| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
0.8.2 is a middle-size update of the 0.8.0 branch:
- ARM32 optimizations for ipred and itx in 10/12bits,
completing the 10b/12b work on ARM64 and ARM32
- Give the post-filters their own threads
- ARM64: rewrite the wiener functions
- Speed up coefficient decoding, 0.5%-3% global decoding gain
- x86 optimizations for CDEF_filter and wiener in 10/12bit
- x86: rewrite the SGR AVX2 asm
- x86: improve msac speed on SSE2+ machines
- ARM32: improve speed of ipred and warp
- ARM64: improve speed of ipred, cdef_dir, cdef_filter, warp_motion and itx16
- ARM32/64: improve speed of looprestoration
- Add seeking, pausing to the player
- Update the player for rendering of 10b/12b
- Misc speed improvements and fixes on all platforms
- Add a xxh3 muxer in the dav1d application
|
|
|
|
this has not been building all year, upstream development is discontinued.
|
|
Fix build problem with php56 maintainer-zts PKG_OPTION is enabled.
|
|
Version 55.0.0 "Waiting For Space"
Bug fixes
* all programs, only on Windows: fixed checking existence of & creating
directories with UNC paths such as `\\server\share\folder\file.mkv`.
* all programs, only on Windows: implemented several workarounds for bugs in
`std::filesystem` wrt. UNC paths. The effect of those bugs was that Blu-rays
and DVDs couldn't be added from UNC paths.
* mkvmerge, only on Windows: fixed non-ASCII characters getting mangled in the
destination file name when splitting is active.
* MKVToolNix GUI, only on Windows: fixed the GUI hanging when trying to browse
for files on drives that don't actually exist.
* MKVToolNix GUI, only on Windows: the GUI did not remember UNC paths as the
"most recently used source directory" properly.
* MKVToolNix GUI: only on Windows: when dragging & dropping files from other
applications to MKVToolNix GUI the GUI will force the drop action to be a
copy action, no matter which keys were pressed by the users. This prevents
Windows Explorer from deleting the dragged file when the user presses shift
while dropping the file onto MKVToolNix GUI.
* MKVToolNix GUI, only on Windows: when using the file dialog for selecting a
file name to save to (e.g. when browsing for the destination file name in
the multiplexer or saving an attachment in the header editor) non-ASCII
characters in the default file name were mangled.
* MKVToolNix GUI: multiplexer: the user can now configure what to do when
using the "Add source files" button independently of the setting for what
happens when files are dragged & dropped onto the GUI. The same options are
available, but they're distinct settings, allowing for different use
cases.
* MKVToolNix GUI: multiplexer: deriving the track language from file names
doesn't use one big regular expression of all enabled languages
anymore. Instead, the file name is split into parts on a list of characters
(such as `.` or `(` and `)`; configurable in the preferences). This prevents
the regular expression becoming too big for the regular expression library
to handle when many languages are enabled.
* MKVToolNix GUI: preferences: the language lists for "deriving track
languages" and "enabling items by language" always included all the ISO
639-3 languages, even if those weren't enabled in the "often-used languages"
pane.
Other changes
* mkvmerge, MKVToolNix GUI multiplexer: the functionality for changing the
length of the NALU "size" field for AVC/H.264 and HEVC/H.265 tracks has been
removed. The corresponding command-line option will continue to be
recognized by mkvmerge as not to break existing third-party applications,
but it won't actually do anything.
|
|
Version 54.0.0 "F Maj Pixie"
New features and enhancements
* mkvmerge: added support for using ISO 639-3 language codes in IETF BF47
language tags.
* mkvmerge: AC-3 parser: added support for byte-swapped AC-3 data.
* mkvmerge: Matroska reader: for audio tracks that have the bit depth track
header set mkvmerge will now keep that header even for codecs that don't
require it for decoding.
* mkvmerge: MPEG transport stream reader, PCM audio tracks: mkvmerge will now
re-order the channels for 5.1, 7.0 and 7.1 channel tracks from the Blu-ray
layout to the WAVEFORMATEXTENSIBLE layout expected in Matroska.
* mkvmerge, mkvinfo, mkvpropedit, MKVToolNix GUI: added support for the
following new track header elements: "hearing impaired" flag, "visual
impaired" flag, "text descriptions" flag, "original" flag, "commentary"
flag.
* MKVToolNix GUI: added support for using ISO 639-3 language codes in IETF
BF47 language tags. As there are several thousand of them, they're
deactivated by default and must be activated in the preferences ("GUI" →
"Often used selections" → "Languages").
* MKVToolNix GUI: multiplexer: when adding Blu-rays the user can select
multiple playlists to add simultaneously in the "select playlist to add"
dialog.
* MKVToolNix GUI: multiplexer: the file name extensions "eb3" and "ec3" were
added for Dolby Digital Plus & "mpl" for Dolby TrueHD in the file
dialogs.
* MKVToolNix GUI: multiplexer: when adding multiple files the dialog asking
the user what to do with them has gained a new checkbox. If enabled, all
files containing at least one video track will always be placed in newly
created multiplex setting.
* MKVToolNix GUI: multiplexer: added a menu entry in the "Multiplexer" for
adding all files that are currently in the clipboard.
Bug fixes
* all: Windows: fixed compatibility with gettext 0.21 and newer on mingw.
* all: Windows: fixed several of the programs having problems with certain
Unicode characters (primarily emojis) in file names (e.g. mkvextract
wrongfully complaining about an "invalid mode" or the GUI not being able to
find parts of Blu-ray file structures).
* mkvextract: AAC: fixed wrong channel mask field in the ADTS headers for 7.1
channel layouts.
* mkvextract: h.265/HEVC extraction: if the first frame starts with the
parameter sets (SPS, PPS & VPS), the ones from CodecPrivate aren't written
and the ones from the first frame are kept.
* mkvmerge: fixed the calculation of chapter timestamps read from NTSC
DVDs.
* MKVToolNix GUI: IETF BCP 47 language widget: the language combo box will now
always contain the language code the user enters in the free-form field,
even if it isn't in the list of often-used languages the user configured in
the preferences.
* MKVToolNix GUI: multiplexer: when browsing for the destination file name the
default directory is now chosen according to the preferences regarding how
the destination file name should be formed. For example, if the policy is
set to "fixed output directory" then that output directory will be the one
initially set when the directory selection dialog is opened.
* MKVToolNix GUI: multiplexer: fixed the removal of appended source files if
the "delete source files" end-of-job action is enabled.
* MKVToolNix GUI: chapter editor: when importing chapters from DVDs the IETF
BCP 47 language elements will be set, too, not just the legacy language
elements.
Build system changes
* libEBML v1.4.2 and libMatroska v1.6.3 are now required. The optional,
bundled copies of both libraries have been updated to those versions. This
bump in requirements fixes several heap overflow bugs in libEBML.
* MKVToolNix is now using the C++17 library feature "file system library"
instead of Boost's "file system" and "system" libraries. For the GNU
Compiler Collection (gcc) libstdc++ this means v8 or newer is required; for
clang's libc++ it means v7 or newer. For macOS this means that provided disk
image will only run on 10.15 "Catalina" or newer.
|
|
|
|
|
|
unbreak build builds.
|
|
qt5/kf5 version.
|
|
qt5/kf5 version
|
|
version 4.3.2:
avcodec/hapdec: Change compressed_offset to unsigned 32bit
avformat/rmdec: Check codec_length without overflow
avformat/mov: Check element count in mov_metadata_hmmt()
avcodec/vp8: Move end check into MB loop in vp78_decode_mv_mb_modes()
avcodec/fits: Check gcount and pcount being non negative
avformat/nutdec: Check timebase count against main header length
avformat/electronicarts: Clear partial_packet on error
avformat/r3d: Check samples before computing duration
avcodec/pnm_parser: Check av_image_get_buffer_size() for failure
avformat/wavdec: Consider AV_INPUT_BUFFER_PADDING_SIZE in set_spdif()
avformat/rmdec: Check remaining space in debug av_log() loop
avformat/flvdec: Treat high ts byte as unsigned
avformat/samidec: Sanity check pts
avcodec/jpeg2000dec: Check atom_size in jp2_find_codestream()
avformat/avidec: Use 64bit in get_duration()
avformat/mov: Check for duplicate st3d
avformat/mvdec: Check for EOF in read_index()
avcodec/jpeglsdec: Fix k=16 in ls_get_code_regular()
avformat/id3v2: Check the return from avio_get_str()
avcodec/hevc_sei: Check payload size in decode_nal_sei_message()
libavutil/eval: Remove CONFIG_TRAPV special handling
avformat/wtvdec: Check len in parse_chunks() to avoid overflow
avformat/asfdec_f: Add an additional check for the extradata size
avformat/3dostr: Check sample_rate
avformat/4xm: Make audio_frame_count 64bit
avformat/mov: Use av_mul_q() to avoid integer overflows
avcodec/vp9dsp_template: Fix integer overflows in itxfm_wrapper
avformat/rmdec: Reorder operations to avoid overflow
avcodec/mxpegdec: fix SOF counting
avcodec/rscc: Check inflated_buf size whan it is used
avformat/mvdec: Sanity check SAMPLE_WIDTH
avcodec/nvenc: fix timestamp offset ticks logic
avformat/rmdec: Fix codecdata_length overflow check
avcodec/simple_idct: Fix undefined integer overflow in idct4row()
avformat/wavdec: Check block_align vs. channels before combining them
avformat/tta: Use 64bit intermediate for index
avformat/soxdec: Check channels to be positive
avformat/smacker: Check for too small pts_inc
avformat/sbgdec: Use av_sat_add64() in str_to_time()
avcodec/cscd: Check output len in zlib as in lzo
avcodec/vp3: Check input amount in theora_decode_header()
avformat/wavdec: Check avio_get_str16le() for failure
avformat/flvdec: Check for EOF in amf_skip_tag()
avformat/aiffdec: Check size before subtraction in get_aiff_header()
avformat/electronicarts: More chunk_size checks
avcodec/cfhd: check peak.offset
avformat/tedcaptionsdec: Check for overflow in parse_int()
avformat/nuv: Check channels
avcodec/siren: Increase noise category 5 and 6
avformat/mpc8: Check size before implicitly converting to int
avformat/nutdec: Fix integer overflow in count computation
avformat/mvi: Use 64bit for testing dimensions
avformat/utils: Check dts in update_initial_timestamps() more
avformat/mpsubdec: Use av_sat_add/sub64() in fracval handling
avformat/flvdec: Check for avio_read() failure in amf_get_string()
avformat/flvdec: Check for nesting depth in amf_skip_tag()
avformat/flvdec: Check for nesting depth in amf_parse_object()
avformat/asfdec_o: Check for EOF in asf_read_marker()
avformat/flvdec: Use av_sat_add64() for pts computation
avformat/utils: Check dts - (1<<pts_wrap_bits) overflow
avformat/bfi: Check chunk_header
avformat/ads: Check size
avformat/iff: Check block align also for ID_MAUD
avcodec/utils: Check for integer overflow in get_audio_frame_duration() for ADPCM_DTK
avformat/fitsdec: Better size checks
avformat/mxfdec: Fix integer overflow in next position in mxf_read_local_tags()
avformat/avidec: dv does not support palettes
avformat/dhav: Break out of infinite dhav search loop
libavformat/utils: consider avio_size() failure in ffio_limit()
avformat/nistspheredec: Check bits_per_coded_sample and channels
avformat/asfdec_o: Check size vs. offset in detect_unknown_subobject()
avformat/utils: check for integer overflow in av_get_frame_filename2()
avutil/timecode: Avoid undefined behavior with large framenum
avformat/mov: Check a.size before computing next_root_atom
avformat/sbgdec: Reduce the amount of floating point in str_to_time()
avformat/mxfdec: Free all types for both Descriptors
uavformat/rsd: check for EOF in extradata
avcodec/wmaprodec: Check packet size
avformat/dhav: Check position for overflow
avcodec/rasc: Check frame before clearing
avformat/vividas: Check number of audio channels
avcodec/alsdec: Fix integer overflow with quant_cof
avformat/mpegts: Fix argument type for av_log
avformat/cafdec: clip sample rate
avcodec/ffv1dec: Fix off by 1 error with quant tables
avformat/mpegts: Increase pcr_incr width to 64bit
avcodec/utils: Check bitrate for overflow in get_bit_rate()
avformat/mov: Check if hoov is at the end
avcodec/hevc_ps: check scaling_list_dc_coef
avformat/iff: Check data_size
avformat/matroskadec: Sanity check codec_id/track type
avformat/rpl: Check the number of streams
avformat/vividas: Check sample_rate
avformat/vividas: Make len signed
avcodec/h264idct_template: Fix integer overflow in ff_h264_chroma422_dc_dequant_idct()
avformat/dsfdec: Check block_align more completely
avformat/mpc8: Check remaining space in mpc8_parse_seektable()
avformat/id3v2: Sanity check tlen before alloc and uncompress
avformat/vqf: Check len for COMM chunks
avformat/mov: Avoid overflow in end computation in mov_read_custom()
avcodec/hevc_cabac: Limit value in coeff_abs_level_remaining_decode() tighter
avformat/cafdec: Check the return code from av_add_index_entry()
avformat/cafdec: Check for EOF in index read loop
avformat/cafdec: Check that bytes_per_packet and frames_per_packet are non negative
avformat/mpc8: correct integer overflow in mpc8_parse_seektable()
avformat/mpc8: correct 32bit timestamp truncation
avcodec/exr: Check ymin vs. h
avformat/avs: Use 64bit for the avio_tell() output
avformat/wavdec: More complete size check in find_guid()
avcodec/mv30: Use unsigned in idct_1d()
avformat/iff: Check size before skip
avformat/rmdec: Check for EOF in index packet reading
avcodec/vp3dsp: Use unsigned constant to avoid undefined integer overflow in ff_vp3dsp_set_bounding_values()
avformat/icodec: Check for zero streams and stream creation failure
avformat/icodec: Factor failure code out in read_header()
avformat/bintext: Check width
avformat/sbgdec: Check that end is not before start
avformat/lvfdec: Check stream_index before use
avformat/au: cleanup on EOF return in au_read_annotation()
avformat/mpegts: Limit copied data to space
avformat/bintext: Check width in idf_read_header()
avformat/iff: check size against INT64_MAX
avformat/vividas: improve extradata packing checks in track_header()
avformat/paf: Check for EOF in read_table()
avformat/gxf: Check pkt_len
avformat/aiffdec: Check packet size
avformat/concatdec: use av_strstart()
avformat/wavdec: Refuse to read chunks bigger than the filesize in w64_read_header()
avformat/rsd: Check size and start before computing duration
avformat/vividas: better check of current_sb_entry
avformat/iff: More completely check body_size
avformat/vividas use avpriv_set_pts_info()
avformat/xwma: Check for EOF in dpds_table read code
avcodec/utils: Check sample rate before use for AV_CODEC_ID_BINKAUDIO_DCT in get_audio_frame_duration()
avcodec/dirac_parser: do not offset AV_NOPTS_OFFSET
avformat/rmdec: Make expected_len 64bit
avformat/pcm: Check block_align
avformat/lrcdec: Clip timestamps
avutil/mathematics: Use av_sat_add64() for the last addition in av_add_stable()
avformat/electronicarts: Check for EOF in each iteration of the loop in ea_read_packet()
avformat/ifv: Check that total frames do not overflow
avcodec/vp9dsp_template: Fix some overflows in iadst8_1d()
avcodec/fits: Check bscale
avformat/nistspheredec: Check bps
avformat/jacosubdec: Use 64bit inside get_shift()
avformat/genh: Check block_align
avformat/mvi: Check count for overflow
avcodec/magicyuv: Check slice size before reading flags and pred
avformat/asfdec_f: Check for negative ext_len
avformat/bethsoftvid: Check image dimensions before use
avformat/genh: Check block_align for how it will be used in SDX2_DPCM
avformat/au: Check for EOF in au_read_annotation()
avformat/vividas: Check for zero v_size
avformat/segafilm: Do not assume AV_CODEC_ID_NONE is 0
avformat/segafilm: Check that there is a stream
avformat/wtvdec: Check dir_length
avformat/ffmetadec: finalize AVBPrint on errors
avcodec/decode/ff_get_buffer: Check for overflow in FFALIGN()
avcodec/exr: Check limits to avoid overflow in delta computation
avformat/boadec: Check that channels and block_align are set
avformat/asfdec_f: Check name_len for overflow
avcodec/h264idct_template: Fix integer overflow in ff_h264_chroma422_dc_dequant_idct()
avformat/sbgdec: Check for timestamp overflow in parse_time_sequence()
avcodec/aacdec_fixed: Limit index in vector_pow43()
avformat/kvag: Fix integer overflow in bitrate computation
avcodec/h264_slice: fix undefined integer overflow with POC in error concealment
avformat/rmdec: sanity check coded_framesize
avformat/flvdec: Check for EOF in amf_parse_object()
avcodec/mv30: Fix multiple integer overflows
avcodec/smacker: Check remaining bits in SMK_BLK_FULL
avcodec/cook: Check subpacket index against max
avcodec/utils: Check for overflow with ATRAC* in get_audio_frame_duration()
avcodec/hevcpred_template: Fix diagonal chroma availability in 4:2:2 edge case in intra_pred
avformat/icodec: Change order of operations to avoid NULL dereference
avcodec/exr: Fix overflow with many blocks
avcodec/vp9dsp_template: Fix integer overflows in idct16_1d()
avcodec/ansi: Check initial dimensions
avcodec/hevcdec: Check slice_cb_qp_offset / slice_cr_qp_offset
avcodec/sonic: Check for overread
avformat/subviewerdec: fail on AV_NOPTS_VALUE
avcodec/exr: Check line size for overflow
avcodec/exr: Check xdelta, ydelta
avcodec/celp_filters: Avoid invalid negation in ff_celp_lp_synthesis_filter()
avcodec/takdsp: Fix negative shift in decorrelate_sf()
avcodec/dxtory: Fix negative stride shift in dx2_decode_slice_420()
avformat/asfdec_f: Change order or operations slightly
avformat/dxa: Use av_rescale() for duration computation
avcodec/vc1_block: Fix integer overflow in ac value
avcodec/mv30: Fix several integer overflows in idct_1d()
avformat/iff: Check data_size not overflowing int64
avcodec/dxtory: Fix negative shift in dx2_decode_slice_410()
avcodec/sonic: Check channels before deallocating
avformat/vividas: Check for EOF in first loop in track_header()
avformat/wvdec: Check rate for overflow
avcodec/ansi: Check nb_args for overflow
avformat/wc3movie: Cleanup on wc3_read_header() failure
avformat/wc3movie: Move wc3_read_close() up
avcodec/tiff: Fix default white level
avcodec/diracdsp: Fix integer anomaly in dequant_subband_*
avutil/fixed_dsp: Fix integer overflows in butterflies_fixed_c()
avcodec/mv30: Check remaining mask in decode_inter()
avcodec/wmalosslessdec: Check remaining space before padding and channel residue
avformat/cdg: Fix integer overflow in duration computation
avcodec/mpc: Fix multiple numerical overflows in ff_mpc_dequantize_and_synth()
avcodec/agm: Fix off by 1 error in decode_inter_plane()
avformat/electronicarts: Check if there are any streams
avcodec/ffwavesynth: Fix integer overflow in wavesynth_synth_sample / WS_SINE
avcodec/vp9dsp_template: Fix integer overflow in iadst8_1d()
avformat/avidec: Fix io_fsize overflow
avcodec/cfhd: Check transform type
avcodec/tiff: Check jpeg context against jpeg frame parameters
avcodec/tiff: Restrict tag order based on specification
avcodec/tiff: Avoid abort with DNG RAW TIFF with YA8
avcodec/tiff: Check the linearization table size
avformat/siff: Reject audio packets without audio stream
avformat/mpeg: Check avio_read() return value in get_pts()
avcodec/tiff: Check bpp/bppcount for 0
avcodec/snowdec: Sanity check hcoeff
avformat/mov: Check comp_brand_size
avformat/ape: Error out in case of EOF in the header
avcodec/alac: Check decorr_shift to avoid invalid shift
avcodec/tdsc: Fix tile checks
opusdec: do not fail when LBRR frames are present
configure: update copyright year
avfilter/vf_framerate: fix infinite loop with 1-frame input
avformat/url: Change () position in ff_make_absolute_url()
avformat/mpegts: make sure mpegts_read_header always stops at the first pmt
avformat/alp: fix handling of TUN files
avformat/argo_asf: fix handling of v1.1 files
swscale/x86/yuv2rgb: fix crashes when loading alpha from unaligned buffers
lavf/url: fix relative url parsing when the query string or fragment has a colon
avformat/libsrt: fix cleanups on failed libsrt_open() and libsrt_setup()
avcodec/cuviddec: backport extradata fixes
avcodec/cuviddec: handle arbitrarily sized extradata
lavf/srt: fix build fail when used the libsrt 1.4.1
avformat/libsrt: close listen fd in listener mode
lavf/url: rewrite ff_make_absolute_url() using ff_url_decompose().
lavf/url: add ff_url_decompose().
avcodec/cbs_av1: fix setting FrameWidth in frame_size_with_refs()
avcodec/cbs_av1: use a more appropiate AV1ReferenceFrameState pointer variable name
avcodec/cbs_av1: fix handling reference frames on show_existing_frame frames
avcodec/cbs_av1: infer frame_type in show_existing_frame frames earlier
avcodec/cbs_av1: add OrderHint to CodedBitstreamAV1Context
avcodec/cbs_av1: infer frame_type when parsing a show_existing_frame frame
cbs_av1: Fix test for presence of buffer_removal_time element
avcodec/cbs_av1: fix storage size for render_{width,height}_minus_1
lavc: Lower MediaFoundation audio encoder priority.
x86/yuv2rgb: fix crashes when storing data on unaligned buffers
checkasm/vf_blend: use the correct depth parameters to initialize the blend modes
x86/vf_blend: fix warnings about trailing empty parameters
x86/h264_deblock: fix warning about trailing empty parameter
avutil/x86inc: fix warnings when assembling with Nasm 2.15
|
|
dav1d is a AV1 decoder that focuses on speed and correctness
|
|
Release v1.6.3.
* libEBML v1.4.2 is now required.
* Added classes for new track header elements:
"KaxFlagHearingImpaired", "KaxFlagVisualImpaired",
"KaxFlagTextDescriptions", "KaxFlagOriginal" and
"KaxFlagCommentary".
|
|
Bump PKGREVISION.
|
|
|
|
Version 53.0.0 "Fool's Gold"
New features and enhancements
* mkvmerge: AVI reader: added support for reading the video aspect ratio from
the video properties header (`vprp` chunk) if present and setting the
display dimensions accordingly.
* mkvmerge: MP4 reader: for h.264/AVC tracks that don't have an
`AVCConfigurationBox` (`avcC` atom) in their sample description (`stsd`)
atom or whose `avcC` atom contains no content `mkvmerge` will now re-derive
the `AVCConfigurationBox` from the bitstream.
* mkvextract: mkvextract will now check if any of the destination file names
is the same as the source file name and abort with an error if that's the
case.
* MKVToolNix GUI: when querying the user for a file name for saving things
(e.g. multiplexer settings or an attachment in the header editor), the
automatically suggested file name will now be based on the
situation-specific file names (e.g. the destination file name for
multiplexer settings or the attachment's name when saving an attachment in
the header editor) instead of the directory's name.
* MKVToolNix GUI: multiplexer: when deriving track languages from file names
the GUI will now select the right-most match instead of the left-most
one. For example, "La.vie.en.rose.(fr).srt" will now be detected as French
(fr) instead of English (en).
* MKVToolNix GUI: preferences: the items in the "pre-defined …" lists can now
be renamed by double-clicking with the mouse or pressing the F2 key.
* Windows installer: the `bluray_dump` command-line utility will be installed
into the `tools` sub-directory. `bluray_dump` can read & dump certain file
types used on Blu-rays: `.mpls` playlists, `.clpi` clip information
databases, `.bdmv` index files, `bdmt_….xml` disc library databases and
`tnmt_….xml` track & chapter name databases.
Bug fixes
* mkvmerge: stretching chapter timestamps with `--chapter-sync` now works
correctly with floating point values including fractions of floating point
numbers (e.g. `12.3/45.67`). The tooltips in the GUI have been adjusted
accordingly.
* mkvmerge: MPEG 1/2 video handling: the "default duration" header field was
often half the value it actually should be, resulting in all video frames
having an explicit block duration with the correct value. This has been
fixed with a patch by Tom Yan.
* mkvmerge: MPEG 1/2 video handling: the data stored in `Codec private` and
`Codec state` doesn't contain extensions other than sequence & sequence
display extensions anymore. Fix by Tom Yan.
* mkvmerge: tag handling: when remuxing a Matroska file with the
`--no-track-tags`, existing `SOURCE_ID` track tags are now skipped, too.
* MKVToolNix GUI: multiplexer: the drop-down boxes with pre-defined track
names now follow the order set in the preferences instead of sorting the
entries alphabetically.
|
|
Changes between 3.0.11.1 and 3.0.12:
----------------------------------
Access:
* Add new RIST access module compliant with simple profile (VSF_TR-06-1)
Access Output:
* Add new RIST access output module compliant with simple profile (VSF_TR-06-1)
Demux:
* Fixed adaptive's handling of resolution settings
Audio output:
* Fix audio distortion on macOS during start of playback
Video Output:
* Direct3D11: Fix some potential crashes when using video filters
macOS:
* Add native support for Apple Silicon / ARM-64
* Visual UI adaptations for macOS Big Sur
* Fix displaying EQ bands in the UI depending on which frequency
presets are set for the EQ in advanced preferences
* Fix UI issues in bookmarks window
Misc:
* Several fixes in the web interface, including privacy and security
improvements
* Update YouTube and Vocaroo scripts
|
|
TheoraPlay is a simple library to make decoding of Ogg Theora videos
easier.
A tiny example to pull data out of an .ogv file is about 50 lines of C
code, and a complete SDL-based media player is about 300 lines of code.
TheoraPlay is optimized for multicore CPUs, and is designed to be
programmer-friendly. You will need libogg, libvorbis, and libtheora, of
course, but then you just drop a .c file and two headers into your
project and you're ready to hook up video decoding, without worrying
about Ogg pages, Vorbis blocks, or Theora decoder state.
Places TheoraPlay can be found in use: - Shank. - Crypt of the
Necrodancer. - 20XX. - FNA.
|
|
it is only used by example programs.
|
|
0.8.0
Added additional parsing for simple CUE-OUT with Duration key
Added _cueout_duration to _parse_cueout
Added AVERAGE-BANDWIDTH to IFramePlaylist
Added VIDEO-RANGE to IFramePlaylist
Added new playlist delta attributes from v10
Added HDCP-LEVEL to StreamInfo and IFramePlaylist
|
|
It's no longer necessary due to the configure script patches and
disabling assembly.
Bump PKGREVISION.
|
|
|
|
Changelog:
26.1.2 macOS Hotfix Changes
Browser sources on macOS now support hardware accelerated rendering on macOS 10.15 and above, reducing resource usage
Fixed a crash on startup (by updating CEF to version 4183)
26.1.1 Hotfix Changes
Fixed macOS 10.13 crash on startup [PatTheMav]
Fixed decklink devices not working correctly [DDRBoxman]
Fixed browser source crashing on certain versions of macOS [eric]
Fixed buttons not displaying text correctly on context bar when using Acri theme [Warchamp7]
Fixed a bug with slideshow source where if you use randomize and have "restart when visible" active, it would not transition on activate [cg2121]
Fixed a bug where duplicated scene items would not have their lock states duplicated [cg2121]
Fixed a bug where the default canvas resolution would be scaled incorrectly if fractional scaling was enabled [RytoEX]
Fixed jack on Linux deadlocking on close [marcan]
Fixed jack on Linux going out of sync sometimes [marcan]
Added webp to image formats that you can browse when using the image source [f3ndot]
26.1 New Features and Additions
Added Virtual Camera output on macOS [johnboiles/PatTheMav]
Added Virtual Camera output on Linux (requires v4l2loopback-dkms) [catxfish/cg2121]
Added the ability to use a separate audio track for the VOD when using Twitch [Jim]
If using Simple output mode, enable "Enable Advanced Encoder Settings", and enable "Twitch VOD Track (Uses Track 2)". Twitch VOD output will then be on audio track 2
If using Advanced output mode in the Streaming tab, enable "Twitch VOD Track" and select the track you'd like to use for it
Special thanks to Twitch for assisting during the development of this feature
Added OpenBSD support [grayed]
Added the ability to ingest captions coming from Decklink devices via "Decklink Captions" from the Tools menu [DDRBoxman]
Added hardware decoding options for stinger transitions [WizardCM]
Added an option to duplicate filters in the right-click context menu of filters [exeldro]
Added ability to copy and paste a single filter between sources [cg2121]
Added HLS support and ingests for YouTube [ushadow]
Added a Replay buffer save event to the frontend API [hgonomeg]
26.1 Improvements/Tweaks
Updated dependencies on Windows and macOS (such as x264 and Qt) to their latest versions for the latest performance improvements [Jim]
On Linux, the program will now detect other instances that are currently running and warn the user about running more than one copies at a time [clockley]
When creating a new profile, you will be now be given the option to run the auto-configuration wizard [JohannMG]
Changed the "Enforce Streaming Service Bitrate" to "Ignore streaming service setting recommendations", moved it to the Streaming section of the Settings window, and made it so it now affects both Simple and Advanced output modes. [Jim]
Maximum limitations for streaming services are now shown in the Streaming section of the Settings window [Jim]
Streaming services may now apply resolution and framerate limits on streams to their services [Jim]
26.1 Bug Fixes
Fixed a bug where the "Save Replay" button would stay highlighted after being clicked [offthegrid-mike]
Fixed a bug where the Media source would have unusual lag playing back certain media files (particularly WMV files) [Jim]
Fixed an issue where Facebook streaming would allow you to select resolutions/framerates that aren't supported [Jim]
Fixed certain connection issues with SRT/mpegts [pkviet]
Fixed a bug where you could disable the Replay Buffer even if it was currently active [Scrxtchy]
Fixed a bug where renaming a source would not return keyboard/mouse focus back to the list [jberenhaus]
Fixed an issue where 'Paste Filters' on sources didn't work in all situations [WizardCM]
Fixed an issue where Virtual Camera & Source Toolbar hotkeys would be duplicated when switching profiles [WizardCM]
Fixed an issue with removing signal handlers in Lua [Scrxtchy]
Fixed a potential hang when repeatedly selecting and deselecting display capture sources on Windows [exeldro]
Fixed an issue where the replay buffer could use settings from CQP mode in CBR mode [R1CH]
Fixed an issue where certain recordings could not be remuxed (typically from the QSV encoder) [R1CH]
|
|
Changelog:
## HandBrake 1.3.3
### All platforms
#### General
- Fixed ISO 639-2/B language codes not set correctly in MKV (affects Hebrew, Indonesian, Javanese, and Yiddish) (#2903)
- Improved support for sources where pixel format cannot be quickly identified, e.g. due to delayed video track start (#2893)
- Added logging to identify where hardware support is disabled
- Miscellaneous bug fixes and improvements
#### Video
- Improved Intel QSV memory footprint by eliminating a buffer pool (#2675)
- Improved Intel QSV H.265 memory buffer size as required by newer Intel Media SDK (#2862)
- Fixed and improved Intel QSV in various situations, especially hardware decoding (#873, #2660, #2661, #2829)
- Fixed full range video being not being identified as limited range after conversion where filters are used (#2859)
#### Subtitles
- Fixed handling of overlapping SSA import subtitles (791adbac)
- Improved support for out-of-order SSA subtitles as allowed by specification (#2906)
#### Command line interface
- Fixed --preset failure unless full path is specified, e.g. --preset="Category Name/Preset Name" (#2838)
#### Build system
- Improved Flatpak to better conform to freedesktop metainfo standards
- Improved Intel QSV Flatpak plugin build effiency using cmake-ninja
- Added a patch to fix cross compiling libdav1d using GCC 10.x (quality of life improvement)
- Official HandBrake 1.3.3 Windows release is built using GCC 9.x and is not directly affected by this issue
#### Third-party libraries
- Updated libraries
- FFmpeg 4.2.3 (decoding and filters)
### Mac
- Fixed preview layout not displaying properly on OS X 10.11 El Capitan
- Fixed incorrect copyright year on About dialog (#2830)
### Windows
- Fixed a crash related to the dark theme (#2816)
- Fixed a potential crash related to preview image memory allocation (#2871)
- Fixed a potential crash due to certain actions causing no preset being selected (#2875)
- Fixed missing E-AC-3 encoder option (#2855)
- Fixed hardware encoder support unavailable in portable build (#2832)
- Miscellaneous bug fixes and improvements
## HandBrake 1.3.2
### All platforms
#### General
- Fixed point to point end detection in certain scenarios (#2603)
- Improved support for H.265 video in AVI container produced by some security cameras (#2622)
- Added logging to identify problematic sources where container and video track pixel aspect ratios differ
- Added logging to help debug potential JSON API issues
#### Video
- Fixed color range conversion being applied twice when scaling video (#2561)
- Fixed incorrect identification of support for QSV HEVC encoder on older Intel hardware (#2558)
- Added logging to identify automatic picture rotation
- Miscellaneous bug fixes and improvements
#### Audio
- Fixed an upstream FFmpeg issue where passing through AAC ADTS audio could produce invalid MKV output in rare cases (error instead) (#2809)
#### Filters
- Fixed uninitialized memory in NLMeans prefilter leading to video corruption at bottom of picture (only affects custom settings) (#2576)
- Fixed a crash in the Detelecine filter with out of bounds parameters (only affects custom settings) (#2560, #2804)
#### Subtitles
- Fixed burned in subtitles position offset where cropscale filter is not used (#2449)
#### Command line interface
- Fixed subtitles not being selected when specifying --all-subtitles without also specifying a non-empty --subtitle-lang-list
#### Build system
- Fixed building the GTK graphical interface for use on Windows (link ole32)
- Updated Flatpak manifest creation script for compatibility with Python 3
- Updated Flatpak runtime and numactl library versions, QSV plugin
- Improved minimum version dependencies to facilitate building on systems with older automake and pkg-config
- Added a workaround for an upstream libdav1d issue affecting installation on FreeBSD (#2662)
- Miscellaneous bug fixes and improvements
### Linux
- Fixed point to point controls not accepting fractional seconds
- Fixed updating presets with identical names in different categories
- Improved parity with other platforms by allowing removal of preset categories (automatic after last preset in category is removed)
- Improved parity with other platforms by showing title and chapter range on the queue summary tab
### Mac
- Fixed selection behavior new track audio mixdown set to DPL2 instead of stereo (#2641)
- Fixed queued job failures related to removable drives by resolving security scoped resources as needed (#2566)
### Windows
- Fixed loading preset files with Unicode characters in path (#2427)
- Fixed clear queue options to prevent them clearing active jobs (#2587)
- Fixed main window status label not always reflecting the true count of queue jobs (#2538)
- Fixed failure loading default settings which could cause various issues in the graphical interface (#2549)
- Fixed preview images displaying incorrectly in some cases (anamorphic none, flip horizontal) (e9675bb, #2764)
- Fixed various issues related to Auto Passthru, including fallback settings (#2619, #2627, #2611)
- Fixed exported presets not importing correctly using the Mac graphical interface (#2531)
- Fixed pause and resume not working correctly in some cases (#2647)
- Fixed display of times greater than 24 hours (estimated time renaming, paused duration) (#2582, #2649)
- Fixed various cosmetic issues in the graphical interface (#2645, #2646)
- Improved display of long filenames in the queue (#2570)
- Improved some UX stress cases related to the queue (#2632)
- Improved error message when importing a preset specifying a nonexistent audio encoder (#2638)
- Improved audio and subtitle languages behavior to preserve selected languages order where "any" is also selected (#2611)
- Improved low disk space preferences and alerts (#2648)
- Added a workaround for an upstream .NET issue causing tooltips to not render correctly in some cases (#2630)
- Miscellaneous bug fixes and improvements
|
|
Simplify, per joerg@'s suggestion "Drop the whole block. Seems much
safer."
|
|
(But still having another problems for now, sorry)
|
|
Fixes PR pkg/55912
|
|
Rather than add to the ifdef stew that's already there, simply don't
try to second-guess the host platform. Otherwise, we'd need more macro
checks added for, at least, NetBSD, FreeBSD, and musl, but including
sys/types.h should "just work".
(There's seemingly another unrelated portability issue directly below
that code block, but it won't break builds, and would require different
means of correction.)
|
|
Version 52.0.0 "Secret For The Mad"
New features and enhancements
* MKVToolNix GUI: job queue: the maximum number of jobs to run concurrently
can now be increased in the preferences. The default remains
at 1.
* MKVToolNix GUI: the GUI will now add a context-specific default extension to
file names selected for saving on platforms that don't add one itself
(e.g. GNOME). For example, when saving multiplexer settings the extension
`.mtxcfg` will be added.
* MKVToolNix GUI: added an option to the preferences for the window to stay on
top of other windows.
Bug fixes
* mkvextract: h.265/HEVC extraction: the code for skipping extraction of
prefix SEI NALUs in the first frame was skipping two bytes too few,
resulting in broken processing of all following bytes. Patch by Mike Chen.
* libEBML: the optional, bundled version of libEBML was updated to v1.4.1.
Build system changes
* libEBML v1.4.1 is now required due to a bug in libEBML that caused pointers
to just-freed memory being returned to the caller under certain invalid data
constellations, causing use-after-free errors in all of MKVToolNix's
programs.
|
|
Make this package build a dependent library consistently on all pkgsrc
platforms, so we don't get PLIST mismatches.
|
|
|
|
Packaged in wip by Charlotte Koch
Open Asset Import Library (short name: Assimp) is a portable Open Source
library to import various well-known 3D model formats in a uniform manner.
The most recent version also knows how to export 3D files and is therefore
suitable as a general-purpose 3D model converter. See the feature list.
Assimp aims to provide a full asset conversion pipeline for use in game
engines/realtime rendering systems of any kind, but it is not limited to
this purpose. In the past, it has been used in a wide range of applications.
Changes from the wip package
-Updated HOMEPAGE
-Adjusted CATEGORIES to be consistent with DragonFly, FreeBSD and OpenBSD
-Re-build buildlink3.mk
|
|
|
|
|
|
|
|
|
|
Presumably a doxygen change means this package now generates different
files
|
|
Noticed by John D. Baker on pkgsrc-users
|
|
I mis-hand-merged the patch and incorrect removed a closing paren.
(It should also be noted that this same patch was sent by Edgar Fuß
earlier.)
|
|
This simply extends the existing list of NetBSD/arm and
NetBSD/aarch64.
(This is a build fix, so no PKGREVISION++.)
From Mark Felder via pkgsrc-users
|
|
Bump PKGREVISION
|
|
|
|
Upstream decided to make wayland support linux-only.
This doesn't actually matter much at all since you can just use the SDL
video output...
|
|
pkgsrc changes:
- Remove `ass' option: libass is now always needed and enabled
- Add `openal' option to enable openal ao (disabled by default)
- Remove pkgsrc patches ao-s that were removed in 0.33.0 (a possible
patch for oss ao is under discussion upstream:
https://github.com/mpv-player/mpv/pull/8312 )
Changes:
0.33.0
======
Features
--------
Added
~~~~~
- scripting: load scripts from directories
- mac: activate logging when started from the bundle
- ytdl_hook.lua: delay load subtitles
- sub: add an option to filter subtitles by regex
- scripting: add a way to run subprocesses as "scripts"
- command: implement asynchronous commands and support for named arguments
- player: add optional separate video decoding thread
- vo_gpu: d3d11: add support for exclusive fullscreen
- w32_common: Support HiDPI on Windows
- vo_x11: add 10 bit support
- vo_gpu: add BT.2390 tone-mapping
- client API: add software rendering API
- audio: add scaletempo2 filter based on chromium
- auto_profiles: add this script
- stream: Implement slice:// for reading slices of streams
- player: add automatic loading of external cover art files
- vo_sixel: implement terminal video output using sixel
Changed
~~~~~~~
- sws_utils: use zimg by default if available
- build: pick up Lua 5.2 by default (preferred over 5.1)
- build: disable RPI vendor blob auto-detection in favor of open-source stack
- build: make C11 atomics mandatory
- build: make libass non-optional
- player: stricter filename matching for external subtitle auto-loading
- lua: support Unicode paths in script loading and IO library
- vo_direct3d: rip out texture video rendering path and dumb down OSD rendering
- audio: rewrite internal audio handling and AO API
- build: disable GLX by default
Removed
~~~~~~~
- stream_libarchive: disable tar support due to bugs
- Remove remains of Libav compatibility
- stream_smb: remove due to lack of thread safety and the abundance
of alternatives, FFmpeg still includes SMB support
- command: remove legacy hook API (has been deprecated for a long time)
- client API: remove deprecated qthelper.hpp header
- removed audio outputs: sndio, rsound, oss
- x11: remove xdg-screensaver invocations that supported dbus based idle inhibit
- client API: deactivate the opengl_cb API
- build system: drop Python 2 compatbility
Options and Commands
--------------------
Added
~~~~~
- demux: add option to disable cache "sharing" between back and forward buffers
- player: add ab-loop-count option/property
- ytdl_hook: add a way to use ytdl's default formats
- ytdl_hook: add all_formats option that loads all formats that were found
- demux_mkv: document probe-start-time option and enable it by default
- command: extend osd-overlay command with bounds reporting
- player: a number of new playlist contol commands/properties
- ipc: add --input-ipc-client option
- options: add option to control display-sync factor
- vo_gpu: add better gamut clipping option
- vo_gpu: vulkan: add ability to disable events
- x11: add option to make window appear on a specific workspace
- wayland: expose wayland-app-id as a user option
- player: add --subs-with-matching-audio option
- command: add read-only focused property
- screenshot: option to use software rendering for screenshots
- command: add delete-watch-later-config
- command: new property mouse-pos with current position and hover state
Changed
~~~~~~~
- options: remove deprecation warning for "-foo bar" syntax
- player: make audio hr-seek default
- ad_lavc: disable decoder downmix by default
- command: support save-position-on-quit for "stop" command too
- command: extend subprocess command
- options: do not accept ":" as separator anymore in key/value lists
Deprecated
~~~~~~~~~~
- demux: deprecate --cache-secs
Removed
~~~~~~~
- wayland: remove wayland-frame-wait-offset option
- input: remove deprecated --input-file option
- vo_vdpau: remove deprecated/inactive --vo-vdpau-deint option
Fixes and Minor Enhancements
----------------------------
- options: stop hiding deprecated options from --help output
- lua, js: add mp.get_script_directory() function
- lua: fix security relevant loading order issue with scripts
- player: make screenshot each-frame mode more accurate
- bash completion: complete ao/af/vo/vf options
- zimg: add alpha support
- wayland: make resizing better
- edl: add mechanism for delay loading streams
- stream_file: fix caching-related performance regression on CIFS
- cocoa-cb: fix crashes and issues with UI updates
- sub, demux: improve behavior with negative subtitle delay/muxed subs
- demux: make seek ranges work for static images + audio
- cocoa-cb: support maximize/minimize on startup
- umpv: change from legacy FIFO to socket
- stats: move input speed to cache page, make it a graph
- stats: add fourth page with performance graphs
- command: print edition title to OSD when cycling
- zimg: add support for big endian input and output
- demux_mkv: add png intra support
- build: detect VT_GETMODE on FreeBSD and DragonFly
- win32: use windows 10 native virtual-terminal if available
- vo_gpu: enable frame caching for still frames
- command: add property to return text subtitles in ASS
- vo_gpu: ra_pl: add timers support
- build: allow vo_wlshm on more Wayland platforms (e.g. FreeBSD)
- zimg: add slice threading and use it by default
- command: add a way to access properties of a current track
- vo_gpu: EGL: fix transparency on X11/EGL/Mesa
- vd_lavc: add AV1 to the default allowed hwdec codec list
|
|
xine-lib (1.2.11) 2020-12-06
* Add gopher input plugin.
* Add user settings for IPv6.
* Split a52 spdif output to separate decoder.
* Code cleanup.
* More engine optimization.
* Small optimizations (opengl2, dvb subtitles, png, mpeg-ts).
* Better A/V sync.
* Fix a few crashes.
* Fix bluray playback.
* Fix VAAPI picture jitter and green flashes.
* Fix Opus and EAC3 audio playback.
* Fix text subtitles.
* Fix mpeg-ts (hdmv mode, detection of cut TV recordings).
* Fix mpeg layer 1-4 audio timing.
* Fix some VDR issues.
* Fix HLS VOD mode.
* Many small fixes.
* Build fixes.
* BSD fixes.
* Update german translation.
* Update internal libdca to upstream 0.0.7.
|
|
|
|
pkgsrc changes:
---------------
* Add patch corresponding to a Github open pull request
* Update from c++03 to c++11 in USE_LANGUAGES to accept nullptr
upstream changes:
-----------------
* No clear ChangeLog in the repository
|
|
No release notes yet, but seems no security updates
(no updates for IE and Edge versions).
https://helpx.adobe.com/flash-player/release-note/fp_32_air_32_release_notes.html
https://helpx.adobe.com/security/products/flash-player.html
|
|
For the Python 3.8 default switch.
|
