[Python-modules-team] Sponsor needed for python-itsdangerous

Jakub Wilk jwilk at debian.org
Sat Apr 6 18:30:12 UTC 2013

* Daniel Kahn Gillmor <dkg at fifthhorseman.net>, 2013-04-05, 11:20:
>>Lintian has already this:
>>Implementing the reverse, i.e. python-helper-but-no-python-depends, 
>>should be very straight-forward and give good results.
>how should the reverse test deal with source packages that use dh 
>--with=python2 that provide multiple binaries?  should all binary 
>packages include ${python:Depends} in that case, or will we be 
>introducing a false-positive somewhere (e.g. on documentation packages, 
>or on packages that build both a C library and its associated python 

Insisting that all binary packages should have ${python:Depends} will 
cause tons of false positives; let's not go that way.

I think Lintian should emit the tag only if none of the binary packages 
have ${python:Depends}. This is not bulletproof, but should catch most 

Jakub Wilk

More information about the Python-modules-team mailing list