[Device Plugin] Kobo Touch Extended Driver

[Katja_hbg]

Originally Posted by ownedbycats

I tried Ignore collections, but that just prevented it from being deleted. [Square brackets] did work though I might also do that for my Kobo Store collection because I don't really need that on my device.

I use tags for collection but like to skip some of them.
Reading that above I entered the values with [] and comma and also restart Calibre. After that I delete those collection manually from my Kobo. But after connection, all tags are created again. Although I did not expect to see [name1], [name2] anymore.

Is my expection or my understanding from your post wrong?

[JSWolf]

Originally Posted by Katja_hbg

I use tags for collection but like to skip some of them.
Reading that above I entered the values with [] and comma and also restart Calibre. After that I delete those collection manually from my Kobo. But after connection, all tags are created again. Although I did not expect to see [name1], [name2] anymore.

Is my expection or my understanding from your post wrong?

The problem you have is that you cannot do what you are trying to do, You remove the collection(s) and when you connect your Kobo to your library, it creates the missing collection yet again.

There are two ways to fix this. One is to delete the tags you don't want as collections or to create a custom column to use as collections and use tags for all the tags. Then you can delete the tags from the custom column when you are done with them.

[ownedbycats]

Not 100% if I'm understanding you right but I'll give it a try.

If you enclose a tag in [Square Brackets] it won't appear on the Kobo, unless they've changed it since my post (I doubt this).

If a collection is deleted with the driver, Kobo sync may bring it back. (I often have a similar issue with Kobo store books reappearing in my 'to be read' collection.) Delete it on the device itself and sync again and it should stop.

[Terisa de morgan]

Originally Posted by Katja_hbg

I use tags for collection but like to skip some of them.
Reading that above I entered the values with [] and comma and also restart Calibre. After that I delete those collection manually from my Kobo. But after connection, all tags are created again. Although I did not expect to see [name1], [name2] anymore.

Is my expection or my understanding from your post wrong?

I have to check the configuration and I'll be back to you this evening, but I think it's something like this: #1,#2,#3. But I don't know if it will be right for you as what I avoid is calibre deleting those collections (they are not in calibre in any way, and I handle them wholly at my Kobo).

[chaley]

Originally Posted by Katja_hbg

I use tags for collection but like to skip some of them.
Reading that above I entered the values with [] and comma and also restart Calibre. After that I delete those collection manually from my Kobo. But after connection, all tags are created again. Although I did not expect to see [name1], [name2] anymore.

Is my expection or my understanding from your post wrong?

You can do this with the template feature in the Kobo driver. See this thread. The idea is that the template filters out the tags you don't want as collections, returning the rest (if any).

[Katja_hbg]

Thanks all for your reply and suppport.

Originally Posted by ownedbycats

If you enclose a tag in [Square Brackets] it won't appear on the Kobo, unless they've changed it since my post (I doubt this).

That was my expectation but it appears again.

Originally Posted by chaley

You can do this with the template feature in the Kobo driver. See this thread. The idea is that the template filters out the tags you don't want as collections, returning the rest (if any).

I already saw that. Now I will have a deeper look into it but maybe I am overwhelmed.

[chaley]

Originally Posted by Katja_hbg

I already saw that. Now I will have a deeper look into it but maybe I am overwhelmed.

Here is a template that removes all tags bracketed by [] and all tags listed in the variable tags_to_ignore.

program:
# Remove any tag bracketed by []
	tags = list_re($tags, ',', '^[^\[].*[^\]]$', '');
# Remove tags listed in tags_to_ignore
	tags_to_ignore = 'AA.BBF.CCC, onDevice';
	list_join(':@:', list_difference(tags, tags_to_ignore, ','), ',')

Example: if a book has the tags

[PBook], Science Fiction, allbooks.foo, Fiction, Fantasy, General, Epic, AA.BBF.CCC, onDevice

and tags_to_ignore is

	tags_to_ignore = 'AA.BBF.CCC, onDevice';

then the template result is

allbooks.foo:@:Epic:@:Fantasy:@:Fiction:@:General:@:Science Fiction

[Katja_hbg]

Originally Posted by chaley

Here is a template that removes all tags bracketed by [] and all tags listed in the variable tags_to_ignore.

Thanks for the details Chaley but I did not make it well.
I do not have tags in brackets (they were added by me in the field "ignore collection" only due to the description in the former forum entry by ownedbycats as a solution to delete instead of preventing to delete).

Field "collection template" is now:

# Remove tags listed in tags_to_ignore
	tags_to_ignore = 'danach, kurz';
	list_join(':@:', list_difference(tags, tags_to_ignore, ','), ',')

Although I deleted them from Kobo before connecting to Calibre, they are added again.
Second try: in addition to the template I update field "collection ignore" showing now: danach, kurz
But still this two tags are created.

In both cases I got an additional tag "Kobo Interpreter: Unknown"

[isarl]

Originally Posted by Katja_hbg

Field "collection template" is now:

# Remove tags listed in tags_to_ignore
	tags_to_ignore = 'danach, kurz';
	list_join(':@:', list_difference(tags, tags_to_ignore, ','), ',')

[…]

In both cases I got an additional tag "Kobo Interpreter: Unknown"

The keyword “program:”, including the colon, at the beginning of chaley's template is important. Your template appears to omit it.

[Katja_hbg]

Originally Posted by isarl

The keyword “program:”, including the colon, at the beginning of chaley's template is important. Your template appears to omit it.

Sorry this is a lazy copy mistake by me only here in the forum.
Of course the template content starts with "program:"

[isarl]

Furthermore, are you using both “collections columns” as well as “collections template”?

If you have the “collections columns” enabled and one or more fields specified, then those will be used as-is, regardless of what you have in the template.

Finally, your template does not initialise the tags variable. Since you are not modifying the tags variable like chaley was to remove tags surrounded in square brackets, you can simply replace “tags” with a column lookup on your source collections. So if your custom column lookup name is #koboshelves then I think your template would look, in full, like:

program:
# Remove tags listed in tags_to_ignore
	tags_to_ignore = 'danach, kurz';
	list_join(':@:', list_difference($#koboshelves, tags_to_ignore, ','), ',')

This seems to work when testing in my own library using the Template Tester.

[isarl]

Originally Posted by Katja_hbg

Sorry this is a lazy copy mistake by me only here in the forum.
Of course the template content starts with "program:"

I thought this might be the case, but it is always worth asking a question even if the answer is seemingly “obvious”. Thank you for clarifying!

[Katja_hbg]

Originally Posted by isarl

Furthermore, are you using both “collections columns” as well as “collections template”?

If you have the “collections columns” enabled and one or more fields specified, then those will be used as-is, regardless of what you have in the template.

And this is the killer.
I use "collection columns" with three fields. One of the is tags.
After being happy with that for years, today I had the idea to reduce the values coming out of "tags".

Thanks again for your help but after having some fun to give it a try, I will continue with former setting.

[isarl]

Originally Posted by Katja_hbg

And this is the killer.
I use "collection columns" with three fields. One of the is tags.
After being happy with that for years, today I had the idea to reduce the values coming out of "tags".

Thanks again for your help but after having some fun to give it a try, I will continue with former setting.

You're so close! There's no need to give up when you're so close.

Any tags you do NOT want processed by the template, leave in your Collections Columns field.

Any tags which you DO want processed by the template, make sure they are NOT in the Collections Columns field, and that your template logic handles these other fields correctly.

For example, if I have Genre (#genre), Collections (#collections), and Tags (tags – this is a builtin column, so no #prefix for it), and I only wanted to perform this custom processing on the tags, then I would put #genre, #collections in the “Collections Columns”. And for the template I would use something like we described previously:

program:
# Remove tags listed in tags_to_ignore
	tags_to_ignore = 'danach, kurz';
	list_join(':@:', list_difference($tags, tags_to_ignore, ','), ',')

Just remember: if you want to process a column before using it as collections, then you need to put that column into a template, not in your “Collections Columns” setting. The “Collections Columns” setting is only for data that will be used as-is.

[chaley]

Originally Posted by isarl

Just remember: if you want to process a column before using it as collections, then you need to put that column into a template, not in your “Collections Columns” setting. The “Collections Columns” setting is only for data that will be used as-is.

My personal choice would be to do it all in the template and leave "Collections Columns" empty/unchecked. That way I have only one place to look and I can more easily fine tune the results.

Example using the columns you mentioned:

program:
# Remove tags listed in tags_to_ignore
	tags_to_ignore = 'danach, kurz';
	list_join(':@:', 
		list_difference($tags, tags_to_ignore, ','), ',',
		$#genre, ',',
		$#collections, ','
	)