From 86788f14457e8263c917d4f9fceebe424379ce9d Mon Sep 17 00:00:00 2001
From: shamoon <4887959+shamoon@users.noreply.github.com>
Date: Mon, 13 Jan 2025 13:02:10 -0800
Subject: [PATCH] Fix: use unmodified original for checksum if exists (#8693)

---
 src/documents/consumer.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/src/documents/consumer.py b/src/documents/consumer.py
index f353afbe7..75fa1da1b 100644
--- a/src/documents/consumer.py
+++ b/src/documents/consumer.py
@@ -718,11 +718,17 @@ class ConsumerPlugin(
                     f"Error occurred parsing title override '{self.metadata.title}', falling back to original. Exception: {e}",
                 )
 
+        file_for_checksum = (
+            self.unmodified_original
+            if self.unmodified_original is not None
+            else self.working_copy
+        )
+
         document = Document.objects.create(
             title=title[:127],
             content=text,
             mime_type=mime_type,
-            checksum=hashlib.md5(self.working_copy.read_bytes()).hexdigest(),
+            checksum=hashlib.md5(file_for_checksum.read_bytes()).hexdigest(),
             created=create_date,
             modified=create_date,
             storage_type=storage_type,