Quote:
Originally Posted by capink
OK. I think I figured out what the problem at your end is. So I modified the module (copy below) to explicitly set is_experimental to False.
|
Thanks! That worked! worked fine on a test of 40 books. running it on 2,000 now :-)
EDIT:
2,000 worked as well. slow and steady wins the race!! no significant CPU or memory build up either. Thank you once again. Action chains are fantastic as a way to change things programatically. And this mechanism allows the chains to run on every book in large libraries..
This would be an excellent addition to the list of examples you already have.. :-)
Note: The problem I was trying to solve with my chains was to reduce the set of tags in my books from ~50,000 to under 200. Most of these tags come from ao3 authors going mad and myself not caring previously.. I finally bit the bullet and wrote some chains to extract only things I care about into tags.. The goal is to make it much easier to find things I care about on my various readers (Android and Kobo) as the number of tags, currently, overwhelms the readers on both those devices and the reader software I use (moon+, lithium, native kobo reader) on those devices don't really look at other attributes beyond title, author, series, and tags.
my other solution, before the "process files in chunks" solution provided by capink, was to edit the sqlite db directly using a sqlite admin tool. However, this has its own issues as the metadata is scattered across multiple tables (needing the right set of join commands) and my SQL foo is weak..
Thanks again!
Bigwoof