[MPlayer-dev-eng] [PATCH] printf --> mp_msg transition (mplayer.c)

[The Wanderer]

Diego Biurrun wrote:

> The Wanderer writes:

>> Well, I did make things harder by forgetting that I could just
>> specify the help/ directory, but the problem was actually that I
>> had tried to create two separate independent patches - one for the
>> two MSGTRs and one for the rest of the changes (since the former is
>> not the same issue as the latter). As a result of that I had made a
>> separate copy of mplayer.c in order to be able to make the diff
>> against my modified version of the file, and when I later said
>> "screw it" I forgot to copy the modified file back into the source
>> tree.
> 
> Probably the easiest way to accomplish that is to make a quick copy
> of your working directory, in case you don't have two already.
> Having two WDs is nice to have a clean "vanilla" working version and
> one you can mess with to your heart's content.

I haven't done that in the past, but now that you mention it I can see
the appeal.

> Editing patches in emacs should also work, it has a nice patch mode,
> I've often done this to split up patches.

I have yet to get to the point where I can actually use emacs; even the
comparatively minimal interface familiarity needed for getting into and
navigating its tutorial/help/internal-documentation mode is more
complication than I've been able to handle. I'd like to use it, but that
hump is one I have just not yet managed to get over.

>> (Speaking of not using CVS properly, I've just noticed a particular
>> type of problem for the second time. If I do 'diff -u
>> main/m_config.h main.working/m_config.h', it reports a difference;
>> if I do 'cvs diff -u m_config.h' from either source tree, it
>> reports no differences. The last time I saw this, in order to get
>> the local copies to sync up I had to remove the allegedly-changed
>> file and have CVS think it'd gotten lost entirely. Any idea what
>> might be going on?)
> 
> I'm in the dark here.  Maybe you have some sticky or something tags
> on parts of your WD, these are known to cause obscure problems now
> and then.  Try "cvs update -dPA" to reset them.

I always use that anyway - I have a nice little "update_cvs" file in the
root of every CVS source tree on my machine, with whatever flags I need
to get the latest version from that repository, to save myself from
having to remember what to use where (or for that matter from typing it
all out every time); '-dPA' is there in all of them. So it looks like
we're both quite confused.

>> (...and, on a complete tangent, why is help_mp.h in .cvsignore?? I
>> certainly don't remember adding it to my local copy...)
> 
> help_mp.h is generated during the configure run.

...right, I knew that, I just forgot to make the connection.

I'm currently going through codecs-cfg.c and removing printfs; I'm not
entirely certain that it really should be changed (because the things
printed seem designed to be written directly into a file, implying that
they don't get printed on the console anyway), but it wasn't mentioned
when I asked about which files shouldn't be transitioned, so because it
does already include mp_msg.h I'm going ahead with it anyway. I am,
however, uncertain as to what messages should and should not be moved
for translation - it would help (though not much) if I at least knew
what the code in that file actually is used for. Many of them seem to be
translation-appropriate, but... something seems different here.

-- 
       The Wanderer

Warning: Simply because I argue an issue does not mean I agree with any
side of it.

A government exists to serve its citizens, not to control them.