How to build custom column based on other columne (if - then statement)

[asogn]

Hi there people

Background.
I have a Kindle 2/3 and I am planning to let the Kindle Collection plugin handle my collections.

The way I use kindle is that as soon as I am finished reading books in one collection I rename it on the kindle to x_NAME.... or X_NAME where NAME is the original collection name.
This works fine and I can set the plugin to not do anything with collections starting with x_

However, the plugin will try to recreate the original collection when I run it again.

So to handle this I am planning to also cater for the x_ named collection.
To do that, I use the plugin to read all existing collections existing on kindle into a custom column called kindlecollections.
Then I have another custom column called collectionname with following template
{#universe:'ifempty($, field('author_sort'))'}{series:| - |}
The template works the way I want. But does not check anything in #kindlecollections

What I would like is for the #collectionname to check if #kindlecollections start with x_ or X_
If it does, copy the content of #kindlecollections to #collectionname
If it doesn't, use {#universe:'ifempty($, field('author_sort'))'}{series:| - |} to populate #collectionname

I have been playing with the template, but can't seem to get my head around it.
Anyone able to help me?
Or is it another way to do what I want?

Cheers, Atle

[asogn]

OK, was able to figure it out. Had some problems due to apostrophe, but following works for me:
{#kindlecollections:'contains($,'^x_*',field('#kin dlecollections'),"{#universe:'ifempty($, field('author_sort'))'}{series:| - |}")'}

Cheers, Atle

[chaley]

Originally Posted by asogn

OK, was able to figure it out. Had some problems due to apostrophe, but following works for me:

{#kindlecollections:'contains($,'^x_*',field('#kindlecollections'),"{#universe:'ifempty($, field('author_sort'))'}{series:| - |}")'}

Cheers, Atle

That template will at some point not work, depending on the values in #universe or series. The problem: bracketed expressions are macros, evaluated before the template is parsed. Thus if #universe or series can contain a character that can affect parsing, most notably " or ' but other characters as well, then the template will fail. Also, nesting Template Program Mode expressions is problematic.

If you want to nest templates or TPM expressions then you should use General Program Mode where no macro processing takes place. They are also easier to read when dealing with this level of complexity. Something like:

program:
u = ifempty(field('#universe'), field('author_sort')); 
s = finish_formatting(field('series'), '', ' - ', '');
kc = field('#kindlecollections');
contains(kc, '^x_*', kc, strcat(u, s))

I can't test it because I don't have your columns.

FWIW: I suspect that the regexp you use in the 'contains' isn't what you really want. That regexp looks for a leading x followed by zero or more underscores. I thing you want '^x_', which is a leading x followed by an underscore.

[asogn]

Originally Posted by chaley

That template will at some point not work, depending on the values in #universe or series. The problem: bracketed expressions are macros, evaluated before the template is parsed. Thus if #universe or series can contain a character that can affect parsing, most notably " or ' but other characters as well, then the template will fail. Also, nesting Template Program Mode expressions is problematic.

If you want to nest templates or TPM expressions then you should use General Program Mode where no macro processing takes place. They are also easier to read when dealing with this level of complexity. Something like:

program:
u = ifempty(field('#universe'), field('author_sort')); 
s = finish_formatting(field('series'), '', ' - ', '');
kc = field('#kindlecollections');
contains(kc, '^x_*', kc, strcat(u, s))

I can't test it because I don't have your columns.

FWIW: I suspect that the regexp you use in the 'contains' isn't what you really want. That regexp looks for a leading x followed by zero or more underscores. I thing you want '^x_', which is a leading x followed by an underscore.

This works like a charm. Changed the regex to be correct (i.e. '^x_').
And it looks much simpler to use program: instead.
So when I have time I will expand this to also strip out quotes. Currently that is not a problem since I have changed all my series to not include quotes....
Turns out that Kindle is not so happy with quotes (or the kindle collections plugin is not so happy)

Next step is to verify that everything works properly and not screw up my kindle on the first run.

Thanks a lot

Cheers, Atle