ValueError. need more than 1 value to unpack

[aleray]

Hi,

with the document below, I get the a "Need more than 1 value to unpack" ValueError:

# Hello{ {: id='foo' }

World

Here is the traceback:

☁  markdown_py -x extra Catalog-ExtendDimensions
Traceback (most recent call last):
  File "/usr/bin/markdown_py", line 34, in <module>
    run()
  File "/usr/lib/python2.7/site-packages/markdown/__main__.py", line 81, in run
    markdown.markdownFromFile(**options)
  File "/usr/lib/python2.7/site-packages/markdown/__init__.py", line 449, in markdownFromFile
    kwargs.get('encoding', None))
  File "/usr/lib/python2.7/site-packages/markdown/__init__.py", line 366, in convertFile
    html = self.convert(text)
  File "/usr/lib/python2.7/site-packages/markdown/__init__.py", line 307, in convert
    output = self.serializer(root)
  File "/usr/lib/python2.7/site-packages/markdown/serializers.py", line 275, in to_xhtml_string
    return _write_html(ElementTree(element).getroot(), format="xhtml")
  File "/usr/lib/python2.7/site-packages/markdown/serializers.py", line 197, in _write_html
    qnames, namespaces = _namespaces(root, default_namespace)
  File "/usr/lib/python2.7/site-packages/markdown/serializers.py", line 263, in _namespaces
    add_qname(key)
  File "/usr/lib/python2.7/site-packages/markdown/serializers.py", line 223, in add_qname
    uri, tag = qname[1:].split("}", 1)
ValueError: need more than 1 value to unpack

Thanks,

Alex

[waylan]

Interesting. This is the first I've ever seen an error from the serializer - which is a trimmed down version of the serializer that ships with the newest version of ElementTree. I wonder if this is really a bug in the serializer, an edge case where we're building an invalid elementtree, or a case of garbage in - garbage out. Either way, we need to address this. We shouldn't be crashing regardless of the input provided. Thanks for the report.

[waylan]

Did a little debugging on this. The attributes set on the h1 are the problem:

>>> h1.items()
[(u'{:', u'{:'), (u'id', u'foo')]

The colon is optional, so the first { starts the attr list, and the second {: is seen as a valueless key. However, a key of {: would not produce valid html - so the serializer is correct to choke on that. Guess the attr list parser needs to be improved.

[mitya57]

The new test passes with Python 2, but fails with Python 3:

ERROR: TestSyntax: "extensions/attr_list"
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/nose/case.py", line 198, in runTest
    self.test(*self.arg)
  File "/home/mandriver/Python-Markdown/build/tests/__init__.py", line 95, in __call__
    output = markdown.markdown(input, **get_args(file, config))
  File "/home/mandriver/Python-Markdown/build/lib/markdown/__init__.py", line 415, in markdown
    return md.convert(text)
  File "/home/mandriver/Python-Markdown/build/lib/markdown/__init__.py", line 304, in convert
    output = self.serializer(root)
  File "/home/mandriver/Python-Markdown/build/lib/markdown/serializers.py", line 275, in to_xhtml_string
    return _write_html(ElementTree(element).getroot(), format="xhtml")
  File "/home/mandriver/Python-Markdown/build/lib/markdown/serializers.py", line 197, in _write_html
    qnames, namespaces = _namespaces(root, default_namespace)
  File "/home/mandriver/Python-Markdown/build/lib/markdown/serializers.py", line 263, in _namespaces
    add_qname(key)
  File "/home/mandriver/Python-Markdown/build/lib/markdown/serializers.py", line 223, in add_qname
    uri, tag = qname[1:].split("}", 1)
ValueError: need more than 1 value to unpack