Bulk auto-delete redudant hierarchical tags

[droopy]

A computer has directories and those directories can have subdirectories. I think we all do better organizing computer files into directories and subdirectories. I think that I would find it beneficial, personally speaking, to have tags in hierarchies where it’s relevant. Thus, I have recently enabled hierarchical tags.

With that as context, I’ve recently enabled a number of Metadata-Source plugins, including the KoboBooks plugin. With both the use of hierarchical tags and KoboBooks plugin active, the tags for a book may now look like this:

Nonfiction.Religion & Spirituality.Philosophy, Nonfiction, Nonfiction.Religion & Spirituality.Philosophy.Aesthetics, Nonfiction.Art & Architecture.Art History, Nonfiction.Social & Cultural Studies.Social Science, Nonfiction.Religion & Spirituality, Nonfiction.Art & Architecture, Nonfiction.Social & Cultural Studies

If we had one tag per line, it would look something like this:
Nonfiction
Nonfiction.Religion & Spirituality
Nonfiction.Religion & Spirituality.Philosophy
Nonfiction.Religion & Spirituality.Philosophy.Aesthetics
Nonfiction.Art & Architecture
Nonfiction.Art & Architecture.Art History
Nonfiction.Social & Cultural Studies
Nonfiction.Social & Cultural Studies.Social Science


What I’d like to do is keep just those tags that have the most data (the longest ones) while excising the shorter, redundant ones. Here’s what I mean. The red tags are deleted. The green, bold ones are preserved.

Nonfiction
Nonfiction.Religion & Spirituality
Nonfiction.Religion & Spirituality.Philosophy
Nonfiction.Religion & Spirituality.Philosophy.Aesthetics
Nonfiction.Art & Architecture
Nonfiction.Art & Architecture.Art History
Nonfiction.Social & Cultural Studies
Nonfiction.Social & Cultural Studies.Social Science

Is there a way for Calibre to auto-search through my entire library and do such a bulk purging?

[droopy]

bump.

[chaley]

Originally Posted by droopy

bump.

No, there isn't an automatic way to do it. The only way I know of with base calibre is book-by-book using the tag editor opened in edit metadata (single) or by Shift F2 on Tags in the book list.

If you know python you could do it using a user-defined template function, sorting the tags by name then deleting any that are a prefix of the next one in the list.

[droopy]

I'm in the queue for a python book from the library.