testing python-markdown

[Yuri T.]

Yes, with -tidy option it does seem to normalize both files to a point
where it makes sense to diff them.   I had to make a lot of small
changes but ended up getting clean diff on most tests eventually.   
The tests that ended up failing seems to do so for three reasons:

1. Lack of support for nested braces ([[braces]](url)) - I can't think
of an easy way of adding this for the lack of python equivalent of
perl's recursive regexps.  I am not sure it's worth it.

2. <p> tags in "loose" lists - I missed this rule earlier and now
figured out that I my current design makes it hard to implement it. 
(When I am processing a list item I don't know what kind of a list
it's a part of.)  This does seem like an important feature, so I will
think about it more when I am in a procrastination mode.

3. Markdown.pl converts "+\tText" to "+   Text", i.e. it uses THREE
tabs.   What is the actual rule there?  I can see that Markdown.pl
handles tabs via $text =~ s{(.*?)\t}{$1.(' ' x ($g_tab_width -
length($1) % $g_tab_width))}ge; but this looks like line noise to
me...  Can someone provide pseudo-code for this?

  - yuri

On 8/12/05, Aaron Swartz <aaronsw at gmail.com> wrote:

> > > Python-markdown produces uses minidom to generate HTML, so it's

> > > hard to get it to agree with the perl implementation on a

> > > character-by-character basis and hence it would help to have a

> > > framework that can ignore meaninless whitespace.

> >

> > Then you should probably try to `diff -w`

> 

> On the other hand, MarkdownTest contains tidy support for exactly this purpose.

> _______________________________________________

> Markdown-Discuss mailing list

> Markdown-Discuss at six.pairlist.net

> http://six.pairlist.net/mailman/listinfo/markdown-discuss

>