Hello all,
I think that I broke the installation abilities of the 0.49 release,
so I'm releasing a new version with that problem fixed. I took the
opportunity to integrate other patches laying around. There is still
many others to fix, and even some patches to integrate. I will try to
work on it soon. Any help would be welcome but I fully understand the
difficulty of entering this code base ;)
Speaking of which. po4a is a mature software, but I have the feeling
that perl is a dying ecosystem. I am playing with the idea of
reimplementing po4a in another language, in a hope to increase further
the contributer circle. A full rewrite is a lot of work given po4a's
size (about 3500 lines of perl), but the soft is well tested, which
will constitute a real asset.
Scala combinators would certainly be a nice technical choice to write
all the parsers that we have, but I'm not sure that it would be a
cleaver move to increase the developper community. Any scala fan in
the room (beyond me, I mean)?
Instead, I think that python is the scripting language of the future,
even if I'm not really fluent in that language (yet). It seems to me
that writing regulat expressions (which constitute 80% of the po4a
complexity) in python is not really pleasant. A very quick googling
tells me that the parsimonious package may be interesting here. But
like I said, I'm not fluent in python, and any insight is welcome.
That being said, my time on po4a is really scarce and I'm not fluent
in python, so I will not be able to do it alone anytime soon. For now,
I'd like to discuss the opportunity of a full rewrite, and the
available technologies that we could leverage. At some point, I hope
to find the time (or to receive the help) needed to develop some proof
of concept implementations before taking a strong decision.
Anyway. For now, the 0.50 release can be downloaded from
https://alioth.debian.org/frs/?group_id=30267 and some other perl
version will follow (I hope) to further integrate the patchs that you
guys provide and fix some bugs when I manage to.
Thanks for your attention,
Mt.
--
Learning and doing is the true spirit of free software; learning without
doing gets you academic sterility, and doing without learning is all too
often the way things are done in proprietary software. -- Raph Levien