[GUI Plugin] Quality Check

[kiwidude]

Changes in this release:

• Properly fix the spaces encoding when using Check unused image files

Thanks @PaulFiera - I used the wrong encoding function in the last release, darned copy/paste...

[kiwidude]

Originally Posted by kovidgoyal

Its not absolutely trivial. IIRC going into the Books section of the Fire requires two things: 1) a cdetype of EBOK 2) the presence of an asin field both in the EXTH header.

Also, the Fire has been known to crash with a few MOBI files when they have non Amazon asin values.

But if you want to do the check, use the MetadataHeader class, you will have to make a couple of changes to the EXTH class to store the raw cdetype and asin values so you plugin can handle them. Easily acheived by subclassing EXTH. See metadata/mobi.py for examples of using MetadataHeader.

Thanks very much Kovid, I had a suspicion it wasn't trivial. I will put it on the rainy day list to take a look at.

[paulfiera]

Originally Posted by kiwidude

Changes in this release:

• Properly fix the spaces encoding when using Check unused image files

Thanks @PaulFiera - I used the wrong encoding function in the last release, darned copy/paste...

You're welcome.

But still...


The space is encoded correctly, but nor the comma, which should be %2C.

Sorry. I know this is nitpicking and I wish they didn't name the files this way. I certainly don't when I'm developing for the Web.

EDIT:

Hold on because I'm finding a couple more:


I don't know which encoding functions are available to you but this looks like standard javascript escape() or encodeURIComponent()

[kiwidude]

There are days I hate Python... (or my lack of knowledge of it)... there is a *third* variant of url encoding function which I should have used... sigh. New release shortly.

[paulfiera]

Originally Posted by kiwidude

There are days I hate Python... (or my lack of knowledge of it)... there is a *third* variant of url encoding function which I should have used... sigh. New release shortly.

I believe you need to choose the one that resembles most javascript escape(). encodeURIComponent() won't encode apostrophes and they seem to be encoded in the epubs I have checked.

[kiwidude]

Changes in this release:

• Yet another tweak to Check unused image files to cope with characters like commas

[paulfiera]

Many thanks, kiwidude.

It's still finding images used in the epubs. Is your check case sensitive? In the same epub, in content.opf, the path to the images is "images", but in the html files is "Images".

Just say the word if you don't want to be bothered any more. I'm feeling like a PITA.

[kiwidude]

It converts everything to lowercase. And it doesn't use the path to the image, it strips that and just looks for the image name. If you have an epub causing you an issue, PM me a link to it.

[kiwidude]

Changes in this release:

• Add a "Search scope" submenu to allow searching selected books rather than all books.
• Change Check unused image files to look for both encoded and non encoded versions of the image names and ignore DRM ebooks

[paulfiera]

Originally Posted by kiwidude

Changes in this release:

• Add a "Search scope" submenu to allow searching selected books rather than all books.
• Change Check unused image files to look for both encoded and non encoded versions of the image names and ignore DRM ebooks

Fantastic addition the Search Scope. Great when adding new books and running the QC only against these new books.

Kudos!!!

[CRussel]

Originally Posted by kiwidude

Thanks very much Kovid, I had a suspicion it wasn't trivial. I will put it on the rainy day list to take a look at.

Somehow, I knew it wasn't trivial, but I could hope.

Kovid is right, it takes both an Amazon ID AND the EBOK tag. The Amazon ID is automatic in my workflow these days, though may well be missing for older books.

Well, if you could do it, it would be a good thing for this Fire user. And I suspect for others, since I know it's a topic that comes up there often.

[jackie_w]

Originally Posted by paulfiera

Fantastic addition the Search Scope. Great when adding new books and running the QC only against these new books.

Kudos!!!

+1 to that. Karma to kiwidude

[Pierre Menard]

Originally Posted by kiwidude

Add a "Search scope" submenu to allow searching selected books rather than all books.

Aw, and here I'd finally just about trained myself to remember to use the "restrict to."

The ability to search for books that have had a calibre conversion has been a fantastic tool--can't tell you how much I appreciate that. So much, in fact, that I'm wondering if that sort of functionality can be extended a bit.

Some of the books I have have not been converted, but they have been edited in Sigil. Sigil seems to add a line to the opf file to indicate this:

<meta content="0.4.0" name="Sigil version" />

I dislike Sigil and don't really touch it, so I'm not sure if all versions or all edits automatically add this line, but I see it often enough. I'd love to be able to automate identifying those.

In fact, there's a few different things that I sometimes wish to check files for. I have in my head an image of the plug-in prompting me for what word I want to look for (eg, "sigil"), maybe even asking if I want to check in the opf file or the html. If that's any sort of possibility and not just crazy talk from someone who has no idea how this actually works, that would be incredibly useful.

[paulfiera]

Originally Posted by Pierre Menard

I have in my head an image of the plug-in prompting me for what word I want to look for (eg, "sigil"), maybe even asking if I want to check in the opf file or the html. If that's any sort of possibility and not just crazy talk from someone who has no idea how this actually works, that would be incredibly useful.

Actually I had the same idea for an enhancement request but perhaps simpler.

An input text field where you could enter any text and search the whole library or a scope.

In my case, I have many epubs generated from MS Word. Normally, these epubs do not have the best formatting and I would like to be able to enter "Microsoft" in an input field and get all the books where that word exists. In the case of MS Word generated epubs there's usually a meta tag defining this.

The possibility of find a string in your library or a scope would be absolutely great.

[kiwidude]

Something like this perhaps... ?