[MPlayer-cvslog] CVS: main configure, 1.1044, 1.1045 Makefile, 1.329, 1.330

[Alexander Strasser]

The Wanderer wrote:
> Rich Felker wrote:
> >On Sun, Aug 21, 2005 at 10:26:08AM +0200, Michael Niedermayer wrote:
> >>On Sun, Aug 21, 2005 at 06:25:58AM +0200, Diego Biurrun wrote:
> >>
> >>>On Sat, Aug 20, 2005 at 09:30:04PM -0400, Rich Felker wrote:
> 
> >>>Let's say that it was inspired by that commit, but I'm talking in
> >>>general, there are many more examples.  Yes, it will be obvious
> >>>to those in the know, but if you look at the output of 'cvs log'
> >>>then 10l is not descriptive unless you see the diff as well,
> >>>while something like "int/int does not produce a float value"
> >>>is...
> >>
> >>this is still not good, the commit log should at least contain
> 
> >>* low level descrioption / diff summary like (int/int does not
> >>produce...)
> >
> >IMO when the diff is short someone can be expected to read it..
> 
> But the diff is not always available ready to hand - it's there when
> reading the CVS-log mailing list, yes, but not when reading the output
> of 'cvs log'. For circumstances like that latter, having this kind of
> explanation can be helpful.

  While i see what you mean, i also know it is not true. If you have access
to cvs log, you usually have access to the repository. That in turn means
just take the to revisions from the log and let cvs produce a diff for
it (cvs diff -r<rev1> -r<rev2> <file>).

  But anyway, cvs log output alone is often suboptimal, as it is only
per file. So you don't necessarily can see the big picture.

  Alex (beastd)