[Python-markdown-discuss] Ticket 80

[Waylan L. <wa...@gm...>]

dtyschenko,

Thanks for you report on  Ticket 80[1] for Python-Markdown. I'm
responding to your report via the list [2] as I'm not sure how or if
we should fix the reported problem.

At first I was a little confused by your report as we don't do
anything with comments - if they exist in the source they are pulled
out as rawhtml and never see ElementTree.  Then I realized you must be
using a third party extension which create html comments and inserts
them into the tree. That is certainly something we want to support and
allow you to do.

However, it appears you are also using the html4 serializer - which we
borrowed (with permission) from the ElementTree 1.3 Preview (still in
alpha since 2007 AFAICT). Any changes to that serializer would ideally
be changes made upstream as well.

In fact, your proposed fix, to also import parts of ElementTree when
using cElementTree feels wrong and defeats the purpose of cElementTree
in the first place. In fact, I see in the notes to ElementTree 1.3
there is a note [3] that it only works with  cElementTree 1.0.6 or
later. What version are you using?

I'm going to say that that is a limitation of the html4 serializer. If
you are using cElementTree, you must be using version 1.0.6 or later.
In other words, this is a bug in cElmentTree, not a bug in Markdown.
Even if you are using a version greater than 1.0.6, I'd say this is
still a bug in cElementTree.

Of course, if anyone has any better suggestions, I'm open to entertain
them - and we accept patches.

Oh, and regarding your final comment regarding `markdown.isBlockLevel`
choking in Comment nodes, I'll patch that. We definitely should
support using Comment nodes there.

[1]: http://www.freewisdom.org/projects/python-markdown/Tickets/000080
[2]: https://lists.sourceforge.net/mailman/listinfo/python-markdown-discuss
[3]: http://effbot.org/zone/elementtree-13-intro.htm#performance

Waylan Limberg