Re: RFS: python-meinheld
* Mikhail Lukyanchenko <ml@uptimebox.ru>, 2011-11-16, 17:36:
http://mentors.debian.net/package/python-meinheld
Alternatively, one can download the package with dget using this command:
dget -x http://mentors.debian.net/debian/pool/main/p/python-meinheld/python-meinheld_0.4.13-1.dsc
I would be glad if someone uploaded this package for me.
(I don't intend to sponsor this package.)
Is there any reason for using a different name for source package than
the name upstream uses?
What is meinheld/server/atomic.h for? It appears to be unused, which is
good, since it wouldn't compile on anything but x86.
meinheld/server/picoev* seems to be an embedded copy of a 3rd party
library. Please package it separately. Also, the files have a different
copyright holder and a difference license[*] than other file, which is
not documented in debian/copyright.
meinheld/server/greenlet.h is taken from the greenlet package, so it has
different license and copyright holders than the rest of files. Again,
this is not documented in debian/copyright.
Lintian emits these tags:
I: python-meinheld source: missing-debian-source-format
P: python-meinheld source: direct-changes-in-diff-but-no-patch-system meinheld.egg-info/SOURCES.txt
P: python-meinheld: no-upstream-changelog
When building, I get this warning:
meinheld/server/server.c: In function 'meinheld_set_process_name':
meinheld/server/server.c:1426:5: warning: implicit declaration of function 'Py_GetArgcArgv' [-Wimplicit-function-declaration]
and a few others about defined but unused variables.
Last but not least, the package doesn't work for me (unstable, i386) at
all. I tried the example from the homepage, and it segfaulted on first
request. Unfortunately, the traceback is not very helpful:
#0 __memcpy_ia32 () at ../sysdeps/i386/i686/multiarch/../memcpy.S:75
#1 0xbfffd240 in ?? ()
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
[*] Apparently, the authors didn't read their license very carefully...
--
Jakub Wilk
Reply to: