Calibre conversion adding classes and screwing the layout?

1v4n0 · 01-23-2015, 05:59 AM

One small bug I've noticed.

Lately, in the epubs, I've started to wrap differently formatted paragraphs inside divs, instead of giving a class to each .

So if until some time ago I'd simply go with something like

Code:

<p> bla bla bla goes like this:</p>
<p class="quotations">This is how the world ends </p>
<p class="quotations">Not with a bang, but with a  whimper</p>
<p> cool, innit?</p>

, recently I thought that it might be cleaner (and easier with top and bottom margin) to switch to

Code:

<p> bla bla bla goes like this:</p>
<div class="quotations">
<p>This is how the world ends </p>
<p>Not with a bang, but with a  whimper</p>
</div>
<p> cool, innit?</p>

Possibly with some minor changes to the css as well.

I noticed, however, that sometimes the conversion to azw3 with calibre screws this layout. As you might know, Calibre add its own classes, named "calibre" followed by a number. I don't like this (oh and btw, any idea on how to quickly get rid of them? Besides removing them with a regex and c/p the original css from the epub), but I don't know if this is the reason.

Anyway this does not happen all the time. I tried with a small file and it works fine, but if I convert the whole book it doesn't.

I attach the 2 epub files.

Any ideas? Ty

pssquirrel · 01-23-2015, 03:25 PM

Quote:

Originally Posted by 1v4n0

...Calibre add its own classes, named "calibre" followed by a number. I don't like this (oh and btw, any idea on how to quickly get rid of them?

I don't have any insight into your main question, but re: generic Calibre classes:

Calibre seems to create a class for every element. I have found that if the element already has a class named, Calibre will respect that name and use it. But if there is no class named, Calibre will create one and call it Calibre1, Calibre2, etc. This happens with every element, including <body> <img> , etc. If you want to avoid generic Calibre classes, add a class to every element: <body class="ebook"> <img class="picture">, , , etc.

kovidgoyal · 01-23-2015, 08:07 PM

calibre *flattens* all css, that means that every CSS rule in the source document (including implicit rules) are converted to class based rules. This is done for excellent reasons, that I have explained many times before, a bit of googling should find you the explanation.

As for your issue, I dont have time to look at it right now, but from your description it sounds like you need to turn off the option to "Remove fake margins" under structure detection in the conversion dialog.

Agama · 01-25-2015, 07:04 AM

Quote:

Originally Posted by 1v4n0

As you might know, Calibre add its own classes, named "calibre" followed by a number. I don't like this (oh and btw, any idea on how to quickly get rid of them? Besides removing them with a regex and c/p the original css from the epub)

My solution for this is a plugin, (a really great calibre feature), which uses a set of customisable regexes and some other custom tidy-up processing. However I am working with Markdown sources, (with my own class extensions), and so the ePub output is predictable. You may be able to do this if your output is also predictable. This fix is quick, practically instant on any ePub, once you have written your plugin.

eschwartz · 01-25-2015, 11:48 AM

You can always use kindlegen...

1v4n0 · 01-26-2015, 02:54 PM

Quote:

Originally Posted by kovidgoyal

turn off the option to "Remove fake margins" under structure detection in the conversion dialog.

Whoa, a reply from The Man Himself. No wonder it's correct

Fixed, ty

1v4n0 · 01-29-2015, 12:57 PM

I got a new one though: I used this solution for numbering the lines. The numbers were wrapper inside a

Code:

<span class="number">

, but calibre decided that from the second span on the class had to be "number1", and the font-size had to be smaller.

I easily fixed it from the stylesheet, but I don't undestand why it happened.

01-29-2015, 12:57 PM	#7
1v4n0 Groupie Posts: 171 Karma: 40000 Join Date: Oct 2013 Device: kindle	I got a new one though: I used this solution for numbering the lines. The numbers were wrapper inside a Code: <span class="number"> , but calibre decided that from the second span on the class had to be "number1", and the font-size had to be smaller. I easily fixed it from the stylesheet, but I don't undestand why it happened.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Automatic entity conversion screwing up search and replacerch and replace	ldolse	Sigil	10	01-05-2012 01:46 AM
txt to Epub - adding classes with textile	Nareshkendel	Conversion	4	11-23-2011 05:30 PM
Why, During ePub to ePub Conversion does Calibre changes existing CSS classes?	JSWolf	Conversion	7	08-06-2011 07:32 PM
Is there a way to prevent Calibre adding to the TOC upon conversion?	PatNY	Conversion	21	03-17-2011 03:09 PM
Calibre conversion without adding to library	Starganderfish	Calibre	2	01-02-2011 04:31 AM

01-23-2015, 08:07 PM	#3
kovidgoyal creator of calibre Posts: 43,835 Karma: 22666666 Join Date: Oct 2006 Location: Mumbai, India Device: Various	calibre flattens all css, that means that every CSS rule in the source document (including implicit rules) are converted to class based rules. This is done for excellent reasons, that I have explained many times before, a bit of googling should find you the explanation. As for your issue, I dont have time to look at it right now, but from your description it sounds like you need to turn off the option to "Remove fake margins" under structure detection in the conversion dialog.

01-25-2015, 11:48 AM	#5
eschwartz Ex-Helpdesk Junkie Posts: 19,422 Karma: 85397180 Join Date: Nov 2012 Location: The Beaten Path, USA, Roundworld, This Side of Infinity Device: Kindle Touch fw5.3.7 (Wifi only)	You can always use kindlegen...