Re: Semantic change for dpkg triggers?
On Mon, 30 May 2011, Joey Hess wrote:
> Raphael Hertzog wrote:
> > My question is thus: are there triggers currently in use where this
> > relaxed behaviour would be wrong? Or more simply are there packages which
> > are really not working before the processing of their awaited triggers?
>
> python-support seems to need that; python does not see files in
> /usr/share/pyshared/, so until update-python-modules has run, a python
> module cannot be used.
That's true but that's precisely why python-support is not using
triggers for this part of the task, they are processed too late
and keep stuff broken for far too long.
The python-support trigger is used for the non-important part (I think
it's the same but for non-standard python versions), and it's actually
activated by /usr/sbin/update-python-modules with the --no-await
option.
> ghc libraries could need it, if any package depends on such a library
> being available to be compiled against at installation time. I don't
> know if any packages use ghc libraries like this currently, but I have
> considered writing one that does, if I ever found the time to rewrite
> ikiwiki in haskell.
I don't know anything about Haskell. What does the trigger do? Is it some
sort of non-optional byte-compilation?
Cheers,
--
Raphaël Hertzog ◈ Debian Developer
Follow my Debian News ▶ http://RaphaelHertzog.com (English)
▶ http://RaphaelHertzog.fr (Français)
Reply to: