TextCritical.net

The TOC cannot be generated because it seems to require ascii.

• https://github.com/aerkalov/ebooklib
• https://pypi.org/project/epub/
• https://github.com/wcember/pypub

Example of what is happening here:

s = 'main \xe1\xbc\x80\xe1\xbd\xb1\xce\xb1\xcf\x84\xce\xbf\xcf\x82'
s.decode('utf-8')
s.decode('utf-8').encode('ascii', 'ignore')

Trace:

• textcritical_net/src/epub.py: self.__writeTocNCX()
• textcritical_net/src/epub.py: fout.write(stream.render('xml'))

The error is generated from:

      <navLabel><text>${node.title}</text></navLabel>

Similar to: https://genshi.edgewall.org/ticket/542

Observations:

1. title is an instance of str with utf-8 encoding
2. The following does not work:
   1. title.decode("utf-8")
   2. title
   3. unicode(title)
   4. unicode(title.decode("utf-8"))
3. The following generates no errors but the content is wrong:
   1. title.decode("utf-8").encode("ascii", "ignore")
   2. title.decode("utf-8").encode("windows-1252", "ignore")
   3. title = Greek.unicode_to_beta_code(unicode(title.decode("utf-8")))

• Why do the main content pages not fail?
  • Genshi is not being imported for that rendering
• Would using Django templates work?
  • Could likely use http://jinja.pocoo.org/