epub2lrf creates file that raises UnicodeDecodeError when loaded

[jace]

I used eCub to generate an epub file from HTML sources. This file is viewable using Calibre's viewer, but not on my PRS-505, where all pages show as blank.

To get around this, I used epub2lrf, but the resultant file freezes the PRS-505 and shows the following error in Calibre's viewer.

  File "build/bdist.linux-i686/egg/calibre/ebooks/lrf/lrfparser.py", line 160, in main
  File "build/bdist.linux-i686/egg/calibre/ebooks/lrf/lrfparser.py", line 28, in parse
  File "build/bdist.linux-i686/egg/calibre/ebooks/lrf/lrfparser.py", line 56, in _parse_objects
  File "build/bdist.linux-i686/egg/calibre/ebooks/lrf/objects.py", line 844, in initialize
  File "build/bdist.linux-i686/egg/calibre/ebooks/lrf/objects.py", line 705, in add_text
  File "/usr/lib/python2.6/encodings/utf_16_le.py", line 16, in decode
    return codecs.utf_16_le_decode(input, errors, True)
UnicodeDecodeError: 'utf16' codec can't decode bytes in position 262-263: illegal UTF-16 surrogate

The attached book's HTML sources are available from  http://twobits.net/

[jace]

Two Bits (epub file)

[kovidgoyal]

Instead of converting to LRf try converting the EPUB to EPUB

[jace]

Oh wow, that fixed it. Who would have thought of epub to epub conversion?