Le samedi 10 octobre 2009 à 19:10 +0200, Daniel Leidert a écrit : > Am Samstag, den 10.10.2009, 18:10 +0200 schrieb Sylvestre Ledru: > > Le samedi 10 octobre 2009 à 17:32 +0200, Daniel Leidert a écrit : > > > Am Freitag, den 09.10.2009, 17:12 +0200 schrieb Sylvestre Ledru: > > > > > > > * Package name : gl2ps > > > > > > I'm not convinced, that this is a good idea. gl2ps is a very small > > > library and several packages (including scientific ones) are using > > > internal copies of gl2ps. > > gl2ps.c is 180 ko and the library 60 ko. > > That's not much. The library consists of 2 source files. and ? There are smaller libraries in Debian.
> > > I fear that packaging it will put pressure on all those maintainers, > > > whose packages use gl2ps source code to use the packaged version of > > > gl2ps. And I cannot see the benfit here. I expect, that this will be a > > > lot of work without much help from upstream authors of affected > > > software. > > I won't force them to do so. > > That's more a question of policy: 4.13. ATM this is a should. But to > make packages using internal copies of gl2ps use the pre-packaged > library, it would be necessary (well, AC_SEARCH_LIBS, ... can be used > withoutthe following) for gl2ps to support pkg-config or aclocal or > something similar. > > > > Which packages are expected to use this library atm (not counting those, > > > who ship a copy of gl2ps.c,h)? > > Scilab. > > How does it detect the library? Same old stuff, a m4 macro to check the existence of a the header file and the library. Anyway, I don't understand your point. Next version of Scilab will use gl2ps as an external dependency. I need this library to make Scilab works. The size of the library or if other packages are embedding the sources or not is irrelevant here. Sylvestre -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
