July 2014 - MPlayer-dev-eng - lists.mplayerhq.hu

Homepage skin download issue
by Ingo Brückl 09 Sep '15

09 Sep '15

Yesterday, I updated hwswskin to 1.3, but it still isn't downloadable. Is the generation of the skin download tarballs working? Ingo

3 8

[PATCH] mencoder slang vobsub fix
by Edward Batutis 09 Sep '14

09 Sep '14

Hi, This is my very first patch for mplayer/mencoder. This patch allows mencoder -slang <lang> <some video file> -vobsub <some sub/idx path> ... to select a language from a SUB/IDX combination. For example: mencoder -slang en /movie/my.avi -vobsub /movie ... where /movie contains a my.sub and my.idx. This will select language 'en' from the idx/sub. Without this patch you always get subtitle zero, whatever it may be. Why this wasn't fixed long ago - who knows. Mplayer does the right thing. Not mencoder. =Ed

2 4

[PATCH] Fix key_down_event SEGFALT
by spunit262 . 04 Sep '14

04 Sep '14

I discovered by messing around that using key_down_event with a large random value (ex SPACE | MP_NO_REPEAT_KEY) can cause a segfalt. I found the problem to be that isprint is only defined for values in the range of an unsigned char and the value EOF. The patch adds a simple check before isprint in mp_input_get_key_name to make sure the value is in isprint's range.

2 1

[PATCH] spudec: fix heap overflow in pal2gray_alpha()
by Matthijs van Otterdijk 04 Sep '14

04 Sep '14

sub/spudec.c:spudec_packet_fill() optionally draws rectangles with an x and y offset, which is used by sub/av_sub.c:avsub_to_spudec() in case of multiple rects. The way this is done now causes a heap overflow in spudec.c:pal2gray_alpha(). spudec_packet_fill() offsets img and aimg by x before calling pal2gray_alpha(). pal2gray_alpha() writes dst_stride pixels for each line in the rect. In case the bottom rectangle (and therefore the rectangle situated at the end of the packet buffer) has an x offset, this will cause x 0s to be written past the end of the packet buffer. The attached patch fixes this by making pal2gray_alpha() handle the x offset, rather than spudec_packet_fill(). - Matthijs

2 1

Danish translation ought to be renamed
by Ingo Brückl 14 Aug '14

14 Aug '14

Does anyone mind me renaming the Danish translation from help_mp-dk.h to help_mp-da.h and protocolling this in the Changelog? (AFAICS, there are no dependencies to be considered.) There is no ISO 639-1 language code "dk", Danish is "da". Ingo

2 2

[patch] gui ignores mplayer volume option
by Stephen Sheldon 30 Jul '14

30 Jul '14

Mplayer has a volume option to set the initial audio volume. It ends up in a variable called start_volume. The GUI does not use it. It is nice to be able to start a video without the sound being embarrassingly loud. I tested it with pulseaudio. This patch does not work with the MS Windows GUI. I tried to get it to work by initializing guiInfo.Volume, it I could not do it. It works for me with the GTK GUI. The volume option does work in the MS Windows non-gui. The patch is attached.

2 10

mplayer embedded in aspx webpage
by Lieven Samijn 21 Jul '14

21 Jul '14

Dear, I'm looking for a longtime, a universal way to display a mjpeg stream on a aspx webpage (for a security camera). I want to do this without activeX or plugin because the site is running as GUI on an embedded system that need to be available for most browsers/OS, like IOS and Android (and windows/macOS of course) I'm able to play the streams for the command line with the mplayer.exe from your site but I cannot find a way to embed this into a html or aspx webpage. (server: win8embedded with IIS8 on it) Could somebody give me some hints or help to find (or a direction) a solution. Thank you very much. Best regards, Lieven

2 1

[ticket2196] mplayer thinks the streams end at the wrong point
by Alexander Strasser 20 Jul '14

20 Jul '14

Hi all, the ticket mentioned in the subject https://trac.mplayerhq.hu/ticket/2196 has a sample where mplayer thinks the streams end way before they actually do. So the audio stops after about 2s and the video at 6.4s. Other players like e.g. ffplay work correctly on that sample. I have made a stupid modification which is attached to that ticket, which makes the provided sample playable for me. It is a hack though and I do not understand that code well enough to take it further. Maybe someone can come up with a proper fix or point me in the right direction. Alexander

2 1

Patch to handle HTTP redirects which contain anchor data (anything following '#')
by badmailserver＠hushmail.com 18 Jul '14

18 Jul '14

The following patch makes redirected URLs (Location header) such as http://example.com/video.mp4?auth=secretauthentication#autoscroll_to_bottom work where it previously wouldn't because "#autoscroll_to_bottom" would be sent as part of the 'auth' GET variable, making authentication fail. --- MPlayer-1.1.1.orig/stream/http.c 2011-12-11 16:41:37.000000000 +0100 +++ MPlayer-1.1.1/stream/http.c 2014-07-18 15:14:44.978000575 +0200 @@ -824,6 +824,9 @@ // TODO: RFC 2616, recommand to detect infinite redirection loops next_url = http_get_field( http_hdr, "Location" ); if( next_url!=NULL ) { + // Avoid sending anchor data to the server + char* hash=strchr(next_url, '#'); + if(hash){hash[0]=0;} int is_ultravox = strcasecmp(stream->streaming_ctrl->url->protocol, "unsv") == 0; stream->streaming_ctrl->url = url_redirect( &url, next_url ); if (!strcasecmp(url->protocol, "mms")) {

1 0

Some thoughts on mode switching
by Jan Andres 06 Jul '14

06 Jul '14

Hi, I don't know about other TVs but mine offers a range of modes at its native resolution with different refresh rates matching those typically found in video content, such as 29.97 Hz, 25 Hz, 23.97 Hz etc. I observe that visual quality improves a lot in some scenes (e.g. when the camera is panning slowly) if I set the refresh rate to match the movie's frame rate prior to playback. Again this may not be the case on other TVs but I suppose it wouldn't hurt, so I'm wondering if one should add code to mplayer to match the refresh rate to the movie frame rate automatically. Ideally this would be an option separate from "-vm", as AFAIK the scalers in TVs differ widely in quality and most users would probably be best off running the TV at its native resolution with scaling performed by mplayer, but matching the refresh rate to the movie's may still be beneficial. What are your thoughts on this? Secondly, I'd like to inquire why the X11-based output drivers contain code to grab the pointer to the output window in case "-vm" is given (and only then). At least for my use-case, this is counterproductive: I have my TV connected to the computer as a secondary display and usually have mplayer play video there while continuing to work normally (even if distracted:) on the primary display. I cannot do this when the pointer gets grabbed by the mplayer output window on the secondary display. So from my perspective, it'd be better if this behavior were at least made optional. Regards, Jan -- Jan Andres <jandres(a)gmx.net>

2 1