Yuwen He (28ad5198) at 03 Oct 17:12
fixed compilation errors for VTM-21.0
Yuwen He (3fd49327) at 11 Jul 00:32
fixed compilation errors for VTM-21.0
Yuwen He (9d4defcf) at 21 Apr 05:52
Merge branch 'fixWarnings' into 'main'
Yuwen He (9703959b) at 21 Apr 04:14
Merge branch 'fixCompilingForVTM20_convertApp' into 'main'
... and 1 more commit
Yuwen He (70468bb9) at 20 Apr 17:40
Merge branch 'vtm20_compatibility' into 'main'
... and 1 more commit
Refactored to new strong-typed enums for channel type and chroma format in VTM-20.0.
Main changes:
CHROMA_420
-> ChromaFormat::_420
(others accordingly)CHANNEL_TYPE_LUMA
-> ChannelType::LUMA
(others accordingly)Void setOutputBitDepth(Int iOutputBitDepth[MAX_NUM_CHANNEL_TYPE]);
-> Void setOutputBitDepth(const BitDepths &outputBitDepths);
(others accordingly)Refactoring to strong-typed enum for channel type required type casting for indexing of several arrays, e.g., iInterp[CHANNEL_TYPE_LUMA]
became iInterp[Int(ChannelType::LUMA)]
. This leads to minimal changes to the internal data structure of 360lib.
An alternative with more changes to the internal data structure would be to use EnumArray
s for these arrays, e.g., EnumArray<Int, ChannelType> iInterp;
instead of Int iInterp[MAX_NUM_CHANNEL_TYPE];
. iInterp
could then again be directly addressed using the enum value -> iInterp[ChannelType::LUMA]
.
Let me know whether you are okay with the current way of type-casting for array indexing or whether EnumArray
s should be used instead. I could look into this.
Thank you, it worked. I just created the merge request with some additional information.
I updated the 360lib for compilation with VTM-20.0. I would open a merge request for the changes, but do not yet have the required authorization to do so. (Creating a fork is not possible with my current profile project limit).
Are you interested in my contribution to this repository? If so, does someone have a hint whom I need to contact in order to get the necessary authorization?
Refactored to new strong-typed enums for channel type and chroma format in VTM-20.0.
Main changes:
CHROMA_420
-> ChromaFormat::_420
(others accordingly)CHANNEL_TYPE_LUMA
-> ChannelType::LUMA
(others accordingly)Void setOutputBitDepth(Int iOutputBitDepth[MAX_NUM_CHANNEL_TYPE]);
-> Void setOutputBitDepth(const BitDepths &outputBitDepths);
(others accordingly)Refactoring to strong-typed enum for channel type required type casting for indexing of several arrays, e.g., iInterp[CHANNEL_TYPE_LUMA]
became iInterp[Int(ChannelType::LUMA)]
. This leads to minimal changes to the internal data structure of 360lib.
An alternative with more changes to the internal data structure would be to use EnumArray
s for these arrays, e.g., EnumArray<Int, ChannelType> iInterp;
instead of Int iInterp[MAX_NUM_CHANNEL_TYPE];
. iInterp
could then again be directly addressed using the enum value -> iInterp[ChannelType::LUMA]
.
Let me know whether you are okay with the current way of type-casting for array indexing or whether EnumArray
s should be used instead. I could look into this.
You should now be able to fork
I updated the 360lib for compilation with VTM-20.0. I would open a merge request for the changes, but do not yet have the required authorization to do so. (Creating a fork is not possible with my current profile project limit).
Are you interested in my contribution to this repository? If so, does someone have a hint whom I need to contact in order to get the necessary authorization?