我正在使用AVAssetWriterVideoInput将缓冲区附加到文件中。我有这段代码:
if ( _assetWriter.status == AVAssetWriterStatusWriting ) {
// If the asset writer status is writing, append sample buffer to its corresponding asset writer input
if (mediaType == AVMediaTypeVideo) {
if (_assetWriterVideoInput.readyForMoreMediaData) {
if (![_assetWriterVideoInput appendSampleBuffer:sampleBuffer]) {
NSLog(@"error: %@", [_assetWriter.error localizedFailureReason]);
NSLog(@"error: %@", [_assetWriter.error localizedRecoveryOptions]);
NSLog(@"error: %@", [_assetWriter.error localizedDescription]);
NSLog(@"error: %@", [_assetWriter.error domain]);
NSLog(@"error: %@", [_assetWriter.error userInfo]);
} else
NSLog(@"frame saved");
}
}
这一行
if (![_assetWriterVideoInput appendSampleBuffer:sampleBuffer]) {
以unknown error
和代码-12738
失败,显然在任何documentation上找不到,正如所有Apple文档所预期的那样。
另外,我怀疑这是一个未知的错误,原因很简单,因为AVFoundation中存在大量未知错误的代码,如果系统选择代码-12738
,它显然知道的不仅仅是说它是未知的。
查看正在保存的文件,它保持0兆字节,因为没有保存缓冲区/帧。
此AVAssetWriterVideoInput
的创建方式如下:
CMVideoDimensions dimensions = CMVideoFormatDescriptionGetDimensions(currentFormatDescription);
NSUInteger numPixels = dimensions.width * dimensions.height;
NSUInteger bitsPerSecond;
// Assume that lower-than-SD resolutions are intended for streaming, and use a lower bitrate
NSUInteger bitsPerPixel = 11.4; // This bitrate matches the quality produced by AVCaptureSessionPresetHigh.
bitsPerSecond = numPixels * bitsPerPixel;
NSDictionary *videoCompressionSettings = @{AVVideoCodecKey : AVVideoCodecH264,
AVVideoWidthKey : @(dimensions.width),
AVVideoHeightKey : @(dimensions.height),
AVVideoCompressionPropertiesKey : @{ AVVideoAverageBitRateKey : @(bitsPerSecond),
AVVideoMaxKeyFrameIntervalKey : @(30)} };
if ([_assetWriter canApplyOutputSettings:videoCompressionSettings forMediaType:AVMediaTypeVideo])
{
// Intialize asset writer video input with the above created settings dictionary
_assetWriterVideoInput = [AVAssetWriterInput assetWriterInputWithMediaType:AVMediaTypeVideo outputSettings:videoCompressionSettings];
_assetWriterVideoInput.expectsMediaDataInRealTime = YES;
和附加缓冲区时具有以下特征:
CMSampleBuffer 0x1009e12a0 retainCount: 1 allocator: 0x1b762cbb8
invalid = NO
dataReady = YES
makeDataReadyCallback = 0x0
makeDataReadyRefcon = 0x0
formatDescription = <CMVideoFormatDescription 0x170443210 [0x1b762cbb8]> {
mediaType:'vide'
mediaSubType:'BGRA'
mediaSpecific: {
codecType: 'BGRA' dimensions: 1920 x 1080
}
extensions: {<CFBasicHash 0x17087c2c0 [0x1b762cbb8]>{type = immutable dict, count = 2,
entries =>
0 : <CFString 0x1b1c6d460 [0x1b762cbb8]>{contents = "Version"} = <CFNumber 0xb000000000000022 [0x1b762cbb8]>{value = +2, type = kCFNumberSInt32Type}
2 : <CFString 0x1b1c6d3e0 [0x1b762cbb8]>{contents = "CVBytesPerRow"} = <CFNumber 0xb00000000001e002 [0x1b762cbb8]>{value = +7680, type = kCFNumberSInt32Type}
}
}
}
sbufToTrackReadiness = 0x0
numSamples = 1
sampleTimingArray[1] = {
{PTS = {290309939228910/1000000000 = 290309.939}, DTS = {INVALID}, duration = {INVALID}},
}
imageBuffer = 0x170321180
I have a sample code here如果您想检查问题。该代码准备以4K拍摄视频。如果您的设备无法执行此操作,请将行AVCaptureSessionPreset3840x2160
更改为AVCaptureSessionPresetHigh inside
ProcessadorVideo.m`。示例代码从视频流中裁剪出一个矩形,并对其应用漫画效果。
感谢
答案 0 :(得分:1)
我试过你的样本但无法重现错误-12783。但是我得到了另一个错误-12780。我使用iPhone 6s所以不确定是否是由于那个原因。
无论如何,我可以修复错误-12780。由于旧时间戳,您将面临此问题。我在您的应用程序中添加了一些日志以进行调试。见下文
2017-01-26 15:00:35.809590 NotWriting [16829:3116125]首字母为401051199680537 2017-01-26 15:00:35.809986 NotWriting [16829:3116125]附加时间戳的视频缓冲区401051199680537 2017-01-26 15:00:35.810008 NotWriting [16829:3116125] presentationTimeStamp小于最后一帧时间戳,所以可能会失败 2017-01-26 15:00:35.815605 NotWriting [16829:3116125]错误:发生未知错误(-12780)
当您编写缓冲区时,您必须确保演示文稿时间戳大于您编写的最后一帧,否则写入将失败。您可以使用以下逻辑添加检查:
CMTime presentationTimeStamp = CMSampleBufferGetPresentationTimeStamp(sampleBuffer);
if(CMTIME_COMPARE_INLINE(presentationTimeStamp, <=, lastpresentationTimeStamp))
{
NSLog(@"presentationTimeStamp is less than last frame timestamp, So rejecting frame");
return;
}
lastpresentationTimeStamp = presentationTimeStamp;