Bug #626
utf8 in conditions in templateparser goes awry
Description
<a href="" class="button">↑</a> is handled just fine, however:
{{ if [paragraph:sort] > 0}} <a href="" class="button">↑</a> {{ endif }}
Gives me an error:
Traceback (most recent call last):
File "/usr/lib/python2.6/dist-packages/mod_python/importer.py", line 1537, in HandlerDispatch
default=default_handler, arg=req, silent=hlist.silent)
File "/usr/lib/python2.6/dist-packages/mod_python/importer.py", line 1229, in _process_target
result = _execute_target(config, req, object, arg)
File "/usr/lib/python2.6/dist-packages/mod_python/importer.py", line 1128, in _execute_target
result = object(arg)
File "/home/underdark/underdark/libs/uweb/__init__.py", line 110, in RequestHandler
response = pages.InternalServerError(*sys.exc_info())
File "/home/underdark/underdark/libs/uweb/pagemaker/__init__.py", line 329, in InternalServerError
'traceback': self._ParseStackFrames(traceback)}))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 263, in Parse
return SafeString(''.join(tag.Parse(**kwds) for tag in self))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 263, in <genexpr>
return SafeString(''.join(tag.Parse(**kwds) for tag in self))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 459, in Parse
output.append(''.join(tag.Parse(**replacements) for tag in self))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 459, in <genexpr>
output.append(''.join(tag.Parse(**replacements) for tag in self))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 459, in Parse
output.append(''.join(tag.Parse(**replacements) for tag in self))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 459, in <genexpr>
output.append(''.join(tag.Parse(**replacements) for tag in self))
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 564, in Parse
value = TAG_FUNCTIONS[func](value)
File "/home/underdark/underdark/libs/uweb/templateparser.py", line 632, in HtmlEscape
text = unicode(text)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 500: ordinal not in range(128)
Associated revisions
History
#1 Updated by Jan Klopper about 12 years ago
Hmm, untrue,
It apears it works fine if I use the correct variable.
However.
I would expect an index error / instead of an encode error if 'sort' doesn't exist.
#2 Updated by Elmer de Looff about 12 years ago
- Description updated (diff)
#3 Updated by Elmer de Looff about 12 years ago
- Status changed from New to In Progress
- % Done changed from 0 to 30
That's a very weird error. These are my results when doing some quick checks (there definitely is an error here):
Encoded UTF8 in the template text (that is, not in a replacement tag, as in your example):
>>> import templateparser
>>> tpl = u'We \u2665 Python'
>>> templateparser.Template(tpl.encode('utf8').Parse()
'We \xe2\x99\xa5 Python'
This works, as expected. The raw bytestream that goes in, is sent raw to the output. Unicode codepoint U2665 translates correctly to the 3-byte UTF8 string. Parsing the same template without first encoding the template to UTF8 (or any other bytestream, though that will result in character set mismatches, and is a very bad idea); that is, storing raw Unicode in the template, does result in an error:
>>> import templateparser
>>> tpl = u'We \u2665 Python'
>>> templateparser.Template(tpl).Parse()
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/elmer/underdark/libs/uweb/templateparser.py", line 248, in __init__
self.AddString(raw_template)
File "/home/elmer/underdark/libs/uweb/templateparser.py", line 308, in AddString
self._ExtendText(node)
File "/home/elmer/underdark/libs/uweb/templateparser.py", line 365, in _ExtendText
for node in self.TagSplit(node):
File "/home/elmer/underdark/libs/uweb/templateparser.py", line 329, in TagSplit
yield TemplateText(node)
UnicodeEncodeError: 'ascii' codec can't encode character u'\u2665' in position 3: ordinal not in range(128)
The error here is what we expect, the Unicode codepoint is not in the ASCII character set. This is a shortcoming of TemplateParser and I'll look into fixing this.
Back to your problem, the reported error mentions \xe2. This is likely not to be a Unicode codepoint (the associated character is Î), but the first character of the up-arrow in UTF8: \xe2\x86\x91.
What I suspect goes wrong here is that when requesting the :sort index, you get the sort method of whatever object you're working on. If the representation of this object contains raw Unicode, the templateparser barfs on this. However, this does still not explain the exact error given, because that does not contain a Unicode codepoint, but the first byte of a multi-byte sequence.
>>> import templateparser
>>> tpl = '[obj:sort]'
>>> tp.Template(tpl.encode('utf8')).Parse(obj=[])
'<built-in method sort of list object at 0x273f998>'
#4 Updated by Elmer de Looff about 12 years ago
- Status changed from In Progress to Resolved
- % Done changed from 30 to 70
Applied in changeset commit:403be453c6ae.
#5 Updated by Elmer de Looff about 12 years ago
- Status changed from Resolved to Feedback
- Assignee changed from Elmer de Looff to Jan Klopper
- Priority changed from High to Normal
Jan,
The main bug has been solved. Please see if you can reproduce the bug you encountered earlier, I'd like to know what causes it.
#6 Updated by Elmer de Looff about 12 years ago
- Status changed from Feedback to Resolved
Applied in changeset commit:138c5a8a1c0e.
#7 Updated by Elmer de Looff about 12 years ago
Applied in changeset 55f11f246503.
#8 Updated by Elmer de Looff about 12 years ago
Applied in changeset 55f11f246503.
#9 Updated by Jan Klopper about 12 years ago
- Assignee changed from Jan Klopper to Elmer de Looff
When triggering this bug now, i get the following error:
<class 'underdark.libs.uweb.templateparser.TemplateKeyError'>
Item has no index, key or attribute 'sort'.
Seems to work as expected, thnx!
#10 Updated by Elmer de Looff about 12 years ago
- Status changed from Resolved to Closed
- % Done changed from 70 to 100
In that case, no problem remains (there might be one hidden somewhere, but without any way to test this, I'll close it for now).
#11 Updated by Elmer de Looff about 12 years ago
- Category set to TemplateParser
Added tests to the TemplateParser suite to test for Unicode support in templates. They may now contain raw codepoints, and these are converted to UTF8. This resolves #626.