[mdb5972]

Possible suggestion:

1. Create a Custom Column text field {disk_filename}
2. Add the books via the CLI calibredb add
3. Use the calibredb add -f #{disk_filename}:{filename}
-- one at a time so will be slower
4. Use RegEx in the multiple edit to parse the custom {disk_namefield} to the Author, Title fields -- saving them for quick use next time.

This would let you leave it on 'Read metadata from file' but allow you to quickly parse the custom field to adjust the title and author.

I use custom tags like '_ADDED' (from Preferences:'Add Books' and Tweaks:'Control how tags are applied...'=True) on add books to quickly search for a subset of records that I need to process after adding new books/records. Or for command line changes to existing records set an custom yes/no updated (#ie #upd) flag to find the changed subset.


Windows: update example

FOR /F "SKIP=1 TOKENS=1,2 DELIMS=," %a in (UNPROCESSED.CSV) DO @ECHO. && @ECHO BOOK ID: %~a && ECHO ISBN : %~b && ECHO. && @IF EXIST E:\xRESULTS\%~b.opf (calibredb set_metadata --dont-notify-gui --with-library="E:\CALIBRE\WORK" -f #upd:yes %~a e:\xRESULTS\%~b.opf)


%~b cuz my export adds double-quotes around the fields; %~ strips them