- NUT-devel - lists.mplayerhq.hu

[nut]: r628 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 16:09:10 2008 New Revision: 628 Log: Issue pcm-frames solution Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 16:09:10 2008 @@ -151,6 +151,8 @@ A. Define an maximum number of samples ( B. Define an maximum timespam (like 0.1 sec) C. Define an maximum number of bytes (like 1024) +Implemented: A + Issue pcm-sample-format -----------------------

1 0

[nut]: r627 - docs/nut.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 16:05:46 2008 New Revision: 627 Log: Strict 1 codec frame rule and RAW audio size limit Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Wed Feb 13 16:05:46 2008 @@ -1068,6 +1068,11 @@ duplicate the role of a container. Any s For example Vorbis in Ogg in NUT is invalid, as is mpegvideo + mpegaudio in MPEG-PS/TS in NUT or dvvideo + dvaudio in DV in NUT. +A demuxer MUST NOT demux a nut frame containing anything else than exactly a +single codec frame. A muxer MUST NOT store anything except exactly 1 single +codec frame per nut frame. +For RAW audio up to 4096 samples shall be considered 1 codec frame, the actual +number of sampes MAY vary between nut frames. Sample code (Public Domain, & untested):

1 0

[nut]: r626 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 15:20:55 2008 New Revision: 626 Log: Issue stream-chapter-overlap consensus (again scream if you disagree!) Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 15:20:55 2008 @@ -51,6 +51,8 @@ issue, that chapters might not end exact all streams. There could be a few samples difference. Do we care about this? +It seems the consensus is: no, we dont care + Issue info-stream-subsets -------------------------

1 0

[nut]: r625 - docs/nutissues.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 15:19:05 2008 New Revision: 625 Log: Issue pcm-sample-format consensus If i misinterpreted anyones position please say so! Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Wed Feb 13 15:19:05 2008 @@ -158,6 +158,8 @@ A. fourcc B. extradata C. New field in the stream header +It seems the consensus is: A + Issue pcm-interleaving ----------------------

1 0

[nut]: r623 - docs/nut.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 15:14:01 2008 New Revision: 623 Log: Header elision support Sizes of tables are limited to allow demuxers with little memory. Headers are stored in a seprate table instead of straight in the frame code table to avoid <max header size>*256 size problems. And to allow elision headers to be selected through an index in the frame header. New demuxers should be able to read old files. Old demuxers can read new files as long as the index is not coded in the frame header, though they will of course be missing the header in the outputed frames if a non empty one has been used. Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Wed Feb 13 15:14:01 2008 @@ -248,6 +248,7 @@ main_header: tmp_mul=1 tmp_stream=0 tmp_match=1-(1<<62) + tmp_head_idx= 0; for(i=0; i<256; ){ tmp_flag v tmp_fields v @@ -261,7 +262,8 @@ main_header: if(tmp_fields>5) count v else count= tmp_mul - tmp_size if(tmp_fields>6) tmp_match s - for(j=7; j<tmp_fields; j++){ + if(tmp_fields>7) tmp_head_idx v + for(j=8; j<tmp_fields; j++){ tmp_reserved[i] v } for(j=0; j<count && i<256; j++, i++){ @@ -277,8 +279,12 @@ main_header: pts_delta[i]= tmp_pts; reserved_count[i]= tmp_res; match_time_delta[i]= tmp_match; + header_idx[i]= tmp_head_idx; } } + header_count_minus1 v + for(i=0; i<header_count_minus1; i++) + elision_header[i+1] vb reserved_bytes stream_header: @@ -326,6 +332,9 @@ frame: if(frame_flags&FLAG_MATCH_TIME){ match_time_delta s } + if(frame_flags&FLAG_HEADER_IDX){ + header_idx v + } if(frame_flags&FLAG_RESERVED) frame_res v for(i=0; i<frame_res; i++) @@ -618,6 +627,7 @@ flags[frame_code], frame_flags (v) otherwise data_size_msb is 0. 6 FLAG_CHECKSUM If set, the frame header contains a checksum. 7 FLAG_RESERVED If set, reserved_count is coded in the frame header. + 10 FLAG_HEADER_IDX If set, header_idx is coded in the frame header. 11 FLAG_MATCH_TIME If set, match_time_delta is coded in the frame header 12 FLAG_CODED If set, coded_flags are stored in the frame header. @@ -693,6 +703,23 @@ match_time_delta[frame_code] (s) match_time_delta, otherwise this value is ignored. MUST be <32768 and >-32768 or =1-(1<<62) +header_idx[frame_code] (v) + The index into the elision_header table. + MUST by <128 + +header_count_minus1 (v) + The number of distinct non empty elision headers. + MUST be <128 + +elision_header[header_idx] (vb) + For frames with a final size <= 4096 this header is prepended to the + frame data. That is if the stored frame is 4000 bytes and the + elision_header is 96 bytes then it is prepended, if it is 97 byte then it + is not. + elision_header[0] is fixed to a length 0 header + The length of each elision_header except header 0 MUST be < 256 and >0 + The sum of the lengthes of all elision_headers MUST be <=1024 + lsb_pts Least significant bits of the pts in time_base precision. Example: IBBP display order

1 0

[nut]: r621 - docs/nut.txt
by cladisch 13 Feb '08

13 Feb '08

Author: cladisch Date: Wed Feb 13 13:31:05 2008 New Revision: 621 Log: fix typo Modified: docs/nut.txt Modified: docs/nut.txt ============================================================================== --- docs/nut.txt (original) +++ docs/nut.txt Wed Feb 13 13:31:05 2008 @@ -691,7 +691,7 @@ match_time_delta (s) match_time_delta[frame_code] (s) If FLAG_MATCH_TIME is not set then this value shall be used for match_time_delta, otherwise this value is ignored. - MUST be <32786 and >-32768 or =1-(1<<62) + MUST be <32768 and >-32768 or =1-(1<<62) lsb_pts Least significant bits of the pts in time_base precision.

1 0

[nut]: r620 - docs/nut4cc.txt
by michael 13 Feb '08

13 Feb '08

Author: michael Date: Wed Feb 13 13:25:03 2008 New Revision: 620 Log: clarify mp4v Modified: docs/nut4cc.txt Modified: docs/nut4cc.txt ============================================================================== --- docs/nut4cc.txt (original) +++ docs/nut4cc.txt Wed Feb 13 13:25:03 2008 @@ -33,6 +33,8 @@ MPG4 MS MPEG4v1 (!NOT ISO MPEG4!) MP42 MS MPEG4v2 MP43 MS MPEG4v3 MP4V ISO MPEG4 Part 2 Video + !for non compliant mpeg4 like old divx4/odivx use the respective + fourcc, this fourcc is only for fully comliant mpeg4! mpg1 ISO MPEG1 Video mpg2 ISO MPEG2 Video MRLE MS RLE

1 0

[nut]: r617 - src/trunk/Makefile
by cladisch 13 Feb '08

13 Feb '08

Author: cladisch Date: Wed Feb 13 09:01:58 2008 New Revision: 617 Log: Fix the nutmerge dependencies to prevent nutmerge.h from being treated as a source file. Modified: src/trunk/Makefile Modified: src/trunk/Makefile ============================================================================== --- src/trunk/Makefile (original) +++ src/trunk/Makefile Wed Feb 13 09:01:58 2008 @@ -2,6 +2,7 @@ include config.mak LIBNUT_OBJS = libnut/muxer.o libnut/demuxer.o libnut/reorder.o libnut/framecode.o NUTUTILS_PROGS = nututils/nutmerge nututils/nutindex nututils/nutparse +NUTMERGE_OBJS = nututils/nutmerge.o nututils/demux_avi.o nututils/demux_ogg.o nututils/framer_mp3.o nututils/framer_mpeg4.o nututils/framer_vorbis.o all: libnut nututils @@ -19,7 +20,8 @@ $(LIBNUT_OBJS): libnut/priv.h libnut/lib nututils: $(NUTUTILS_PROGS) -nututils/nutmerge: nututils/demux_avi.c nututils/demux_ogg.c nututils/framer_mp3.c nututils/framer_mpeg4.c nututils/framer_vorbis.c nututils/nutmerge.h libnut/libnut.a +$(NUTMERGE_OBJS): nututils/nutmerge.h +nututils/nutmerge: $(NUTMERGE_OBJS) libnut/libnut.a $(NUTUTILS_PROGS): CFLAGS += -Ilibnut

4 4

[nut]: r618 - src/trunk/nututils/nutparse.c
by cladisch 13 Feb '08

13 Feb '08

Author: cladisch Date: Wed Feb 13 09:03:14 2008 New Revision: 618 Log: Do not print the unneeded intermediate value of the sync point back pointer calculation. Modified: src/trunk/nututils/nutparse.c Modified: src/trunk/nututils/nutparse.c ============================================================================== --- src/trunk/nututils/nutparse.c (original) +++ src/trunk/nututils/nutparse.c Wed Feb 13 09:03:14 2008 @@ -490,7 +490,7 @@ static void parse_syncpoint(uint64_t pos for (i = 0; i < stream_count; ++i) streams[i].last_pts = convert_ts(pts, tbid, streams[i].time_base_id); value = read_var(); - printf(" back_ptr_div16: %"PRIu64" (%"PRIu64") -> 0x%"PRIx64"\n", value, value * 16 + 15, pos - (value * 16 + 15)); + printf(" back_ptr_div16: %"PRIu64" (0x%"PRIx64")\n", value, pos - (value * 16 + 15)); } static void parse_index(void)

1 0

[nut]: r616 - docs/nutissues.txt
by michael 12 Feb '08

12 Feb '08

Author: michael Date: Tue Feb 12 23:04:46 2008 New Revision: 616 Log: Mention last consensus about raw and stream headers. Modified: docs/nutissues.txt Modified: docs/nutissues.txt ============================================================================== --- docs/nutissues.txt (original) +++ docs/nutissues.txt Tue Feb 12 23:04:46 2008 @@ -167,3 +167,5 @@ D. Only allow 1 standard interleaving What about the interleaving of non raw codecs, do all specify the interleaving, or does any leave it to the container? If so, our options would be down to only C. + +Also a field specific to just raw does not belong in the stream headers.

1 0