self.filename))
             text = document_parser.get_text()
             date = document_parser.get_date()
+            archive_path = document_parser.get_archive_path()
         except ParseError as e:
             document_parser.cleanup()
             raise ConsumerError(e)
@@ -178,8 +179,16 @@ class Consumer(LoggingMixin):
                 # place. If this fails, we'll also rollback the transaction.
 
                 create_source_path_directory(document.source_path)
-                self._write(document, self.path, document.source_path)
-                self._write(document, thumbnail, document.thumbnail_path)
+
+                self._write(document.storage_type,
+                            self.path, document.source_path)
+
+                self._write(document.storage_type,
+                            thumbnail, document.thumbnail_path)
+
+                if archive_path and os.path.isfile(archive_path):
+                    self._write(Document.STORAGE_TYPE_UNENCRYPTED,
+                                archive_path, document.archive_path)
 
                 # Delete the file only if it was successfully consumed
                 self.log("debug", "Deleting file {}".format(self.path))
@@ -258,10 +267,10 @@ class Consumer(LoggingMixin):
             for tag_id in self.override_tag_ids:
                 document.tags.add(Tag.objects.get(pk=tag_id))
 
-    def _write(self, document, source, target):
+    def _write(self, storage_type, source, target):
         with open(source, "rb") as read_file:
             with open(target, "wb") as write_file:
-                if document.storage_type == Document.STORAGE_TYPE_UNENCRYPTED:
+                if storage_type == Document.STORAGE_TYPE_UNENCRYPTED:
                     write_file.write(read_file.read())
                     return
                 self.log("debug", "Encrypting")

src/documents/models.py

@@ -224,6 +224,19 @@ class Document(models.Model):
     def source_file(self):
         return open(self.source_path, "rb")
 
+    @property
+    def archive_path(self):
+        fname = "{:07}{}".format(self.pk, ".pdf")
+
+        return os.path.join(
+            settings.ARCHIVE_DIR,
+            fname
+        )
+
+    @property
+    def archive_file(self):
+        return open(self.archive_path, "rb")
+
     @property
     def file_name(self):
         return slugify(str(self)) + self.file_type

src/documents/parsers.py

@@ -141,6 +141,9 @@ class DocumentParser(LoggingMixin):
         self.tempdir = tempfile.mkdtemp(
             prefix="paperless-", dir=settings.SCRATCH_DIR)
 
+    def get_archive_path(self):
+        return None
+
     def get_thumbnail(self):
         """
         Returns the path to a file we can use as a thumbnail for this document.

src/documents/signals/handlers.py

@@ -168,11 +168,17 @@ def run_post_consume_script(sender, document, **kwargs):
 
 @receiver(models.signals.post_delete, sender=Document)
 def cleanup_document_deletion(sender, instance, using, **kwargs):
-    for f in (instance.source_path, instance.thumbnail_path):
-        try:
-            os.unlink(f)
-        except FileNotFoundError:
-            pass  # The file's already gone, so we're cool with it.
+    for f in (instance.source_path,
+              instance.archive_path,
+              instance.thumbnail_path):
+        if os.path.isfile(f):
+            try:
+                os.unlink(f)
+            except OSError as e:
+                logging.getLogger(__name__).warning(
+                    f"While deleting document {instance.file_name}, the file "
+                    f"{f} could not be deleted: {e}"
+                )
 
     delete_empty_directories(os.path.dirname(instance.source_path))