[MPlayer-dev-eng] MPLAYER_VERBOSE environment variable?

[Chuck Blake]

>Feeling towards this patch seems to be positive in general.

Cool.  Just a couple of quick responses.

 - Moving all output earlier could be do-able, but this was simpler.

 - ANSI C89 guarantees environment variables and atoi() as well, and that
   they're in stdlib.h which mp_msg.c already #include's.  Both even work
   in MS-DOS, Windows, etc.  It should really be an extremely low risk for
   portability problems, though I obviously cannot test every platform by
   myself...I have some trouble with getenv() being hella slow on Windows
   relative to Unix, but that is only a relative thing meaning ~10s of micro
   seconds rather than sub microsecond, but we only call it once so that's
   not an issue.

>Nevertheless the environment variable needs to be documented in the man
>page for this to be committed...

  Agreed. [ Seemed premature if you guys were going to reject it. ]
  How about this?  Short and sweet...No need to burden the docs with long-
  winded rationales/justifications/etc.

  The other reason I held off originally is that I cannot translate it into
  all the various languages MPlayer provides with its docs.

diff -ruw mplayer-1.0.20060102.orig/DOCS/man/en/mplayer.1 mplayer-1.0.20060102/DOCS/man/en/mplayer.1
--- mplayer-1.0.20060102.orig/DOCS/man/en/mplayer.1     2006-01-01 16:53:32.000000000 -0500
+++ mplayer-1.0.20060102/DOCS/man/en/mplayer.1  2006-01-18 20:18:33.569674375 -0500
@@ -9141,6 +9141,11 @@
 .B MPLAYER_HOME
 Directory where MPlayer looks for user settings.
 .
+.B MPLAYER_VERBOSE
+The default verbosity level across all modules.
+If this variable is unset MPlayer/MEncoder uses the value 0.
+See also the -v and -msglevel options.
+.
 .SS libaf:
 .
 .TP