* Peter T. Breuer <p...@inv.it.uc3m.es>, 2009-07-08, 14:32:
>Package: python-urlgrabber
>Version: 3.1.0-4
>Severity: normal
>
>Probably something doesn't compile under python 2.3 (which is safely
>installed on system here) or vice versa.
Python 2.3 was removed[1] from Debian and it obviously not supported.

Is it too much trouble to supply the missing support layer for 2.3?

I don't know about february 21 of 2008 (when this bug was filed), but
it seems to me you are playing a little fast and loose with your
tenses here :).  Waiting a year and a half and then announcing that the
support has disappeared isn't really "fixing bugs".
Sorry you had to wait so long. (Actually, it's not my responsibility to triage bugs; I just spotted this one by chance.)

And you are right, no bug was fixed, because there is no bug here.

Your reference

[1] http://bugs.debian.org/401407

is to debian UNSTABLE, when 2.3 was retired from THAT set of experimental
packages (in dec 2006).

Okay, here is the full timeline:
- Jan 2007: python2.3 was removed [1] from unstable and testing.
- Apr 2007: etch was released [2] (without python2.3).
- Feb 2008: this report was filed.
- Feb 2009: lenny was released [3] (without python2.3).

[2] http://lists.debian.org/debian-announce/2007/msg00002.html
[3] http://lists.debian.org/debian-announce/2009/msg00002.html

I maintain servers which are in many cases 10 years old, and going
strong.  I don't know which machine this was from - probably my laptop,
serving as a test base - but sure, I can compile python 2.4 for them
(and probably have) but why should I when 2.3 works fine?  If one only
have to provide a few support functions missing from 2.3, then shucks,
that's just par for the course in maintaining a piece of application
level code!
Actually, the culprit are not missing functions, but missing language constructs.
There are codes I wrote twenty years ago for which I am
still providing support and compatibility layers for many different
kinds of support architectures ..  one writes a line or two in
configure.in and generates the autoconf output and provides a function
of ones own. That's maintaining!
It's nice of you that you support legacy systems. However, we (Debian) have enough work with supporting modern ones.

So please give some thought to fixing this - it ought to be a no
brainer for you.
I insist that there nothing to fix here. It you think it's not the case, feel free to reopen the bug.

--
Jakub Wilk



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to