Fix for Kindle 3 problem with Sections and Articles

[nickredding]

Originally Posted by kovidgoyal

Merged your code with some minor cleanups:

http://bazaar.launchpad.net/~kovid/c.../mobi/writer2/

Incidentally, have you tested a recipe with a single feed, that becomes a "blog" on the Kindle?

Yes, the code inserts the periodical and (single) section TOC but because the MOBI type is 102 the Kindle v3.1 doesn't provide the Sections and Articles view--just the Articles List. It works fine, though.

[nickredding]

Kovid - there is a strange problem with the mobi writer2 code--it seems to shuffle the images in the periodical somehow (it looks random) so images appear in the wrong articles. I checked the mobi writer2 code with and without my changes--same result. I also checked OEB recipe output and the images are in the right places, so it is something that happens in MOBI conversion. I can't see the issue immediately and I'll try t run it down, but I thought I'd give you the heads up in case you have any ideas what might be causing this.

[kovidgoyal]

Probably a bug in the code that maps image urls to indexes. See generate_images(). I vaguely recall noting to myself that I needed to test that code more thoroughly. I'm a little busy right now so I wont be able to look at it until later.

[kovidgoyal]

Fix committed.

[PeterT]

Originally Posted by kovidgoyal

I'm a little busy right now so I wont be able to look at it until later.

Originally Posted by kovidgoyal

Fix committed.

I'd hate to see how long it takes to fix something when you are NOT busy!

[kovidgoyal]

When I'm not busy, I'm usually goofing off, so it would take considerably longer

[nickredding]

I get this on all recipes with the 10200 revision

34% Download finished
Parsing all content...
Python function terminated unexpectedly
  '2to3' (Error Code: 1)
Traceback (most recent call last):
  File "site.py", line 132, in main
  File "site.py", line 109, in run_entry_point
  File "C:\Users\Nick\calibre\src\calibre\ebooks\conversion\cli.py", line 286, in main
    plumber.run()
  File "C:\Users\Nick\calibre\src\calibre\ebooks\conversion\plumber.py", line 946, in run
    encoding=self.input_plugin.output_encoding)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\conversion\plumber.py", line 1084, in create_oebbook
    reader()(oeb, path_or_stream)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\oeb\reader.py", line 71, in __call__
    self._all_from_opf(opf)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\oeb\reader.py", line 610, in _all_from_opf
    self._metadata_from_opf(opf)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\oeb\reader.py", line 132, in _metadata_from_opf
    mi = OPF(stream).to_book_metadata()
  File "C:\Users\Nick\calibre\src\calibre\ebooks\metadata\opf2.py", line 609, in to_book_metadata
    ans = MetaInformation(self)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\metadata\__init__.py", line 264, in MetaInformation
    return Metadata(title, authors, other=mi)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\metadata\book\base.py", line 108, in __init__
    self.smart_update(other)
  File "C:\Users\Nick\calibre\src\calibre\ebooks\metadata\book\base.py", line 567, in smart_update
    other_lang = getattr(other, 'languages', [])
  File "C:\Users\Nick\calibre\src\calibre\ebooks\metadata\opf2.py", line 955, in fget
    l = canonicalize_lang(t.strip())
  File "C:\Users\Nick\calibre\src\calibre\utils\localization.py", line 216, in canonicalize_lang
    m2to3 = iso639['2to3']
KeyError: '2to3'

[nickredding]

Kovid - I just saw your message about running from source and today's release, and that was the problem.