tsMuxer has the ability to detect audio delay for TS/M2TS/MPG/VOB/EVO sources, but not from MKV source. This patch adds ability to detect audio delay for MKV source.
Fixes#156.
All pgStreams3D[pgTrackNum] (coming from clpi) must be changed to streamInfo (coming from mpls)
When 3d-plane is "FF", this means it is 'undefined' (i.e. 2D), not 'zero'.
QString::toDouble (and ::toFloat) behaviour changed between Qt4 and Qt5 : in
the older version, the function tried to convert the string according to the
current locale and, if unsuccessful, fell back to parsing according to the "C"
locale.
In Qt5, the conversion is always performed in the "C" locale, which means that
all the shenanigans related to finding and replacing the decimal point are not
needed anymore, since the main tsMuxer binary always runs in the "C" locale due
to not calling setlocale() at all.
Also, if Qt 4.8 documentation is to be believed, all this code wasn't ever
needed in the first place, as the conversion function would've just used the
"C" locale after failing to convert the string due to a mismatch in the used
decimal point character - and thus succeeded.
Because of the addition of reader.skipBits(toPassBits); , tsMuxer crashes when toPassBits is above 32 bits.
The commit 45fbb43 is useless, as the various parts of mpls and clpi are always contiguous.
GitHub runners only checkout the particular revision that's going to be built
instead of the whole tree, which results in the revision count always being set
to 1. The short revision checksum is enough anyway.
This replaces the occurrences of the static version number with a version string
if building in a Git repository and Git is available on the build host.
This isn't perfect, since the revision is only going to be fetched when running
qmake or cmake, while it's possible that it changes between consecutive
compilations, not just reconfigurations. However, the main purpose of this is
to add a version tag to the builds built centrally by the GitHub runners, which
always configure the project from scratch.
Fixes#127.
Including raw C headers caused a certain version of libc++ not to find the floating-point version of abs(), which caused compilation errors due to argument type ambiguity.
The repository now contains a clang-format file which should be used when committing new code. Additionally, a new workflow job is added which checks the conformance of source files to the specified formatting rules.
Fixes#47.
For w64, data length includes data metadata (16 bytes) and size (8 bytes)
So 24 bytes have to be substracted from the length read in the stream.
For lpcm, tsMuxer considers frames of 5 ms and rounds up the last block for frame alignment.
So it is normal for the demux to be slightly larger thant the original wav stream.
Fixes#136.
* Don't include <QtGui>, which is a leftover from Qt4 and causes literally
everything to be pulled into the TU.
* Use forward declarations instead of includes where possible.
* Replace aggregated Ui* classes with aggregation via pointer, which allows
these classes to be forward-declared and thus further reduce the number of
included headers. Aggregation via pointer has been the default in Qt Creator
for some time now.
* Fix minor warnings reported by Clang.
* Move QnCheckBoxedHeaderView to a separate file.
* Move QtvCodecInfo to a separate file, change initializer list to inline member initialization.
* Create an actual QMake .pro file instead of stuffing it in MXE build scripts.
* Fix 32-bit integers being used as pointers to QtvCodecInfo objects in 64-bit Windows builds.
* Update information in the "About" tab.
Limiting the reading pace of packets to the Bluray Tranfer Rate value actually creates another issue -timegap between two PCR packets create by tsMuxer can be larger than 100ms.
This issue can be worse than the one solved...
I haven't managed to solve this new issue, so for the time being I prefer to revert on the previous merge #101.
Version is automatically switched to V3 when HEVC is detected.
In case the user wants to force the V3 format for AVC, he will be able to do it via the V3 option in the Blu-ray tab.
tsMuxer called FT_Done_Face also if FT_New_Face reported an error. This caused
FT_Done_Face to be fed with a pointer that was either already freed in a
previous iteration of the loop, or was totally uninitialised from the start.
Fixes#63
fps info can be either in VPS or SPS nal, or both, or none.
tsMuxer reads fps from VPS nal only.
This patch allows tsMuxer to find fps in SPS when not present in VPS.
Plus bug sorted out in flags reading (8 bits in lieu of 1 bit)
I_end values in ClipInf CPI Table are determined by thresholds on I-frame sizes.
These threshold sizes are different for TS_Recording_Rate <= 60000000 (HD), and for TS_Recording_Rate > 60000000 (4K).
This patch solves issue #107
The commit must be merged after commit #111 for V3_flags / Dolby Vision Integration.
- Separate the V3 flags (for index.bdmv) from the HDR metadata (for .mpls)
- With V3_flags the UHD_BLURAY DiskType is not needed anymore
- Add subpath for Dolby Vision
- Change BD Type to 66/100 GB and TS-Recording_Rate to 109 mbps when 4K is detected
Since AppImage essentially creates a separate filesystem that the GUI binary
runs from, it's not possible for it to be aware of the existence of the main
binary that's alongside the actual AppImage file that's supposed to be run by
the end user.
The only solution to this that I can see now is copying the main binary into the
AppDir, which makes the file reside alongside the GUI binary inside the
filesystem that AppImage creates.
Fixes#103.
