[MPlayer-dev-eng] Re: x11_common.c: it still doesn't work in fullscreen
diego at biurrun.de
Wed Mar 17 18:42:11 CET 2004
Greg Stark writes:
> Diego Biurrun <diego at biurrun.de> writes:
> > Just please don't "reverse" revisions that are not the last ones
> > again, it's a bad idea anyway as it creates a "hole" in the revision
> > history.
> I don't know how it works with mplayer, but on other projects I've worked on
> we never "removed" revisions even if it *was* the last one. Doing so removes
> history information from the repository that really ought to be preserved even
> if it doesn't cause problems, and it seems bound to cause problems.
> We always just did cvs "update -j newversion -j oldversion" then committed.
> This means all the right things happen to anybody who has a checked out copy
> of the old tree, namely they get told to update before committing. Also it
> means there's a log of the patch being applied and reversed so people later
> can see what happened.
I agree fully. Having the information about what was bad and backed
out in the CVS history can be extremely useful. IMHO the only
justified use of 'cvs admin -o' is to remove the latest revision of a
file if you committed some kind of embarassing goof and recommit an
improved version immediately thereafter.
But obviously 'cvs admin -o' is being abused and misused, so it might
be a good idea to completely refrain from using it.
More information about the MPlayer-dev-eng