[GUI Plugin] Modify ePub

[kiwidude]

@zetmolm - I'm not overly keen on doing so to be honest. There are a bunch of complications around it - for instance it is only going to work if (a) the book has a stylesheet, and (b) each html page links to that stylesheet. I don't like the idea of adding a feature that only works "some of the time" and even more the fact that it is to workaround a bug in someone else's software that in all likelihood could get easily fixed in another version. I known absolutely nothing about FBReader but I would suggest either chasing up whoever develops that or find another bit of reader software. This just has a "too bespoke" smell to me.

Of course if it was me in your situation and there was no alternatives I would say to heck with it and add it, but I'm not, so I'm going to use that time on other things instead .

[Zetmolm]

@Kiwidude: Thanks for your explanation. I see your points. I had hoped I asked for something simple, but apparently I did not. You are right that it would be better to get the bug resolved rather than work around it. But then, I am afraid that requests sent to those developers may end up in some sort of bottomless pit, whereas you actually respond to requests (even though it is a 'no' this time). Well, I'll give it a try anyway, and in the meantime use alternative solutions (edit the ePub manually / use another reader program / convert ePubs to another format).

[JSWolf]

Any chance of being able to access the font subsetting?

[Man Eating Duck]

Originally Posted by JSWolf

Any chance of being able to access the font subsetting?

Alas, no. That would have been handy

Here's what kiwidude wrote about it:

Originally Posted by kiwidude

@Man Eating Duck - the code Kovid has written is dependent on having the data structure used by the conversion pipeline that Modify ePub intentionally avoids using. Which means I have to rewrite what he has done basically. Given that it is listed as "experimental" currently if it ever happens then it is only going to be after it has been in calibre for a while longer as I don't want to have to continually be changing my version of it.

[pirl8]

@KiwiDude

Could it be possible to implement this in Modify ePub?

[JSWolf]

Originally Posted by pirl8

@KiwiDude

Could it be possible to implement this in Modify ePub?

Do you know how much larger the eBook would actually be with &shy added to every place possible that a hyphen could be? Given that the PW has less memory, I don't think this is realistic.

[pirl8]

Originally Posted by JSWolf

Do you know how much larger the eBook would actually be with &shy added to every place possible that a hyphen could be?

Yes, of course I know.

The html part is rougly twice the original, estimated with a prefix/suffix of 2 characters, which gives the highest amount of hyphenation, considering an average of 2.5 characters per syllable and estimating 5 characters to store the "­" entity (which gets actually easily compressed).

Spoiler:

Just put the HTML files in a directory and run this script to have an estimate

#!/bin/bash
for F in *.html *.xhtml
  do lynx --assume-charset UTF8 --dump $F | 
  tr '[:alpha:]' x | 
  tr -cs 'x' '[\n*]' 
done | 
sort | 
uniq -c | 
sort -n | 
grep xxxx | 
awk 'BEGIN { \
        SL=2.5; \
        TOT=0; \
        SHY=5; \
    } \
    { 
        print $1 " instances of " length($2) " character long words" ; \
        TOT += $1 * (length($2) / SL + 0);
    } \
    END {
        print "Estimated text increase: " SHY*TOT \
    } ' 
echo -n "Total text: "
du -c -b *.html | tail -1 | awk '{print $1}'

There are no memory concerns (embedding fonts - even with subsampling - or adding oversized images for a best view with KPW which has an higher resolution uses much more memory and images cannot be compressed like text) and - at the moment - this is the ONLY way to have hyphenation on AZW3 files on e-ink Kindles.

[calvin]

I would like it too if the &shy/hyphen feature could be implemented.

The epub has to be properly tagged, especially the language tag and there has to be an algorithm for detecting the syllables. But that should be possible because ixtab has implemented this in his kindle jbpatch package.

[pirl8]

Originally Posted by calvin

The epub has to be properly tagged, especially the language tag and there has to be an algorithm for detecting the syllables. But that should be possible because ixtab has implemented this in his kindle jbpatch package.

The algorithm is widely available (there's for sure a python class which implements the TeX algorithm). It should be normally implemented in the reader. Ixtab has found a way to inject code in the java framework of the kindle and thus he could modify the Mobi reader to implement hyphenation. This is not possible with the AZW3 reader which is a standalone application and does not rely on that framework.

It seems to me such a nonsense that Amazon has implemented a reader which can handle the ­ entity without integrating the hyphenation algorithm...

[kerb]

Hello,

I have a Kobo Glo and I use it to read .epub files. However, if I want to have the full kobo reading features I have to change the file to kepub.epub. The device can read the file with no problem. However, "modify ePub" does not work with this format as the cover will be missing and only the title of the book will show in the device.

Any suggestions as to what I could do to get my cover back?

Thanks

[davidfor]

Originally Posted by kerb

Hello,

I have a Kobo Glo and I use it to read .epub files. However, if I want to have the full kobo reading features I have to change the file to kepub.epub. The device can read the file with no problem. However, "modify ePub" does not work with this format as the cover will be missing and only the title of the book will show in the device.

Any suggestions as to what I could do to get my cover back?

Thanks

This happens on the Kobo device. It is how they work. There is absolutely nothing that the Modify ePub or any other of kiwidudes plugins can do about it. If you want to discuss this, it belongs in the Kobo forum.

[kerb]

Originally Posted by davidfor

This happens on the Kobo device. It is how they work. There is absolutely nothing that the Modify ePub or any other of kiwidudes plugins can do about it. If you want to discuss this, it belongs in the Kobo forum.

The reason why I posted my query here is that there is someone who has already modified the plugin to solve this matter (see by the very end of the entry). Unfortunately, his modified plugin does not work for me and he does not offer support, which is why I was wondering whether this could be made "official"

http://dsandrei.blogspot.fr/2012/07/koboish-ebooks.html

[Thasaidon]

Originally Posted by kerb

The reason why I posted my query here is that there is someone who has already modified the plugin to solve this matter (see by the very end of the entry). Unfortunately, his modified plugin does not work for me and he does not offer support, which is why I was wondering whether this could be made "official"

http://dsandrei.blogspot.fr/2012/07/koboish-ebooks.html

I would be interested in plug-in that Kepubed ePub files and others have expressed similar views in the past.

[davidfor]

Originally Posted by kerb

The reason why I posted my query here is that there is someone who has already modified the plugin to solve this matter (see by the very end of the entry). Unfortunately, his modified plugin does not work for me and he does not offer support, which is why I was wondering whether this could be made "official"

http://dsandrei.blogspot.fr/2012/07/koboish-ebooks.html

What he did was modify the plugin to change the code inside the epub so that annotations worked. This has no relationship to the covers. This happens purely on the Kobo device.

Also, the modifications to convert from ePub to kepub is NOT something that should be in the Modify ePub plugin. What those changes are doing is to convert the file to another format. That is not a sensible thing for this plugin.

[kerb]

Thanks for all your answers. I think you're absolutely right as it seems to be about the Calibre Kobo driver. I went back to the blog entry I posted and read all the comments to see if there was something I was missing.

Thanks so much to Joel Goguen. He has published a driver which sorts out the problem of covers for kobofied files. I'm so happy now!!! Just follow the instructions

https://github.com/jgoguen/calibre-kobo-driver#readme

His comment is at the very end of the blog entry
http://dsandrei.blogspot.co.uk/2012/...19243870487287