On Wed, Apr 24, 2013 at 10:40 AM, Edward K. Ream <[email protected]>wrote:
> Two major discoveries:
>
Alas, neither python-modernize, nor six, nor the "unicode_escape" encoding
is a magic bullet.
1. six.u can fail for unicode input. Sheesh. This looks safer::
def u(s):
if sys.version_info < (3,):
return s if isinstance(s,unicode) else unicode(s,"unicode_escape")
else:
return s if isinstance(s,str) else str(s,"unicode_escape")
2. python-modernize is pretty disappointing. It doesn't automatically
convert imports! And the print fixer inserts a 'from __future__ import
print_function' for every print statement contains one of the new keywords.
Furthermore, python-modernize always wraps u literals in calls to six.u.
(There are actually other unicode literal options, but they are useless.)
Otoh, it's easy to convert calls to six.u to just u and convert 'import
six' to 'from docutils.utils.u import u'.
The bottom line is that unicode issues are still causing subtle problems.
The strategy remains:
- start with the original docutils sources,
- make *small* changes to the code,
- run all of docutils' unit tests
- figure out what went wrong :-)
The goal: create a relatively simple recipe for converting docutils to a
common code base *automatically*.
Not there yet. I had hoped to avoid writing my own fixers, but it looks
like that is not really an option.
Edward
--
You received this message because you are subscribed to the Google Groups
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/leo-editor?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
