within few days, i'll write a REST service to CRUD the sympa list defs.

I will not use the parser from the sympa code because:

• i don't want sympa to be a a software dependency: there is no need
• i want to be able to extend the format without editing the sympa code

i'll probably push this parser on CPAN but for now, TIMTOWTDI and i consider 3 of those:

• wrap roughly copy/paste from the sympa code
• use pegex
• write my incremental m//cxmsg parser (i still don't understand if it's bottom/up or top/down :))

all with pros and cons.

copy/paste

would be harder to maintain (the current code is a cargo cult parsing in a while loop) but every improvement can be contributed back to sympa.

Pegex

pegex grammars are very simple to maintain and read which could be usable in other languages (i haven't tested yet but javascript could be an example). It would add a dependency to Pegex but i really don't think it's a cons argument in 2014.

incremental m//cxmsg parser

as i'm used to it, i think i can provide simple, extensible and readble code following this way but is it worth to do it just to avoid the Pegex dependency? does Pegex add an runtime overload ?