bundled/markdown2/TODO.txt @ 431b3e74013f

- add "html-classes" extra to wiki
- bug: can't have '<\w+' in a code span or code block with safe_mode if there
  is a '>' somewhere later in the document. E.g. code.as.com-beta/CHANGES.md.
  It captures all of that. Right answer is to not count code spans or code
  blocks.
  - add an issue for this
  - test cases
  - idea: better sanitation re-write? lot of work
  - idea: Change all <,>,& emission from markdown processing to something
    like {LT}, {GT}, {AMP}, {OPENTAG:$tag[:$class]} (first checking for
    conflicts and escaping those out of the way). Then do sanitization at the
    end:
        escape: escape all <,>,& with entities
        remove: not supported
        whitelist: (new) build a reasonable default whitelist of patterns to
            keep. Takes "extras" argument (and hook for subclassing) to
            for custom whitelist. Google Code (was it?) had some list
            of reasonable whitelist stuff.
    Then unescape these special chars. The use of OPENTAG above would make
    "html-classes" extra trivial.

- fix the r135 xml option, add xml extra for it (see email)
- look at http://code.google.com/p/markdownsharp/
- add description of pyshell and demote-headers extras to wiki
- to bring up on markdown-discuss:
    - the trailing '#' escaping in DoHeaders (provide a patch for this)
    - the discussion of backticks and backslash-escapes in code spans:
        also bring in python-markdown-discuss on this
    - The link for backslash escapes doesn't mention '>', but I believe it
      should -- viz Markdown.pl's `%g_escape_table` which *does* include '>'.
      TODO: bring this up on markdown-discuss list.
- wiki: add an "Other Markdown implementations page"
    http://daringfireball.net/projects/markdown/
    http://www.michelf.com/projects/php-markdown/
    http://www.freewisdom.org/projects/python-markdown/Features
- test safe_mode on HTML in footnotes
- compare setup.py stuff from Yannick to what I have now. Also:
    http://gitorious.org/projects/git-python/repos/mainline/trees/master
    http://www.python.org/~jeremy/weblog/030924.html
- http://www.freewisdom.org/projects/python-markdown/Available_Extensions
- Extras.wiki desc of code-color option. Not sure I love the ":::name"
  markup for the lexer name.
- find more unicode edge cases (look for any usage of md5() and make that
  unicode)
- update MDTest 1.1? (see
  http://six.pairlist.net/pipermail/markdown-discuss/2007-September/000815.html)
  update MDTest tests from http://git.michelf.com/mdtest/
- I see ref to Markdown.pl 1.0.2
  (http://six.pairlist.net/pipermail/markdown-discuss/2007-August/000756.html)
  Update to that? Yes. Copy, at least, in showdown package.
- take a look at other examples/test-cases from
  http://adlcommunity.net/help.php?file=advanced_markdown.html
- googlecode site: Why another Python impl? Test info. Usage/Features page.
- get on http://en.wikipedia.org/wiki/Markdown
- ask about remaining two MarkdownTest test failures
- put in recipes site
- perhaps some extras from Maruku and PHP Markdown extra
  (http://maruku.rubyforge.org/maruku.html#extra)
    - tables (tho I don't really like the syntax, prefer google codes, see
      below)
    - markdown inside literal HTML (if 'markdown="1|true"' attr)
    - automatic toc generation (wanted that anyway, no a fan of maruku syntax
      for this)
    - weird markup in headers and links (does markdown2.py handle this?)
    - meta-data syntax? One example of this is ids for headers. How about
      automatically assigning header ids from the name (a la rest)?
    - at-the-top email-style headers?
    - maruku's footnote links are 'fn:1' and 'fnref:1' for a footnote id of
      'blah'. If this is the PHP Markdown Extras way, then should follow
      that.
- googlecode wiki markup ideas?
  (http://code.google.com/p/support/wiki/WikiSyntax)
    - ~~strikeout~~
    - ||tables||simple||syntax||
- <http://daringfireball.net/2004/12/markdown_licensing> at bottom has a wish
  list:
    - simple "cite" for blockquote. How about:
        [Zaphod Breeblebrox]
        > blah blah
        > blah
- do perf comparison with the other Markdown impls (if compare horribly then
  do something about it)
- submit a Markdown.py (or .pl?) fix based on revision 1895 (on tm svn)
- see about using html5lib (for speed and/or for better raw HTML handling)
- see about plugins (SmartyPants, others available)