core: output --playing-msg message only after at least one frame is shown

This way it's possible to retrieve correct information about video, like
actual width/height, which in general are available only after at least
one frame has been sent to the video output, such as dwidth/dheight.

mpv_identify.sh becomes a bit slower, because we let it decode enough
audio and video to fill the audio buffers and to send one frame to the
video output. Also, --playing-msg isn't shown anymore with --frames=0
(could be fixed by special-casing it, should this break any use cases).

Note that in some corner cases, like when the demuxer for some reason
returns lots of audio packets but no video packets at the start, but
video actually starts later, the --playing-msg will still be output
before video starts.

DOCS/man/en/options.rst

@@ -1455,7 +1455,7 @@
     *WARNING*: works with the deprecated ``mp_http://`` protocol only.
 
 --playing-msg=<string>
-    Print out a string before starting playback. The string is expanded for
+    Print out a string after starting playback. The string is expanded for
     properties, e.g. ``--playing-msg=file: ${filename}`` will print the string
     ``file:`` followed by a space and the currently played filename.
 

TOOLS/mpv_identify.sh

@@ -100,7 +100,7 @@ for __midentify__key in $__midentify__allprops; do
     eval unset $__midentify__nextprefix$__midentify__key
 done
 
-__midentify__output=`mpv --playing-msg="$__midentify__propstr" --vo=null --ao=null --frames=0 "$@"`
+__midentify__output=`mpv --playing-msg="$__midentify__propstr" --vo=null --ao=null --frames=1 "$@"`
 __midentify__fileindex=0
 __midentify__prefix=
 while :; do

core/mp_core.h

@@ -255,6 +255,7 @@ typedef struct MPContext {
     int step_frames;
     // Counted down each frame, stop playback if 0 is reached. (-1 = disable)
     int max_frames;
+    bool playing_msg_shown;
 
     bool paused_for_cache;
 

core/mplayer.c

@@ -3397,6 +3397,13 @@ static void run_playloop(struct MPContext *mpctx)
             update_subtitles(mpctx, a_pos);
     }
 
+    if (opts->playing_msg && !mpctx->playing_msg_shown && new_frame_shown) {
+        mpctx->playing_msg_shown = true;
+        char *msg = mp_property_expand_string(mpctx, opts->playing_msg);
+        mp_msg(MSGT_CPLAYER, MSGL_INFO, "%s\n", msg);
+        talloc_free(msg);
+    }
+
     /* It's possible for the user to simultaneously switch both audio
      * and video streams to "disabled" at runtime. Handle this by waiting
      * rather than immediately stopping playback due to EOF.
@@ -4013,12 +4020,6 @@ goto_enable_cache: ;
         goto terminate_playback;
     }
 
-    if (opts->playing_msg) {
-        char *msg = mp_property_expand_string(mpctx, opts->playing_msg);
-        mp_msg(MSGT_CPLAYER, MSGL_INFO, "%s\n", msg);
-        talloc_free(msg);
-    }
-
     // Disable the term OSD in verbose mode
     if (verbose)
         opts->term_osd = 0;
@@ -4079,6 +4080,7 @@ goto_enable_cache: ;
     mpctx->step_frames = 0;
     mpctx->total_avsync_change = 0;
     mpctx->last_chapter_seek = -2;
+    mpctx->playing_msg_shown = false;
 
     // If there's a timeline force an absolute seek to initialize state
     double startpos = rel_time_to_abs(mpctx, opts->play_start, -1);