ValueError: All strings must be XML compatible: Unicode or ASCII, no NULL by

[nimblebooks]

What does this mean? (error is at bottom) building an ePub from a collection of HTML documents.

Where does it think the NULL character is?

...

Generating default cover
Traceback (most recent call last):
File "/usr/bin/ebook-convert", line 19, in <module>
sys.exit(main())
File "/usr/lib/calibre/calibre/ebooks/conversion/cli.py", line 254, in main
plumber.run()
File "/usr/lib/calibre/calibre/ebooks/conversion/plumber.py", line 886, in run
self.opts, self.log)
File "/usr/lib/calibre/calibre/ebooks/epub/output.py", line 176, in convert
oeb_output.convert(oeb, tdir, input_plugin, opts, log)
File "/usr/lib/calibre/calibre/ebooks/oeb/output.py", line 31, in convert
results = oeb_book.to_opf2(page_map=True)
File "/usr/lib/calibre/calibre/ebooks/oeb/base.py", line 1907, in to_opf2
results[NCX_MIME] = (href, self._to_ncx())
File "/usr/lib/calibre/calibre/ebooks/oeb/base.py", line 1880, in _to_ncx
self.toc.to_ncx(navmap)
File "/usr/lib/calibre/calibre/ebooks/oeb/base.py", line 1593, in to_ncx
element(point, NCX('content'), src=urlunquote(node.href))
File "/usr/lib/calibre/calibre/ebooks/oeb/base.py", line 240, in element
return etree.SubElement(parent, *args, **kwargs)
File "lxml.etree.pyx", line 2437, in lxml.etree.SubElement (src/lxml/lxml.etree.c:47345)
File "apihelpers.pxi", line 210, in lxml.etree._makeSubElement (src/lxml/lxml.etree.c:11391)
File "apihelpers.pxi", line 205, in lxml.etree._makeSubElement (src/lxml/lxml.etree.c:11327)
File "apihelpers.pxi", line 283, in lxml.etree._initNodeAttributes (src/lxml/lxml.etree.c:12162)
File "apihelpers.pxi", line 1242, in lxml.etree._utf8 (src/lxml/lxml.etree.c:19848)

[pdurrant]

Perhaps in one of the names of the HTML documents? It's hard to tell without looking at the source code.

[nimblebooks]

This is the command that invokes:

sudo ebook-convert tmp/temptoc.html zz_delivery/zimzaztest.epub

(run without options to isolate errors)

this is the directory it is running against

rw-r--r-- 1 bitnami root 105892 2011-11-04 14:21 1.html
-rw-r--r-- 1 bitnami root 318260 2011-11-04 14:21 2.html
-rw-r--r-- 1 bitnami root 47016 2011-11-04 14:21 3.html
-rw-r--r-- 1 bitnami root 204 2011-11-04 14:21 temptoc.html

I did a little digging around and used od -b to convert these files to octal, then grep ' 000' to see if there are any null characters; nope.

So I'm stumped.

The full script is pretty long and doing a lot of stuff to gather the html files so I am trying to avoid posting it.

[pdurrant]

Originally Posted by nimblebooks

this is the directory it is running against

rw-r--r-- 1 bitnami root 105892 2011-11-04 14:21 1.html
-rw-r--r-- 1 bitnami root 318260 2011-11-04 14:21 2.html
-rw-r--r-- 1 bitnami root 47016 2011-11-04 14:21 3.html
-rw-r--r-- 1 bitnami root 204 2011-11-04 14:21 temptoc.html

I have a vague memory that XML ids shouldn't begin with numbers, so it might be that the file names 1.html, etc are causing problems with various bits of the generated ePub. It might be worth trying again with them names a1.html, a2.html, etc.

[nimblebooks]

I substituted a new vanilla 1.html file for the 3 existing files and the build worked fine, so that probably wasn't the problem. The previous files are about Japanese aircraft carriers so it may have something to do with non-Unicode characters (just guessing). I am looking for a way to verify whether a file contains any non-Unicode characters.

[kovidgoyal]

From the traceback that looks like an outdated calibre version. I suggest updating to the latest, 0.8.24. Also on windows, calibre can occassionaly have trouble with files whose names contains non ascii characters, removing those characters usually resolves the issue.