Archived:MDFDevVideo Record Symbian API
m (Protected "CS001065 - MDFDevVideo Record API" [edit=sysop:move=sysop]) |
|||
| Line 153: | Line 153: | ||
[[Image:DevvideoEx.zip]] | [[Image:DevvideoEx.zip]] | ||
| − | [[Category:Symbian C++]][[Category:S60 3rd Edition, Feature Pack 2 ]][[Category:Code Examples | + | [[Category:Symbian C++]][[Category:S60 3rd Edition, Feature Pack 2 ]][[Category:Code Examples]][[Category:Multimedia]][[Category:Video]] |
Revision as of 13:36, 7 July 2008
| ID | CS001065 | Creation date | July 7, 2008 |
| Platform | S60 3rd Edition, FP2 | Tested on devices | |
| Category | Symbian C++ | Subcategory | Multimedia, Video |
| Keywords (APIs, classes, methods, functions): MDFDevVideo Record API |
- This API is not part of the public SDK. It can be found in the SDK API Plug-in.
Overview
The Video Encoding and Recording API is also known as the DevVideoRecord API. Multimedia Framework (MMF) plug-ins use this interface to access video encoders and preprocessors in a platform and hardware-independent manner.
Header files
#include <devvideorecord.h>Link against
LIBRARY devvideo.libSource file
The MMMFDevVideoObserver class defines the observer mixin interface that any client using CMMFDevVideoRecord must implement.
//MMMFDevVideoRecordObserver
virtual void MdvroReturnPicture(TVideoPicture* aPicture) = 0;
virtual void MdvroSupplementalInfoSent() = 0;
virtual void MdvroNewBuffers() = 0;
virtual void MdvpoError(TInt aError) = 0;
virtual void MdvroInitializeComplete(TInt aError);
Virtual void MdvroStreamEnd();
//NewL()
delete iDvr;
iDvr = NULL;
iDvr = CMMFDevVideoRecord::NewL(*this);
// Find an accelerated encoder if possible, non-accelerated if not
RArray<TUid> encs;
iDvr->FindEncodersL(KMime, 0, encs, EFalse);
TUid enc = TUid::Uid(0);
TBool foundAccel = EFalse;
for ( TInt i = 0; i < encs.Count(); i++ ) {
CVideoEncoderInfo *info = iDvr->VideoEncoderInfoLC(encs[i]);
if ( info->Accelerated() ) {
enc = encs[i];
foundAccel = ETrue;
} else if ( !foundAccel ) {
LFPRINT((_L("Not accelerated, but haven't seen one so far --using this")));
enc = encs[i];
} else {
LFPRINT((_L("Not accelerated and already have one -- skipping")));
}
CleanupStack::PopAndDestroy();
}
if ( enc.iUid == 0 ) {
User::Leave(KErrNotSupported);
}
// Select encoder
iEnc = iDvr->SelectEncoderL(enc);
// Input format
TSize size(176, 144);
TUncompressedVideoFormat fmt;
fmt.iDataFormat = EYuvRawData;
fmt.iYuvFormat.iCoefficients = EYuvBt601Range1;
fmt.iYuvFormat.iPattern = EYuv420Chroma2;
fmt.iYuvFormat.iDataLayout = EYuvDataPlanar;
fmt.iYuvFormat.iYuv2RgbMatrix = NULL;
fmt.iYuvFormat.iRgb2YuvMatrix = NULL;
fmt.iYuvFormat.iAspectRatioNum = 1;
fmt.iYuvFormat.iAspectRatioDenom = 1;
iDvr->SetInputFormatL(iEnc, fmt, size);
// Output format
CCompressedVideoFormat *cfmt = CCompressedVideoFormat::NewL(KMime);
CleanupStack::PushL(cfmt);
iDvr->SetOutputFormatL(iEnc, *cfmt, EDuCodedPicture,EDuElementaryStream,
EFalse);
CleanupStack::PopAndDestroy(); //cfmt
// Buffer options
TEncoderBufferOptions bopt;
bopt.iMaxPreEncoderBufferPictures = 1;
bopt.iHrdVbvSpec = EHrdVbvNone;
bopt.iHrdVbvParams.Set(NULL, 0);
bopt.iMaxOutputBufferSize = 10240;
bopt.iMaxCodedPictureSize = 10240;
bopt.iMaxCodedSegmentSize = 10240; // ignored in non-segmented mode
bopt.iMinNumOutputBuffers = 2;
iDvr->SetBufferOptionsL(bopt);
// Source is memory
iDvr->SetSourceMemoryL(15.0, EFalse, EFalse); // not real-time
// Initialize
iDvr->Initialize();
//WRITE PIC
// Generate data
TInt y = iPicNum % 14;
Mem::Fill(iYuv, 176*y*10, 200);
Mem::Fill(iYuv + 176*y*10, 176*144 - 176*y*10, 0);
Mem::Fill(iYuv + 176*144, 176*144/2, 128);
// Write picture
iPic.iData.iDataFormat = EYuvRawData;
iPic.iData.iDataSize = TSize(176,144);
iPicDataPtr.Set(iYuv, 176*144*3/2, 176*144*3/2);
iPic.iData.iRawData = &iPicDataPtr;
iPic.iOptions = TVideoPicture::ETimestamp;
iPic.iTimestamp = (1000000 * iPicNum) / 15;
TRAPD(err, iDvr->WritePictureL(&iPic));
if ( err != KErrNone ) {
LFPRINT((_L("WritePictureL failed with %d"), err));
TestComplete(err);
}
iPicNum++;
Example project
This application does very basic MPEG-4 video encoding from a memory source and writes the results to a file. It does just about the least amount of work to get everything up and running, and while the settings may be too basic for a real application, it should be useful for comparison purposes. Once the application is running, you can start adding additional rate control and other settings as needed.
