[chaley]

Originally Posted by TicklishOwl

I'm trying to set up a composite custom column (built from Calibre's tags column), but I'm having trouble with the template code.

...
Example: A book tagged 'classics, france, historical, middle-grade' would add 'Historical Fiction, Middle-Grade' to #koboshelf.

It's easy to get a single tag to work:

{tags:contains(historical, Historical Fiction,)}

Unfortunately, I can't get multiple tags to work. Clearly I do not have a gift for this language. I'd be most grateful for some guidance.

Use a general program mode template like the following

program:
	t = field('tags');
	l = '';
	l = list_union(l, in_list(t, ',', 'space', 'Space Opera', ''), ',');
	l = list_union(l, in_list(t, ',', 'science', 'Science Fiction', ''), ',');
	l = list_union(l, in_list(t, ',', 'alternate', 'Alternate History', ''), ',');
	l = list_union(l, in_list(t, ',', 'a', 'b', ''), ',');

This template constructs a list of tags returned by the in_list functions.

Note 1: in_list takes a regular expression for the match pattern. If there are multiple tags that result in the same shelf, you can write it like this

list_union(l, in_list(t, ',', 'tag1|tag2', 'Space Opera', ''), ',');

Note 2: if you want exact matching on the tags, use something like

list_union(l, in_list(t, ',', '^space$', 'Space Opera', ''), ',');

or in the case of a multiple match

list_union(l, in_list(t, ',', '^(tag1|tag2)$', 'Space Opera', ''), ',');

Note 3: this template is at the limit of what the template system is intended to do. If you have a large library, it will probably be slow. What "large" means depends on the computer running calibre. If you have python skills, or if you know someone with those skills, a custom calibre template function will have much better performance.