[MPlayer-dev-eng] [PATCH] verbosity reduction in streaming, input, CPU detection

[Ivan Kalvachev]

Standard declaimer "I hate redusing verbosity"

Useful comments bellow:

2006/9/29, Diego Biurrun <diego at biurrun.de>:
> > > > > --- stream/tcp.c        (revision 19966)
> > > > > +++ stream/tcp.c        (working copy)
> > > > > @@ -153,7 +153,7 @@
> > > > >  #else
> > > > >         inet_ntop(af, our_s_addr, buf, 255);
> > > > >  #endif
> > > > > -       if(verb)
> > > > > mp_msg(MSGT_NETWORK,MSGL_STATUS,MSGTR_MPDEMUX_NW_ConnectingToServer,
> > > > > host, buf , port );
> > > > > +       if(verb) mp_msg(MSGT_NETWORK,MSGL_V,"Connecting to server
> > > > > %s[%s]: %d...\n", host, buf , port );
> > > >
> > > > I think also this should stay there
> > >
> > > But it's redundant:

If you don't want redundancy, remove the server name from the
"connecting" message.

The DNS query is relatively slow process and sometime could take extra
time, meanwhile mplayer will hang doing nothing and nobody would know
why. (This is especally true if it tries IPv6 query first and there is
no ipv6 servers).

These messages are also presented when seeking.


> > > MPlayer dev-SVN-r19982-4.1.2 (C) 2000-2006 MPlayer Team
> > > Connecting to server www.nasa.gov[212.201.100.134]: 80...
> > >
> > > Playing rtsp://a661.l1856741582.c18567.g.lr.akamaistream.net/live/D/661/18567/v0001/reflector:41582.
> > > Connecting to server a661.l1856741582.c18567.g.lr.akamaistream.net[207.40.149.87]: 554...
> > >
> > > The "Playing ..." line already conveys the interesting information, no
> > > need to repeat it 2 times I'd say.
> >
> > It can be useful when url changes, like with playlists or redirections
> >
> > r at tower2:~$ mplayer -playlist http://www.nasa.gov/ram/35037main_portal.ram
> > MPlayer dev-SVN-r19965-4.1.2 (C) 2000-2006 MPlayer Team
> > CPU: Intel(R) Pentium(R) 4 CPU 2.40GHz (Family: 15, Model: 2, Stepping: 4)
> > CPUflags:  MMX: 1 MMX2: 1 3DNow: 0 3DNow2: 0 SSE: 1 SSE2: 1
> > Compiled for x86 CPU with extensions: MMX MMX2 SSE SSE2
> > STREAM_HTTP(1), URL: http://www.nasa.gov/ram/35037main_portal.ram
> >
> > This will be removed ^^^

You'd better don't. Streams are opened also when reading subtitles,
Hiding it would make user wonder why it takes so long for a movie to start.

> >
> > Resolving www.nasa.gov for AF_INET...
> >
> > This will be removed ^^^

Already commented.


> > Connecting to server www.nasa.gov[213.254.17.128]: 80...
> >
> > This is the only line showing the connection and the port used ^^^
> >
> > Playing rtsp://a661.l1856741582.c18567.g.lr.akamaistream.net/live/D/661/18567/v0001/reflector:41582.
> > STREAM_RTSP, URL: rtsp://a661.l1856741582.c18567.g.lr.akamaistream.net/live/D/661/18567/v0001/reflector:41582
> >
> > This will be removed ^^^

I wonder if we still support playlist, so the given address and video
stream would differ?

> > Resolving a661.l1856741582.c18567.g.lr.akamaistream.net for AF_INET...
> >
> > This will be removed ^^^
> >
> > Connecting to server a661.l1856741582.c18567.g.lr.akamaistream.net[195.22.198.7]: 554...
> > Cache size set to 320 KBytes
> > ...
>
> You have a point here as well, but the first "Connecting to server" line
> is redundant, it just echoes the command line, HTTP is port 80 ..
>
> The second one can be interesting for playlists when the URL changes,
> but only then.
>
> > > > > --- cpudetect.c (revision 19966)
> > > > > +++ cpudetect.c (working copy)
> > > >
> > > > Why?
> > >
> > > Because it's not interesting information that might be useful outside of
> > > debugging.  For me it prints
> > >
> > > AltiVec found
> > > CPU: PowerPC
> > >
> > > Well, I knew what my CPU was already ..
> >
> > <rant>
> > Tell it to 'fat binary' lovers...
> > </rant>
>
> :)
>
> > Btw it prints a lot more for Intel cpus
>
> Yes, I'm aware of that, on my x86 box:
>
> CPU: AMD-K6(tm)-III Processor (Family: 5, Model: 13, Stepping: 0)
> CPUflags:  MMX: 1 MMX2: 0 3DNow: 1 3DNow2: 1 SSE: 0 SSE2: 0
> Compiled for x86 CPU with extensions: MMX 3DNow 3DNowEx
>
> But I still don't consider it useful outside of debugging.

It is good idea to say to the users that we are aware what their
processor features are and that we use them:)

I don't consider hiding network messages good idea, as generally
network is way slower than file operations.