Split AudioProfile between fwk and android.media.audio.common
a.m.a.c adds a version of AudioProfile struct which
is similar to Audio HIDL HAL V7. It lacks some bits used only
by the framework. These are carried in AudioProfileSys.
AudioProfileSys does not contain AudioProfile. This is done
to simplify further splitting of other types like AudioPort.
Updated to/from AIDL conversions in libaudiofondation.
They encapsulate the fact that there are two AIDL structures
per AudioProfile class instance.
Bug: 198812639
Test: atest audiofoundation_parcelable_test
Change-Id: I6834650f9406c552b6dfa42b120769eb38d6f588
diff --git a/media/libaudiofoundation/AudioProfile.cpp b/media/libaudiofoundation/AudioProfile.cpp
index 2d44f4a..15f2d68 100644
--- a/media/libaudiofoundation/AudioProfile.cpp
+++ b/media/libaudiofoundation/AudioProfile.cpp
@@ -156,31 +156,38 @@
return *this;
}
-ConversionResult<media::AudioProfile>
+ConversionResult<AudioProfile::Aidl>
AudioProfile::toParcelable(bool isInput) const {
- media::AudioProfile parcelable;
+ media::audio::common::AudioProfile parcelable;
parcelable.name = mName;
- parcelable.format = VALUE_OR_RETURN(legacy2aidl_audio_format_t_AudioFormatDescription(mFormat));
+ parcelable.format = VALUE_OR_RETURN(
+ legacy2aidl_audio_format_t_AudioFormatDescription(mFormat));
+ // Note: legacy 'audio_profile' imposes a limit on the number of
+ // channel masks and sampling rates. That's why it's not used here
+ // and conversions are performed directly on the fields instead
+ // of using 'legacy2aidl_audio_profile_AudioProfile' from AidlConversion.
parcelable.channelMasks = VALUE_OR_RETURN(
convertContainer<std::vector<AudioChannelLayout>>(
mChannelMasks,
[isInput](audio_channel_mask_t m) {
return legacy2aidl_audio_channel_mask_t_AudioChannelLayout(m, isInput);
}));
- parcelable.samplingRates = VALUE_OR_RETURN(
+ parcelable.sampleRates = VALUE_OR_RETURN(
convertContainer<std::vector<int32_t>>(mSamplingRates,
convertIntegral<int32_t, uint32_t>));
- parcelable.isDynamicFormat = mIsDynamicFormat;
- parcelable.isDynamicChannels = mIsDynamicChannels;
- parcelable.isDynamicRate = mIsDynamicRate;
parcelable.encapsulationType = VALUE_OR_RETURN(
legacy2aidl_audio_encapsulation_type_t_AudioEncapsulationType(mEncapsulationType));
- return parcelable;
+ media::AudioProfileSys parcelableSys;
+ parcelableSys.isDynamicFormat = mIsDynamicFormat;
+ parcelableSys.isDynamicChannels = mIsDynamicChannels;
+ parcelableSys.isDynamicRate = mIsDynamicRate;
+ return std::make_pair(parcelable, parcelableSys);
}
-ConversionResult<sp<AudioProfile>>
-AudioProfile::fromParcelable(const media::AudioProfile& parcelable, bool isInput) {
+ConversionResult<sp<AudioProfile>> AudioProfile::fromParcelable(
+ const AudioProfile::Aidl& aidl, bool isInput) {
sp<AudioProfile> legacy = new AudioProfile();
+ const auto& parcelable = aidl.first;
legacy->mName = parcelable.name;
legacy->mFormat = VALUE_OR_RETURN(
aidl2legacy_AudioFormatDescription_audio_format_t(parcelable.format));
@@ -190,23 +197,24 @@
return aidl2legacy_AudioChannelLayout_audio_channel_mask_t(l, isInput);
}));
legacy->mSamplingRates = VALUE_OR_RETURN(
- convertContainer<SampleRateSet>(parcelable.samplingRates,
+ convertContainer<SampleRateSet>(parcelable.sampleRates,
convertIntegral<uint32_t, int32_t>));
- legacy->mIsDynamicFormat = parcelable.isDynamicFormat;
- legacy->mIsDynamicChannels = parcelable.isDynamicChannels;
- legacy->mIsDynamicRate = parcelable.isDynamicRate;
legacy->mEncapsulationType = VALUE_OR_RETURN(
aidl2legacy_AudioEncapsulationType_audio_encapsulation_type_t(
parcelable.encapsulationType));
+ const auto& parcelableSys = aidl.second;
+ legacy->mIsDynamicFormat = parcelableSys.isDynamicFormat;
+ legacy->mIsDynamicChannels = parcelableSys.isDynamicChannels;
+ legacy->mIsDynamicRate = parcelableSys.isDynamicRate;
return legacy;
}
ConversionResult<sp<AudioProfile>>
-aidl2legacy_AudioProfile(const media::AudioProfile& aidl, bool isInput) {
+aidl2legacy_AudioProfile(const AudioProfile::Aidl& aidl, bool isInput) {
return AudioProfile::fromParcelable(aidl, isInput);
}
-ConversionResult<media::AudioProfile>
+ConversionResult<AudioProfile::Aidl>
legacy2aidl_AudioProfile(const sp<AudioProfile>& legacy, bool isInput) {
return legacy->toParcelable(isInput);
}
@@ -331,16 +339,19 @@
}
ConversionResult<AudioProfileVector>
-aidl2legacy_AudioProfileVector(const std::vector<media::AudioProfile>& aidl, bool isInput) {
- return convertContainer<AudioProfileVector>(aidl,
- [isInput](const media::AudioProfile& p) {
- return aidl2legacy_AudioProfile(p, isInput);
+aidl2legacy_AudioProfileVector(const AudioProfileVector::Aidl& aidl, bool isInput) {
+ return convertContainers<AudioProfileVector>(aidl.first, aidl.second,
+ [isInput](const media::audio::common::AudioProfile& p,
+ const media::AudioProfileSys& ps) {
+ return aidl2legacy_AudioProfile(std::make_pair(p, ps), isInput);
});
}
-ConversionResult<std::vector<media::AudioProfile>>
+ConversionResult<AudioProfileVector::Aidl>
legacy2aidl_AudioProfileVector(const AudioProfileVector& legacy, bool isInput) {
- return convertContainer<std::vector<media::AudioProfile>>(legacy,
+ return convertContainerSplit<
+ std::vector<media::audio::common::AudioProfile>,
+ std::vector<media::AudioProfileSys>>(legacy,
[isInput](const sp<AudioProfile>& p) {
return legacy2aidl_AudioProfile(p, isInput);
});