[Python-apps-team] Bug#491109: Bug#491109: mercurial: Unreliable test for "wish" in postinst.

[Vincent Danjean]

tag 491109 +confirmed +help
thanks

Daniel Kobras wrote:
> Hi!
> 
> When mercurial's postinst checks whether to enable extension hgk, it
> runs
> 
> 	which wish > /dev/null || enable=false
> 
> but wish is handled via alternatives, and might only become available
> later on if mercurial and tk8.4 are upgraded in one batch. Lacking
> something like a "Pre-Suggests" to avoid this situation, maybe the test
> could be expanded to eg.
> 
> 	which wish > /dev/null || test -d /usr/share/doc/tk8.4 || enable=false
> 
> or even a version that's less kludgy.

The problem is that lots of tk versions are available (tk8.3 to tk8.6 (the latter
in experimental only))
Perhaps, a good thing to do would be to always enable the hgk extension and make
the check (with a error message telling to install one tkX.Y package) in the
code.
   Any idea on this part is welcome but I would like to avoid to hardcode in the
package the different tk version supported (it is always a mess when backport comes).

   Regards,
     Vincent

-- 
Vincent Danjean       GPG key ID 0x9D025E87         vdanjean at debian.org
GPG key fingerprint: FC95 08A6 854D DB48 4B9A  8A94 0BF7 7867 9D02 5E87
Unofficial pacakges: http://www-id.imag.fr/~danjean/deb.html#package
APT repo:  deb http://perso.debian.org/~vdanjean/debian unstable main