[Moin-devel] handling paragraphs as long lines

[Werner LEMBERG]

IMHO one of the weakest points in MoinMoin is how wiki.py parses the
input.  Using regular expressions simply doesn't work properly for
nested balanced expressions, and we have already far too many
exceptions which nested expressions work, and which not.

What do you think of using a completely different paradigm, namely to
do bison-like parsing of the input data to get the appropriate tokens?

To use a solution in pure Python the following looks very promising:

  http://lava.net/~newsham/pyggy/html/pyggy.html

BTW, is any other part of MoinMoin dependent on wiki.py?  I mean, can
I simply replace the HTML generating engine with a different one,
similar to the `colorizing' parsers for other languages like C or C++?
Where can I find the documentation of the affected interface?


    Werner