[GUI Plugin] Quality Check

[DoctorOhh]

Originally Posted by kiwidude

using the "Check books without jackets".

I have a question, how do you check for a Book jacket? During ePub to ePub conversions calibre checks for a properly implemented jacket and if found it removes the old book jacket and replaces it with the new book jacket. This update was in calibre 0.6.50.

Release: 0.6.50 [30 Apr, 2010]

Bug Fixes

* When converting EPUB to EPUB multiple times and creating book jacket from metadata, if an existing book jacket is found, replace it. This will only work with EPUBs created with the current release onwards

I ask because any book jacket created prior to 0.6.50 is not a proper book jacket and if I update any book created before 0.6.50 I will end up with a old book jacket and a new book jacket.

Does your QA check ever identify books converted prior to 0.6.50 as having a book jacket? It seems you should be able to use the code Kovid uses to identify proper book jackets.

This was really my motivation in wanting to know when a book was created. Maybe a QA check ePubs created before version 0.x.xx would work. Or maybe your QA check for a book jacket already takes this into account.

[kiwidude]

Originally Posted by dwanthny

I have a question, how do you check for a Book jacket?

I'm afraid Calibre 0.6.50 is the equivalent of being before I was born . So I have no idea about the old logic versus the new. What I can do is tell you what my check does...

• Looks for a file named 'jacket.xhtml'. If found, looks inside it for either of:
• <meta content="jacket" name="calibre-content"
• <meta name="calibre-content" content="jacket"

EDIT: At purely a guess I would say perhaps in 0.6.50 they added the <meta> tag, which is why the comment says it will only work from that version onwards. In which case my check won't pickup such jackets in books either. That is speculation but is what I would infer from the description you posted if I had written the code, but I am only guessing.

[theducks]

Originally Posted by kiwidude

Hmmm, you can't unless you use the version attached to this post. Run in debug and it should hopefully give you the book id and title.

The most likely cause I can think of for that is some sort of corruption, perhaps a file locked or something. Does Check Library show any issues?

Ran your new PI in debug mode. Ran fine

THEN I checked the DB for errors:
There were errors (file 632).

Thanks

[kovidgoyal]

Yes, the <meta> tag is what is used to identify jackets.

[DoctorOhh]

Originally Posted by kovidgoyal

Yes, the <meta> tag is what is used to identify jackets.

Regardless of the name of the file, that explains what I'm seeing. Thanks.

Originally Posted by kiwidude

EDIT: At purely a guess I would say perhaps in 0.6.50 they added the <meta> tag, which is why the comment says it will only work from that version onwards. In which case my check won't pickup such jackets in books either. That is speculation but is what I would infer from the description you posted if I had written the code, but I am only guessing.

You answered my question. I have quite a few that calibre recognizes as having a book jacket and automatically updates on a new conversion that the QA plugin marks as having no jacket. It seems these are labeled jacket1 vice jacket.

I'll send you a link to a sample book in PM.

[kiwidude]

Thanks Walt, I've changed the logic to cater for that jacket naming in the EPUB. I will include it in the next release shortly.

[nynaevelan]

Howdy Kiwidude:

Me again. A question for you in the check ISBN section, I have several books that are either only available as an ebook or I do not have an ISBN for them, instead I have Amazon's ASIN. Is there a way to avoid these books failing during the ISBN check? Also, I have some books that I created myself, therefore they do not have an ISBN either, I would like to have these books filtered out also. What would you suggest is the best way of going about doing this??

Nyn

[kiwidude]

Check ISBN only checks books that have a value in the ISBN, so for your own books you should not have anything in that field?

As for ASIN, maybe Kovid will correct me if wrong but I believe that once the new metadata API is finished in Calibre 0.8 that ASIN will be retrieved and stored in a separate ASIN identifier field. So Quality Check will be correct in flagging ASIN values that are in ISBN as these should be "moved" into the new proper place.

[nynaevelan]

Thank you, once again you set me on the right path. I will remove the ISBN numbers from my books and just use a saved search to filter out the ASIN books for now.

[kiwidude]

Changes in this release:

• For very small libraries ensure not divide by zero error from displaying status
• If an error retrieving the format path for a book, dump to debug output
• For the EPUB jackets check, include numbered jacket files from legacy Calibre conversions
• Fix bug in check no html comments which had incorrect logic after 1.3 rewrite
• Add ability to customise the menu to allowing hiding menu items not of interest

[Noughty]

Is it possible to add a feature to search for books without summaries? For books with no comments?

[kiwidude]

You don't need the plugin to do that:
comments:false

[nynaevelan]

Originally Posted by kiwidude

[*]Add ability to customise the menu to allowing hiding menu items not of interest

Works perfectly

[kiwidude]

Originally Posted by nynaevelan

Works perfectly

Phew. Thanks for the feedback. Hopefully that should be the last change for this plugin for a while - well until someone comes up with the next idea for it of course.

[ldolse]

Originally Posted by kiwidude

Phew. Thanks for the feedback. Hopefully that should be the last change for this plugin for a while - well until someone comes up with the next idea for it of course.

You jinxed yourself I've got a couple ideas I've been thinking about for a bit now. I think this plugin is probably the best destination for these functions, although currently the plugin is really a library quality check vs. a book/epub quality check, so I'm not certain if these would be better suited for another plugin.

• Check epub for Adobe xpgt files. Extra credit for tying in the Calibre feature in the conversion pipeline which removes the margin from these files.
• Check the epub manifest for embedded fonts
• Check to see if the file was created through Calibre's conversion pipeline (css might be the most reliable check for epub, not sure about other formats).
• Check for use of SVG files in epub (admittedly more complicated as they can either be defined inline in the html or as files.)