This doesn't fail for me in the host system, even with LC_ALL=C or env -i.
OTOH in sbuild there are additional failures, like this:
________________ EncodingTest.test_unicode_literal_in_expr_file ________________
Traceback (most recent call last):
File "/<<BUILDDIR>>/mako-1.0.6+ds1/test/test_template.py", line 168, in
test_unicode_literal_in_expr_file
lambda t:t.strip()
File "/<<BUILDDIR>>/mako-1.0.6+ds1/test/__init__.py", line 43, in
_do_file_test
unicode_=unicode_, template_args=template_args)
File "/<<BUILDDIR>>/mako-1.0.6+ds1/test/__init__.py", line 56, in _do_test
output = template.render_unicode(**template_args)
File "/<<BUILDDIR>>/mako-1.0.6+ds1/mako/template.py", line 471, in
render_unicode
as_unicode=True)
File "/<<BUILDDIR>>/mako-1.0.6+ds1/mako/runtime.py", line 839, in _render
return context._pop_buffer().getvalue()
File "/<<BUILDDIR>>/mako-1.0.6+ds1/mako/util.py", line 160, in getvalue
return self.delim.join(self.data)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 63:
ordinal not in range(128)
There are 5 of them, with the same mako/template.py:471 cause.
Anyway, I tried to compare the
ExceptionsTest::test_utf8_format_exceptions_pygments behavior in sbuild
and in the host system, it turned out the HTML generated is different (I'm
sure ExceptionsTest::test_utf8_html_error_template_pygments failure is the
same).
Failing one:
<span class="cp">${</span><span class="sa">u</span><span class="s1">'
Expected one:
<span class="cp">${</span><span class="s1">u'
And the failing one contains additional lines in the CSS:
.syntax-highlighted .sa { color: #BA2121 } /* Literal.String.Affix */
.syntax-highlighted .dl { color: #BA2121 } /* Literal.String.Delimiter */
.syntax-highlighted .fm { color: #0000FF } /* Name.Function.Magic */
.syntax-highlighted .vm { color: #19177C } /* Name.Variable.Magic */
Looks like the problem is caused by something in Pygments but apart from
making sure the version used is the same I don't know how to proceed.
--
WBR, wRAR
signature.asc
Description: PGP signature
