[GUI Plugin] EpubSplit

[JimmXinu]

I'm not sure what you mean--when splitting an epub, you are creating new books. Naturally, the metadata gets updated.

But you can control which metadata is copied, and prevent the cover from being copied, in the EpubSplit configuration.

[JSWolf]

Originally Posted by JimmXinu

I'm not sure what you mean--when splitting an epub, you are creating new books. Naturally, the metadata gets updated.

But you can control which metadata is copied, and prevent the cover from being copied, in the EpubSplit configuration.

I did uncheck all the fields in the plugin and it seemed to work except, when the plugin creates the OPF, it does not add proper line endings so the OPF entries are just one long line.

Also, what I'd like is the top of the original OPF copied down to the </metadata> copied over as the top of the metadata. This gives the original info which I can then change as I want. Calibre will have done nothing to monkey about with the code.

[JimmXinu]

Originally Posted by JSWolf

I did uncheck all the fields in the plugin and it seemed to work except, when the plugin creates the OPF, it does not add proper line endings so the OPF entries are just one long line.

Also, what I'd like is the top of the original OPF copied down to the </metadata> copied over as the top of the metadata. This gives the original info which I can then change as I want. Calibre will have done nothing to monkey about with the code.

Is there anything actually broken or inaccurate in the OPF generated?

Because it sounds like you are just being extremely particular about the appearance and format of a meta data file that isn't intended for users to be viewing directly anyway...

[JSWolf]

Originally Posted by JimmXinu

Is there anything actually broken or inaccurate in the OPF generated?

Because it sounds like you are just being extremely particular about the appearance and format of a meta data file that isn't intended for users to be viewing directly anyway...

Take a look at this OPF file. It's a mess because there are no proper line endings.

<?xml version="1.0" encoding="utf-8"?><package unique-identifier="epubsplit-id" version="2.0" xmlns="http://www.idpf.org/2007/opf"><metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:opf="http://www.idpf.org/2007/opf"><dc:identifier id="epubsplit-id">epubsplit-uid-1402088581</dc:identifier><dc:title>Spenser 05 - The Judas Goat Split</dc:title><dc:creator opf:role="aut">Robert B. Parker</dc:creator><dc:contributor opf:role="bkp">epubsplit</dc:contributor><dc:rights>Copyrights as per source stories</dc:rights><dc:language>eng</dc:language><dc:description>&lt;div&gt;&lt;p class=&quot;description&quot;&gt;Split from:&lt;/p&gt;&lt;p class=&quot;description&quot;&gt;&lt;/p&gt;&lt;p class=&quot;description&quot;&gt;Spenser has gone to London—and not to see the Queen. He's gone to track down a bunch of bombers who've blown away his client's wife and kids. His job is to catch them. Or kill them. His client isn't choosy.&lt;br&gt;&lt;br&gt;But there are nine killers to one Spenser—long odds. Hawk helps balance the equation. The rest depends on a wild plan. Spenser will get one of the terrorists to play Judas Goat—to lead him to others. Trouble is, he hasn't counted on her being very blond, very beautiful and very dangerous.&lt;/p&gt;&lt;/div&gt;</dc:description><meta name="cover" content="coverimageid"/></metadata><manifest><item href="toc.ncx" id="ncx" media-type="application/x-dtbncx+xml"/><item href="cover.jpg" id="coverimageid" media-type="image/jpeg"/><item href="cover.xhtml" id="cover" media-type="application/xhtml+xml"/><item href="OEBPS/Park_9780307569660_epub_cvi_r1.htm" id="a0" media-type="application/xhtml+xml"/><item href="OEBPS/Park_9780307569660_epub_adc_r1.htm" id="a1" media-type="application/xhtml+xml"/><item href="OEBPS/Park_9780307569660_epub_tp_r1.htm" id="a2" media-type="application/xhtml+xml"/><item href="OEBPS/Park_9780307569660_epub_c01_r1.htm" id="a3" media-type="application/xhtml+xml"/><item href="OEBPS/Park_9780307569660_epub_c02_r1.htm" id="a4" media-type="application/xhtml+xml"/><item href="OEBPS/Park_9780307569660_epub_css_r1.css" id="a5" media-type="text/css"/><item href="OEBPS/images/Park_9780307569660_epub_tp_r1.jpg" id="a6" media-type="image/jpeg"/><item href="OEBPS/images/Park_9780307569660_epub_cvi_r1.jpg" id="a7" media-type="image/jpeg"/></manifest><spine toc="ncx"><itemref idref="cover" linear="yes"/><itemref idref="a0" linear="yes"/><itemref idref="a1" linear="yes"/><itemref idref="a2" linear="yes"/><itemref idref="a3" linear="yes"/><itemref idref="a4" linear="yes"/></spine><guide><reference href="cover.xhtml" title="Cover" type="cover"/></guide></package>

As you saw in the code box, the OPF has no line endings and thus, I have to do the work of manually adding them in.

[jackie_w]

The Beautify file(s) option in the calibre Editor will tidy up the opf (and/or any other file) for missing line ends.

[JSWolf]

Originally Posted by jackie_w

The Beautify file(s) option in the calibre Editor will tidy up the opf (and/or any other file) for missing line ends.

Sorry, doesn't work. Then I end up with XML files that I have to edit because of the new formatting. Yes, sometimes that helps, but not always and not in most cases.

All we need is line endings put in the proper places in the new OPF. Unchecking all of the fields in the plugin options solved the first problem and now it's led to a new problem that once solved it's good to go.

[JimmXinu]

Attached is a test version that uses .toprettyxml(indent=' '...) instead of just .toxml() for the content.opf and toc.ncx files.

I honestly don't see the existing output as problem--it's a machine written file for machines to read. But I'm willing to change to pretty print output as long as it doesn't break machine readability.

I don't remember the details, but I do have some past evidence that pretty printed XML was a problem at one time.

[JSWolf]

Originally Posted by JimmXinu

Attached is a test version that uses .toprettyxml(indent=' '...) instead of just .toxml() for the content.opf and toc.ncx files.

I honestly don't see the existing output as problem--it's a machine written file for machines to read. But I'm willing to change to pretty print output as long as it doesn't break machine readability.

I don't remember the details, but I do have some past evidence that pretty printed XML was a problem at one time.

Thanks. I'll give this a try sometime this weekend when I have the time and let you know how it goes. I never thought to check the NCX file. Thanks for that.

[JimmXinu]

With Kovid's gracious help, I've switched over to using Transifex for translations for my plugins.

You can now help out by providing translations without having to install anything. All you need is to create an account on a web site and jump in to translating. The link to EpubSplit's translations specifically is here.

Thanks!

[JimmXinu]

Version 1.3.4 - 02 Jul 2014
Update Spanish and add Ukrainian translations, courtesy of the volunteers at Transifex.
Pretty format XML of generated metadata files.

[JimmXinu]

Attached is a test version that should work for both the existing 1.X calibre and the upcoming 2.X version calibre. 2.x will use PyQt5 which requires changes to virtually all plugins.

This version needs to be tested with the existing calibre.

Brave souls to test with with the beta 2.X version would also be appreciated. Links to the 2.X beta and discussion of the transition (for plugins) can be see at: Attn plugin developers: calibre moving to Qt 5.

[samy2]

I tested it briefly under Linux. Seem to work well. Thank you!

[BetterRed]

@JimmXinu - tested with V1.201 portable and V1.44 64 bit - works fine for me

BR

[JimmXinu]

Originally Posted by BetterRed

@JimmXinu - tested with V1.201 portable and V1.44 64 bit - works fine for me

BR

Thanks!

[gbm]

Originally Posted by JimmXinu

Attached is a test version that should work for both the existing 1.X calibre and the upcoming 2.X version calibre. 2.x will use PyQt5 which requires changes to virtually all plugins.

This version needs to be tested with the existing calibre.

Brave souls to test with with the beta 2.X version would also be appreciated. Links to the 2.X beta and discussion of the transition (for plugins) can be see at: Attn plugin developers: calibre moving to Qt 5.

When tying to save setting in Customize Plugin had the following error using calibre 1.201.1

calibre 1.201.2  isfrozen: True is64bit: True
Linux-3.13.0-24-generic-x86_64-with-debian-jessie-sid Linux ('64bit', 'ELF')
('Linux', '3.13.0-24-generic', '#46-Ubuntu SMP Thu Apr 10 19:11:08 UTC 2014')
Python 2.7.8
Linux: ('debian', 'jessie/sid', '')
Successfully initialized third party plugins: DeDRM && Baen && EpubSplit && Modify ePub
Traceback (most recent call last):
  File "site-packages/calibre/gui2/preferences/plugins.py", line 292, in customize_plugin
  File "site-packages/calibre/gui2/preferences/plugins.py", line 368, in modify_plugin
  File "site-packages/calibre/customize/__init__.py", line 171, in do_user_config
  File "calibre_plugins.epubsplit.__init__", line 81, in save_settings
  File "calibre_plugins.epubsplit.config", line 171, in save_settings
AttributeError: 'NoneType' object has no attribute 'toString'

Running Linux Mint 17 Cinnamon 64-bit

bernie