ffmpeg.git
17 years agobegan to simplify code - use modified API for stream reading
Fabrice Bellard [Mon, 20 May 2002 16:32:49 +0000 (16:32 +0000)]
began to simplify code - use modified API for stream reading

Originally committed as revision 548 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoconverted to new API
Fabrice Bellard [Mon, 20 May 2002 16:31:13 +0000 (16:31 +0000)]
converted to new API

Originally committed as revision 547 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agosuppressed mpeg demux mess - use now dynamic stream creation api
Fabrice Bellard [Mon, 20 May 2002 16:29:40 +0000 (16:29 +0000)]
suppressed mpeg demux mess - use now dynamic stream creation api

Originally committed as revision 546 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agosplit mux/demux related structures - added file probing support - improve media file...
Fabrice Bellard [Mon, 20 May 2002 16:28:47 +0000 (16:28 +0000)]
split mux/demux related structures - added file probing support - improve media file reading API

Originally committed as revision 545 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded first version of MPEG/DVB transport stream demux
Fabrice Bellard [Mon, 20 May 2002 16:27:23 +0000 (16:27 +0000)]
added first version of MPEG/DVB transport stream demux

Originally committed as revision 544 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoremoved unused code
Fabrice Bellard [Mon, 20 May 2002 16:25:09 +0000 (16:25 +0000)]
removed unused code

Originally committed as revision 543 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agompeg1/2 identifier - fixed frame rate for some bad mpeg1 streams
Fabrice Bellard [Mon, 20 May 2002 16:24:39 +0000 (16:24 +0000)]
mpeg1/2 identifier - fixed frame rate for some bad mpeg1 streams

Originally committed as revision 542 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agobetter av_freep()
Fabrice Bellard [Mon, 20 May 2002 16:23:27 +0000 (16:23 +0000)]
better av_freep()

Originally committed as revision 541 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded avcodec_set_bit_exact() for testing
Fabrice Bellard [Mon, 20 May 2002 16:22:51 +0000 (16:22 +0000)]
added avcodec_set_bit_exact() for testing

Originally committed as revision 540 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoupdate
Fabrice Bellard [Mon, 20 May 2002 16:22:05 +0000 (16:22 +0000)]
update

Originally committed as revision 539 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded regression test mention
Fabrice Bellard [Mon, 20 May 2002 16:21:46 +0000 (16:21 +0000)]
added regression test mention

Originally committed as revision 538 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded libav regression tests
Fabrice Bellard [Mon, 20 May 2002 16:20:05 +0000 (16:20 +0000)]
added libav regression tests

Originally committed as revision 537 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago64bit incompatibility bug, patch by Artur Grabowski <art@blahonga.org>
Artur Grabowski [Mon, 20 May 2002 04:00:25 +0000 (04:00 +0000)]
64bit incompatibility bug, patch by Artur Grabowski <art@blahonga.org>

Originally committed as revision 6141 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

17 years agobig cosmetics patch, cleanup of messages printed by mplayer and libs.
Arpi [Mon, 20 May 2002 03:25:26 +0000 (03:25 +0000)]
big cosmetics patch, cleanup of messages printed by mplayer and libs.
some printf->mp_msg conversion, and some debug messages moved from warn/info to v/dbg2
mplayer's output is now shorter, readable and consistent

Originally committed as revision 6139 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

17 years agoFix a segfault when handling errors or .asx or .ram files. Silly bug
Philip Gladstone [Mon, 20 May 2002 03:02:09 +0000 (03:02 +0000)]
Fix a segfault when handling errors or .asx or .ram files. Silly bug
on my part.

Originally committed as revision 536 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agofixing alignment for alignment picky cpus
Michael Niedermayer [Sun, 19 May 2002 01:49:53 +0000 (01:49 +0000)]
fixing alignment for alignment picky cpus

Originally committed as revision 535 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agooops - removed obsolete license draft
Fabrice Bellard [Sat, 18 May 2002 23:31:08 +0000 (23:31 +0000)]
oops - removed obsolete license draft

Originally committed as revision 534 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoupdate
Fabrice Bellard [Sat, 18 May 2002 23:20:40 +0000 (23:20 +0000)]
update

Originally committed as revision 533 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded first version of regression tests - simply type 'make test' to test every codec...
Fabrice Bellard [Sat, 18 May 2002 23:18:22 +0000 (23:18 +0000)]
added first version of regression tests - simply type 'make test' to test every codec and some mux/demuxes - added simple synthetic audio and video bitstreams generators so that no big streams needs to be used

Originally committed as revision 532 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agouse av memory functions
Fabrice Bellard [Sat, 18 May 2002 23:17:41 +0000 (23:17 +0000)]
use av memory functions

Originally committed as revision 531 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded bigendian support - added cross compilation support - added generation of objec...
Fabrice Bellard [Sat, 18 May 2002 23:13:26 +0000 (23:13 +0000)]
added bigendian support - added cross compilation support - added generation of objects in another directory for simultaneous compilations (needs testing) - simplified redundant code

Originally committed as revision 530 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded VPATH support - added regression targets
Fabrice Bellard [Sat, 18 May 2002 23:11:25 +0000 (23:11 +0000)]
added VPATH support - added regression targets

Originally committed as revision 529 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agouse av memory handling functions
Fabrice Bellard [Sat, 18 May 2002 23:11:09 +0000 (23:11 +0000)]
use av memory handling functions

Originally committed as revision 528 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agofixed pcm read
Fabrice Bellard [Sat, 18 May 2002 23:09:43 +0000 (23:09 +0000)]
fixed pcm read

Originally committed as revision 527 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded VPATH support
Fabrice Bellard [Sat, 18 May 2002 23:07:57 +0000 (23:07 +0000)]
added VPATH support

Originally committed as revision 526 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded dummy crc output format for easy codec testing
Fabrice Bellard [Sat, 18 May 2002 23:06:13 +0000 (23:06 +0000)]
added dummy crc output format for easy codec testing

Originally committed as revision 525 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoremoved unmaintained vc stuff
Fabrice Bellard [Sat, 18 May 2002 23:05:19 +0000 (23:05 +0000)]
removed unmaintained vc stuff

Originally committed as revision 524 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoremoved not maintained vc stuff
Fabrice Bellard [Sat, 18 May 2002 23:04:30 +0000 (23:04 +0000)]
removed not maintained vc stuff

Originally committed as revision 523 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoremoved useless header includes - use av memory functions
Fabrice Bellard [Sat, 18 May 2002 23:03:29 +0000 (23:03 +0000)]
removed useless header includes - use av memory functions

Originally committed as revision 522 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded VPATH support for multiple dir compilation
Fabrice Bellard [Sat, 18 May 2002 23:01:20 +0000 (23:01 +0000)]
added VPATH support for multiple dir compilation

Originally committed as revision 521 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoproper memory handling functions
Fabrice Bellard [Sat, 18 May 2002 22:59:50 +0000 (22:59 +0000)]
proper memory handling functions

Originally committed as revision 520 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoremoved debug function
Fabrice Bellard [Sat, 18 May 2002 22:59:12 +0000 (22:59 +0000)]
removed debug function

Originally committed as revision 519 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoput all integer init code to compute n^(4/3) - memory alloc and header fixes
Fabrice Bellard [Sat, 18 May 2002 22:58:08 +0000 (22:58 +0000)]
put all integer init code to compute n^(4/3) - memory alloc and header fixes

Originally committed as revision 518 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoimproved mpeg1/2 decoding speed
Fabrice Bellard [Sat, 18 May 2002 22:56:50 +0000 (22:56 +0000)]
improved mpeg1/2 decoding speed

Originally committed as revision 517 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoadded proper memory handling functions - fixed include paths
Fabrice Bellard [Sat, 18 May 2002 22:56:25 +0000 (22:56 +0000)]
added proper memory handling functions - fixed include paths

Originally committed as revision 516 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agotried to avoid gcc 2.95.2 bug by puting explicit register constraints - added comment...
Fabrice Bellard [Sat, 18 May 2002 22:49:11 +0000 (22:49 +0000)]
tried to avoid gcc 2.95.2 bug by puting explicit register constraints - added comment about rounding bug in some functions (need to correct or suppress them for regression tests)

Originally committed as revision 515 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* If a stream gets stuck in WAIT_FEED, then disconnecting the other end
Philip Gladstone [Sat, 18 May 2002 21:33:05 +0000 (21:33 +0000)]
* If a stream gets stuck in WAIT_FEED, then disconnecting the other end
  will clear out the stream. I think that this is really a linux bug in
  the handling of poll, but I did a workaround anyway.
* Improve the statistics output and prevent a buffer overrun when lots
  of clients are connected.
* Process ffm input when it is received and don't always be one ffm
  packet behind.
* Try to avoid going through the poll loop when not required.

Originally committed as revision 514 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoFix the WAIT_FEED problem. It turns out that when you open up an FFM
Philip Gladstone [Sat, 18 May 2002 21:30:16 +0000 (21:30 +0000)]
Fix the WAIT_FEED problem. It turns out that when you open up an FFM
file and seek to an FFM packet, then it is important that the packet
found has a frame header within it. If not, then terrible things happen.
This fixes the problem.

Originally committed as revision 513 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoupdated for my short term goals
Fabrice Bellard [Sat, 18 May 2002 10:28:55 +0000 (10:28 +0000)]
updated for my short term goals

Originally committed as revision 512 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoqmin==1 bugfix
Michael Niedermayer [Fri, 17 May 2002 15:31:33 +0000 (15:31 +0000)]
qmin==1 bugfix

Originally committed as revision 511 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoshared lib support (req by kabi) ...
Michael Niedermayer [Fri, 17 May 2002 13:01:01 +0000 (13:01 +0000)]
shared lib support (req by kabi) ...

Originally committed as revision 510 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Added a note that says that WMP cannot stream .mpg files.
Philip Gladstone [Fri, 17 May 2002 01:54:39 +0000 (01:54 +0000)]
* Added a note that says that WMP cannot stream .mpg files.

Originally committed as revision 509 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoAdded use of StartSendOnKey
Philip Gladstone [Fri, 17 May 2002 01:54:14 +0000 (01:54 +0000)]
Added use of StartSendOnKey

Originally committed as revision 508 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Change the default behaviour to start streaming as soon as possible (i.e.
Philip Gladstone [Fri, 17 May 2002 01:53:28 +0000 (01:53 +0000)]
* Change the default behaviour to start streaming as soon as possible (i.e.
  no waiting for key frames).
* Provide StartSendOnKey paramter for a stream to wait until we get key frames
  before sending.
* Add the codec names into the status page. May help debugging problems.

Originally committed as revision 507 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agohopefully faster mmx2&3dnow MC
Michael Niedermayer [Fri, 17 May 2002 01:04:14 +0000 (01:04 +0000)]
hopefully faster mmx2&3dnow MC

Originally committed as revision 506 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agonew hopefully faster MC
Michael Niedermayer [Thu, 16 May 2002 23:29:09 +0000 (23:29 +0000)]
new hopefully faster MC

Originally committed as revision 505 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoprint frame decoding time support (x86 only)
Michael Niedermayer [Thu, 16 May 2002 22:01:30 +0000 (22:01 +0000)]
print frame decoding time support (x86 only)

Originally committed as revision 504 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* fixed --enable-shared handling
Zdenek Kabelac [Thu, 16 May 2002 18:02:50 +0000 (18:02 +0000)]
* fixed --enable-shared handling

Originally committed as revision 503 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Updated with new stream configuration items
Philip Gladstone [Thu, 16 May 2002 02:03:34 +0000 (02:03 +0000)]
* Updated with new stream configuration items

Originally committed as revision 502 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add code to configure the following:
Philip Gladstone [Thu, 16 May 2002 02:03:07 +0000 (02:03 +0000)]
* Add code to configure the following:
  * prebuffering/preroll a live stream -- this improves startup time
  * videoqmin/videoqmax/videoqdiff -- codec parameters
  * maximum bandwidth for live streams
* Add support for .ram and .rpm extensions mapping onto .rm
* Make the status page show bandwidth. Also make the .asf and .rm
  links go to .asx and .ram files.
* Make a stream only start streaming when it gets a keyframe on each
  stream. This is arguable, and it maybe ought to be restricted to
  live streams. However, since I don't think that file streams work,
  this is a step in the right direction. It improves the startup delay.
* Log an error if we are unable to delete the temp feed file.

Originally committed as revision 501 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoPass over the bit_rate_tolerance field to ffmpeg
Philip Gladstone [Thu, 16 May 2002 01:57:55 +0000 (01:57 +0000)]
Pass over the bit_rate_tolerance field to ffmpeg

Originally committed as revision 500 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agofixing end overwrite bugs (some at least)
Michael Niedermayer [Wed, 15 May 2002 02:27:41 +0000 (02:27 +0000)]
fixing end overwrite bugs (some at least)
this needs testing, untested parts should be reversed before release
i tested 24->32, 15->16, bgr24->rgb, bgr32->rgb (retesting them isnt a bad idea either ...)

Originally committed as revision 6097 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

17 years ago- Bug fix for AVStream->r_frame_rate not being initialized for live video
Juanjo [Wed, 15 May 2002 02:02:24 +0000 (02:02 +0000)]
- Bug fix for AVStream->r_frame_rate not being initialized for live video
capture.

Originally committed as revision 499 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Now ffmpeg handles MPEG-2 streams with pull down, it gets the real frame
Juanjo [Tue, 14 May 2002 22:06:13 +0000 (22:06 +0000)]
- Now ffmpeg handles MPEG-2 streams with pull down, it gets the real frame
rate of the stream.

Originally committed as revision 498 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Added r_frame_rate in AVStream to store the real frame rate.
Juanjo [Tue, 14 May 2002 21:50:00 +0000 (21:50 +0000)]
- Added r_frame_rate in AVStream to store the real frame rate.
- This stuff could be done in AVCodecContext and the real frame rate
calculated in utils.c, but we don't changes in that by now ;)

Originally committed as revision 497 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- repeat_pict meaning changed, now it signals the extra delay for the
Juanjo [Tue, 14 May 2002 19:56:42 +0000 (19:56 +0000)]
- repeat_pict meaning changed, now it signals the extra delay for the
decoded frame.
- extra_delay = (repeat_pict / 2) * (1/fps)

Originally committed as revision 496 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoenable simple-idct by default (needed cuz otherwise files encoded with ffmpeg will...
Michael Niedermayer [Tue, 14 May 2002 15:42:54 +0000 (15:42 +0000)]
enable simple-idct by default (needed cuz otherwise files encoded with ffmpeg will show stripes with the reference IDCT during decoding)

Originally committed as revision 495 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* first shot for the new avcodec API
Zdenek Kabelac [Tue, 14 May 2002 14:17:11 +0000 (14:17 +0000)]
* first shot for the new avcodec API
  - comments, critics, improvements on the ffmpeg list are welcomed

Originally committed as revision 494 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Mea culpa: it seems that I broke encoding to 8-bit pcm files. This fixes it.
Philip Gladstone [Tue, 14 May 2002 02:36:23 +0000 (02:36 +0000)]
* Mea culpa: it seems that I broke encoding to 8-bit pcm files. This fixes it.

Originally committed as revision 493 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Video CD option for ffmpeg
Juanjo [Sun, 12 May 2002 21:39:44 +0000 (21:39 +0000)]
- Video CD option for ffmpeg

Originally committed as revision 492 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- VCD MPEG-1 compliant stream support (set AVF_FLAG_VCD)
Juanjo [Sun, 12 May 2002 21:38:54 +0000 (21:38 +0000)]
- VCD MPEG-1 compliant stream support (set AVF_FLAG_VCD)

Originally committed as revision 491 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agobuild num ++
Nick Kurshev [Sun, 12 May 2002 16:06:45 +0000 (16:06 +0000)]
build num ++

Originally committed as revision 490 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agobuild info for ABI
Nick Kurshev [Sun, 12 May 2002 15:53:25 +0000 (15:53 +0000)]
build info for ABI

Originally committed as revision 489 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoI was too lazy to be a hero ;)
Nick Kurshev [Sun, 12 May 2002 15:30:10 +0000 (15:30 +0000)]
I was too lazy to be a hero ;)

Originally committed as revision 488 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- ffmpeg relies on libavcodec to repeat frames from 3:2 sources.
Juanjo [Sun, 12 May 2002 02:56:50 +0000 (02:56 +0000)]
- ffmpeg relies on libavcodec to repeat frames from 3:2 sources.

Originally committed as revision 487 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- repeat_pict added to AVCodecContext to signal if the decoder must repeat
Juanjo [Sun, 12 May 2002 02:55:48 +0000 (02:55 +0000)]
- repeat_pict added to AVCodecContext to signal if the decoder must repeat
  frames from 3:2 pulldown sources.
- repeat_pict honoured by MPEG-1/2 decoder.

Originally committed as revision 486 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agovarious openbsd and general warning fixes - patch by Björn Sandell <biorn@dce.chalmer...
Björn Sandell [Sun, 12 May 2002 02:18:52 +0000 (02:18 +0000)]
various openbsd and general warning fixes - patch by Björn Sandell <biorn@dce.chalmers.se>

Originally committed as revision 6058 to svn://svn.mplayerhq.hu/mplayer/trunk/postproc

17 years ago- Fixes for MPEG-PS which have more than 1 system header, now we don't take
Juanjo [Sat, 11 May 2002 23:53:04 +0000 (23:53 +0000)]
- Fixes for MPEG-PS which have more than 1 system header, now we don't take
  in account the system header just walk through the file searching streams.
  MPEG-PS demuxer needs more work anyway.

Originally committed as revision 485 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- pict_type exported to AVCodecContext
Juanjo [Sat, 11 May 2002 23:42:16 +0000 (23:42 +0000)]
- pict_type exported to AVCodecContext
- Added real_pict_num to AVCodecContext, it represent the number of the
  previous encoded frame, usefull when coding with B frames.
- Warning fix in motion_est.c

Originally committed as revision 484 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Bug fix in map stream parser
Juanjo [Sat, 11 May 2002 23:27:14 +0000 (23:27 +0000)]
- Bug fix in map stream parser

Originally committed as revision 483 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Improve logging -- now actually gives you the number of bytes sent
Philip Gladstone [Fri, 10 May 2002 02:20:27 +0000 (02:20 +0000)]
* Improve logging -- now actually gives you the number of bytes sent
* Print out the logging information that comes from WMP (you'd be suprised what
  it sends!
* Fix a remotely exploitable buffer overflow (argh!)
* Add support for automatically serving up .asx files. It generates an automatic
  redirect to the associated .asf file (with the same parameters). I guess that
  someone who understands the realaudio equivalent could hack that it as well.

Originally committed as revision 482 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add implementation of strlcpy
Philip Gladstone [Fri, 10 May 2002 02:17:41 +0000 (02:17 +0000)]
* Add implementation of strlcpy
* Fix endless loop in find_info_tag if given specific arguments

Originally committed as revision 481 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add prototype for strlcpy
Philip Gladstone [Fri, 10 May 2002 02:16:29 +0000 (02:16 +0000)]
* Add prototype for strlcpy

Originally committed as revision 480 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add the 'flags' field to the list of codec fields passed from ffserver to
Philip Gladstone [Thu, 9 May 2002 12:37:44 +0000 (12:37 +0000)]
* Add the 'flags' field to the list of codec fields passed from ffserver to
  ffmpeg. This gives the VideoHighQuality patch a fighting chance.

Originally committed as revision 479 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- HQ option for ffserver. Thanks to D Richard Felker III.
Juanjo [Thu, 9 May 2002 07:43:07 +0000 (07:43 +0000)]
- HQ option for ffserver. Thanks to D Richard Felker III.

Originally committed as revision 478 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoversion info for so too
Nick Kurshev [Thu, 9 May 2002 06:23:37 +0000 (06:23 +0000)]
version info for so too

Originally committed as revision 477 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Warning fix.
Juanjo [Thu, 9 May 2002 05:42:27 +0000 (05:42 +0000)]
- Warning fix.

Originally committed as revision 476 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Updating docs....
Juanjo [Thu, 9 May 2002 04:37:38 +0000 (04:37 +0000)]
- Updating docs....

Originally committed as revision 475 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Changed default qmin = 3 as ffmpeg.c
Juanjo [Thu, 9 May 2002 02:33:27 +0000 (02:33 +0000)]
- Changed default qmin = 3 as ffmpeg.c
- Less bitrate tolerance since we are streaming.

Originally committed as revision 474 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago- Segfault fixed when mpeg audio decoder returns a negative data_size.
Juanjo [Thu, 9 May 2002 02:32:22 +0000 (02:32 +0000)]
- Segfault fixed when mpeg audio decoder returns a negative data_size.

Originally committed as revision 473 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Don't allocate 0 bytes of memory. It upsets electricFence!
Philip Gladstone [Thu, 9 May 2002 01:24:27 +0000 (01:24 +0000)]
* Don't allocate 0 bytes of memory. It upsets electricFence!

Originally committed as revision 472 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Fix a problem with the first sample when down sampling.
Philip Gladstone [Thu, 9 May 2002 01:23:49 +0000 (01:23 +0000)]
* Fix a problem with the first sample when down sampling.
* Note that this code needs to be fixed -- the rate conversion from 48000->44100
  sounds horrible!

Originally committed as revision 471 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Every frame is a key_frame
Philip Gladstone [Thu, 9 May 2002 01:22:51 +0000 (01:22 +0000)]
* Every frame is a key_frame

Originally committed as revision 470 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Start using enumerated types (makes debugging much easier)
Philip Gladstone [Thu, 9 May 2002 01:22:30 +0000 (01:22 +0000)]
* Start using enumerated types (makes debugging much easier)

Originally committed as revision 469 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Make put_wav_header return the number of bytes that it put.
Philip Gladstone [Thu, 9 May 2002 01:21:56 +0000 (01:21 +0000)]
* Make put_wav_header return the number of bytes that it put.

Originally committed as revision 468 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Change extern inline to static inline so that it will compile without optimization
Philip Gladstone [Thu, 9 May 2002 01:21:24 +0000 (01:21 +0000)]
* Change extern inline to static inline so that it will compile without optimization

Originally committed as revision 467 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Try and fill out some more of the RV header fields correctly
Philip Gladstone [Thu, 9 May 2002 01:20:53 +0000 (01:20 +0000)]
* Try and fill out some more of the RV header fields correctly

Originally committed as revision 466 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add defaults to switch statements
Philip Gladstone [Thu, 9 May 2002 01:19:33 +0000 (01:19 +0000)]
* Add defaults to switch statements

Originally committed as revision 465 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add a sleep when doing the post to ffserver. Yes, this is the wrong
Philip Gladstone [Thu, 9 May 2002 01:19:15 +0000 (01:19 +0000)]
* Add a sleep when doing the post to ffserver. Yes, this is the wrong
  solution.

Originally committed as revision 464 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years agoIntialize the codec type to video.
Philip Gladstone [Thu, 9 May 2002 01:18:42 +0000 (01:18 +0000)]
Intialize the codec type to video.

Originally committed as revision 463 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add the gop_size to the video parameters. Also the audio framesize.
Philip Gladstone [Thu, 9 May 2002 01:18:08 +0000 (01:18 +0000)]
* Add the gop_size to the video parameters. Also the audio framesize.
* Copy the duration over as well, though I'm not 100% certain that that is
  still needed.

Originally committed as revision 462 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Change extern inline to static inline. This improves compiling without optimization.
Philip Gladstone [Thu, 9 May 2002 01:16:45 +0000 (01:16 +0000)]
* Change extern inline to static inline. This improves compiling without optimization.

Originally committed as revision 461 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add default cases to switch statements.
Philip Gladstone [Thu, 9 May 2002 01:16:08 +0000 (01:16 +0000)]
* Add default cases to switch statements.

Originally committed as revision 460 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Add a duration field to the AVPacket
Philip Gladstone [Thu, 9 May 2002 01:15:45 +0000 (01:15 +0000)]
* Add a duration field to the AVPacket

Originally committed as revision 459 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Make it work with sound cards (like mine) that can only capture in stereo.
Philip Gladstone [Thu, 9 May 2002 01:15:21 +0000 (01:15 +0000)]
* Make it work with sound cards (like mine) that can only capture in stereo.
* Add a kludge to allow the left channel to be inverted -- my tv card/sound card
  ends up with the left channel = minus right channel. Converting to mono by
  adding the channels doesn't work well.

Originally committed as revision 458 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Many, many changes to make it work (mostly). Information snagged from
Philip Gladstone [Thu, 9 May 2002 01:13:43 +0000 (01:13 +0000)]
* Many, many changes to make it work (mostly). Information snagged from
  staring at headers of real ASF files. ASFRecorder and avifile were
  other useful sources.

Originally committed as revision 457 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Updated so that some streams actually work!
Philip Gladstone [Thu, 9 May 2002 01:11:49 +0000 (01:11 +0000)]
* Updated so that some streams actually work!

Originally committed as revision 456 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* These are my notes on streaming
Philip Gladstone [Thu, 9 May 2002 01:11:27 +0000 (01:11 +0000)]
* These are my notes on streaming

Originally committed as revision 455 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Fix a nasty problem with output buffering not have enough (or large enough)
Philip Gladstone [Thu, 9 May 2002 01:11:08 +0000 (01:11 +0000)]
* Fix a nasty problem with output buffering not have enough (or large enough)
  buffers. In fact, the code was pretty much shot.
* Try to fool WMP into thinking that we are a microsoft server.
* When we establish a stream to a user, copy the codec information from
  that saved as part of the stream. This gives us the real frame_size and
  other important parameters.
* ASF needs to know about key frames, so add some logic to copy this information
  around.
* When we get the data from ffmpeg as part of a feed, make sure that we save
  the actual codec parameters.
* Allow configuration of AudioCodec and VideoCodec
* Make sure that we delete the feed file before starting. This is not ideal
  but it makes things work a whole lot better!

Originally committed as revision 454 to svn://svn.ffmpeg.org/ffmpeg/trunk

17 years ago* Fix nasty problem with unitialized memory in the file_table
Philip Gladstone [Thu, 9 May 2002 01:06:59 +0000 (01:06 +0000)]
* Fix nasty problem with unitialized memory in the file_table
* Add default entries to switch statements to catch programming errors
* Fix problem with high CPU usage when capturing live (the audio grabber
  is non-blocking for some reason). This is the stream_no_data stuff.
* Handle the video grabber getting behind in its work -- PKT_FLAG_DROPPED_FRAME

Originally committed as revision 453 to svn://svn.ffmpeg.org/ffmpeg/trunk