Here’s another release of MKVToolNix. Several bugs have been fixed. A rather important one is an update to the JSON library used which fixes immense memory consumption when loading stored settings (e.g. a ~500 KB JSON file leading to consumption of more than 2 GB of memory).
Enhancements include support for WebVTT subtitles, much improved support for teletext subtitles in MPEG transport streams and several more fields in the JSON/verbose identification formats.
There have been no changes for package maintainers since the previous release.
Here’s the full ChangeLog since v8.9.0:
Here are the usual links:
…to the source code: http://mkvtoolnix.download/source.html
…to the binaries: http://mkvtoolnix.download/downloads.html
The Windows and Mac OS binaries are available. Most of the Linux binaries are still being built and will be available in a couple of hours.
Here’s the full ChangeLog since v9.0.1:
- 2016-04-23 Moritz Bunkus <email@example.com>
- Released v9.1.0 "Little Earthquakes".
- mkvmerge: MPEG TS/teletext enhancement: included the teletext page number in the JSON/verbose identification output as track property "teletext_page".
- mkvmerge: MPEG TS bug fix: the "text_subtitles" property of the JSON/verbose identification modes was always set to true for all subtitle tracks, even for those that aren’t text subtitles (VobSub, PG).
- mkvmerge: MPEG TS/teletext enhancement: if a teletext track contains multiple teletext pages then mkvmerge will now recognize all of those pages as separate tracks to merge instead of only merging the first page. See #1662.
- 2016-04-21 Moritz Bunkus <firstname.lastname@example.org>
- mkvmerge: MPEG TS/teletext enhancement: mkvmerge will now ignore obviously bogus PTS values for teletext tracks and use PTS from earlier audio or video packets instead. See #1662.
- 2016-04-20 Moritz Bunkus <email@example.com>
- mkvmerge: MPEG TS/teletext bug fix: the language code signaled in the MPEG TS PMT is taken into account when selecting the character encoding to use during decoding of the teletext subtitles, not just the "national character set" stored in the teletext page headers. For example, a German teletext page may signal "national character set" 0 (English) whereas it’s actually 4 (German). See #1662.
- mkvmerge: teletext decoding bug fix: fixed dropping of certain non-ASCII characters in rare circumstances due to wrong filtering of already UTF-8 encoded strings.
- mkvmerge: MPEG TS reader enhancement: teletext tracks of type 5 (hearing impaired) are recognized as subtitles, too. Implements #1662.
- 2016-04-16 Moritz Bunkus <firstname.lastname@example.org>
- MKVToolNix GUI: merge tool enhancement: characters that aren’t valid in path names are automatically removed from the output file name. Implements #1647.
- MKVToolNix GUI: bug fix (Windows only): the GUI didn’t start if the USERNAME environment variable contained characters that aren’t allowed in file names (e.g. : or ?).
- 2016-04-13 Moritz Bunkus <email@example.com>
- mkvmerge: AVI reader bug fix: fixed reading files where the file ends in the middle of an audio chunk. Fixes #1657.
- 2016-04-11 Moritz Bunkus <firstname.lastname@example.org>
- mkvextract: new feature: added support for extracting WebVTT subtitles. Implements the extraction part of #1592.
- mkvmerge: new feature: added support for reading WebVTT subtitles from WebVTT and Matroska files. Implements the merge part of #1592.
- mkvmerge: enhancement: when reading Matroska files not created by mkvmerge that contain chapters the existing edition UIDs and chapter UIDs are removed and random ones created. This is necessary as not only HandBrake but other tools assign sequential numbers starting at 1 for each file. Therefore there are two chapter entries with the UID 1, two with the UID 2 etc. and those should, strictly speaking, be treated as if they were a single chapter whereas the user expects those entries to stay separate entries.
- 2016-04-09 Moritz Bunkus <email@example.com>
- MKVToolNix GUI: new feature: added an option in the preferences ("Merge" → "Output") for controlling whether or not the GUI clears the "output file name" input upon removal of the last file.
- MKVToolNix GUI: new feature: added an option in the preferences ("Merge" → "Default values") for controlling whether or not the GUI clears the "file title" input upon removal of the last file.
- mkvmerge: bug fix: mkvmerge will no longer abort reading a Matroska file with a structural error right before the first cluster. Fixes #1654.
- 2016-04-03 Moritz Bunkus <firstname.lastname@example.org>
- mkvmerge: enhancement: added the muxing date in both local time zone and UTC to verbose/JSON identification outputs (keys "date_local" and "date_utc", formatted after ISO 8601) when identifying Matroska files.
- 2016-04-02 Moritz Bunkus <email@example.com>
- mkvmerge: enhancement: added the minimum timestamp for each track in verbose/JSON identification outputs (key "minimum_timestamp") when identifying Matroska files. At most the first ten seconds are probed; if no block is found for a track within that range then the key is not output for the track. Also added "muxing_application" and "writing_application" to the "container" section of the output. Currently those are only set for Matroska files.
- 2016-03-31 Moritz Bunkus <firstname.lastname@example.org>
- MKVToolNix GUI: merge tool change: attachments from source files have been moved from the "Tracks, chapters, tags and attachments" list on the "sources" tab to a new list on the "attachments" tab. That way all existing attachments and all the ones to newly add will be shown in a single tab. This makes it easier to decide which attachments will have to be added and which can be removed.
- MKVToolNix: merge tool bug fix: when adding playlists the GUI won’t ask the user whether or not to scan if there’s only a single playlist in that directory.
- 2016-03-29 Moritz Bunkus <email@example.com>
- mkvmerge: bug fix: AVC/h.264: fixed handling of interlaced frames with bottom field first.
- MKVToolNix GUI: bug fix: fixed huge memory consumption (e.g. allocation of 2 GB for a JSON file of 650 KB) in the JSON library by updating said JSON library. Fixes #1631.
Have fun :)