On Thu, 30 Apr 2020 02:03:20 +0200 Matthias Andree matthias.and...@gmx.de said
Am 28.04.20 um 22:02 schrieb Chris:
> In sentiment I am inline with your thoughts as well.
> Would it be a worthy project to create a mailman(2)-lts port?
> I'd be fully up for helping, and or creating it myself.
> There's a port that's a shim for py2.x-->py3.x called 2to3, or something
> like that. It also wouldn't be that difficult to simply modify mailman(2)
> to adopt the py3.x language changes.
Given that Mailman is mainly a text processing machine with various
heads (mail, web and CGI, command line) interfaces, and one of the ideas
driving the incompatible Python 3 was to clean up the delineation of the
strings/bytes/unicode types from one another and see to encoding. we'd
be in for lots of - ironically speaking - "fun" - meaning code audits,
revisions, possibly explicit code to write for the front lines to
properly decode external input and encode internal output.
The string stuff is the challenge (in python), and having just taken a
closer look, 3.x makes some changes in this area as well. Which only makes
the same challenges _different_. :(
I haven't looked into too much detail, but attaining 100% conversion and
test coverage would be a challenge and possibly a major undertaking.
I'd assume had it been as simple as 2to3 or py-futurize or adopting
py-six, someone might have done that already.
After looking closer, I'm inclined to say the (work) load appears close
to the same. Only the tasks have changed.
I really like Perl a lot more for all this string/byte handling stuff.
Maybe Majordomo? B-}
--Chris
_______________________________________________
freebsd-ports@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
