To recreate: 1 - Export some records form Koha 2 - Strip all the subject fields 3 - Stage for import - Set matching rule on 999 or isbn, whatever works - Set action when matching record found to ignore 4 - View the batch and see the matches 5 - Click the diff view 6 - It shows the imported record with subject stripped 7 - It shows the subject fields in the existing record crossed out 8 - This makes it look like the import will remove fields 9 - Import the batch 10 - Confirm the records are still complete, subject not removed
+1 for this! The diff view follows the standard conventions for git, but it's confusing and misleading with regards to what's actually happening with the record. It would be ideal if the green and red highlighting could show what data is actually kept based on the import settings. Or failing that, we could highlight the differences in yellow on both sides, to at least avoid giving a wrong impression when the incoming file is being ignored.