Young_Allen
Young_Allen

MediaMuxer源码分析

1.介绍：

MediaMuxer是Android中音视频复用器，可以合成video和audio。android kitkat版本上应用层目前只支持对Mp4文件的合成，TS文件(或TS包)的合成需要修改stagefright源码，本文对MediaMuxer源码作出分析。

2.干货：

MediaMuxer.h文件源码：

/*
 * Copyright 2013, The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#ifndef MEDIA_MUXER_H_
#define MEDIA_MUXER_H_

#include 
#include 
#include 
#include 

namespace android {

struct ABuffer;
struct AMessage;
struct MediaAdapter;
struct MediaBuffer;
struct MediaSource;
struct MetaData;
struct MPEG4Writer;

// MediaMuxer is used to mux multiple tracks into a video. Currently, we only
// support a mp4 file as the output.
// The expected calling order of the functions is:
// Constructor -> addTrack+ -> start -> writeSampleData+ -> stop
// If muxing operation need to be cancelled, the app is responsible for
// deleting the output file after stop.
struct MediaMuxer : public RefBase {
public:
    // Please update media/java/android/media/MediaMuxer.java if the
    // OutputFormat is updated.
    enum OutputFormat {
        OUTPUT_FORMAT_MPEG_4 = 0,
        OUTPUT_FORMAT_LIST_END // must be last - used to validate format type
    };

    // Construct the muxer with the output file path.
    MediaMuxer(const char *path, OutputFormat format);

    // Construct the muxer with the file descriptor. Note that the MediaMuxer
    // will close this file at stop().
    MediaMuxer(int fd, OutputFormat format);

    virtual ~MediaMuxer();

    /**
     * Add a track with its format information. This should be
     * called before start().
     * @param format the track's format.
     * @return the track's index or negative number if error.
     */
    ssize_t addTrack(const sp &format);

    /**
     * Start muxing. Make sure all the tracks have been added before
     * calling this.
     */
    status_t start();

    /**
     * Set the orientation hint.
     * @param degrees The rotation degrees. It has to be either 0,
     *                90, 180 or 270.
     * @return OK if no error.
     */
    status_t setOrientationHint(int degrees);

    /**
     * Set the location.
     * @param latitude The latitude in degree x 1000. Its value must be in the range
     * [-900000, 900000].
     * @param longitude The longitude in degree x 1000. Its value must be in the range
     * [-1800000, 1800000].
     * @return OK if no error.
     */
    status_t setLocation(int latitude, int longitude);

    /**
     * Stop muxing.
     * This method is a blocking call. Depending on how
     * much data is bufferred internally, the time needed for stopping
     * the muxer may be time consuming. UI thread is
     * not recommended for launching this call.
     * @return OK if no error.
     */
    status_t stop();

    /**
     * Send a sample buffer for muxing.
     * The buffer can be reused once this method returns. Typically,
     * this function won't be blocked for very long, and thus there
     * is no need to use a separate thread calling this method to
     * push a buffer.
     * @param buffer the incoming sample buffer.
     * @param trackIndex the buffer's track index number.
     * @param timeUs the buffer's time stamp.
     * @param flags the only supported flag for now is
     *              MediaCodec::BUFFER_FLAG_SYNCFRAME.
     * @return OK if no error.
     */
    status_t writeSampleData(const sp &buffer, size_t trackIndex,
                             int64_t timeUs, uint32_t flags) ;

private:
    sp mWriter;
    Vector< sp > mTrackList;  // Each track has its MediaAdapter.
    sp mFileMeta;  // Metadata for the whole file.

    Mutex mMuxerLock;

    enum State {
        UNINITIALIZED,
        INITIALIZED,
        STARTED,
        STOPPED
    };
    State mState;

    DISALLOW_EVIL_CONSTRUCTORS(MediaMuxer);
};

}  // namespace android

#endif  // MEDIA_MUXER_H_

可以看到MPEG4Writer是继承MediaWriter 的。

/*
 * Copyright (C) 2009 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#ifndef MPEG4_WRITER_H_

#define MPEG4_WRITER_H_

#include 

#include 
#include 
#include 

namespace android {

class MediaBuffer;
class MediaSource;
class MetaData;

class MPEG4Writer : public MediaWriter {
public:
    MPEG4Writer(const char *filename);
    MPEG4Writer(int fd);

    // Limitations
    // 1. No more than 2 tracks can be added
    // 2. Only video or audio source can be added
    // 3. No more than one video and/or one audio source can be added.
    virtual status_t addSource(const sp &source);

    // Returns INVALID_OPERATION if there is no source or track.
    virtual status_t start(MetaData *param = NULL);
    virtual status_t stop() { return reset(); }
    virtual status_t pause();
    virtual bool reachedEOS();
    virtual status_t dump(int fd, const Vector& args);

    void beginBox(const char *fourcc);
    void writeInt8(int8_t x);
    void writeInt16(int16_t x);
    void writeInt32(int32_t x);
    void writeInt64(int64_t x);
    void writeCString(const char *s);
    void writeFourcc(const char *fourcc);
    void write(const void *data, size_t size);
    inline size_t write(const void *ptr, size_t size, size_t nmemb);
    void endBox();
    uint32_t interleaveDuration() const { return mInterleaveDurationUs; }
    status_t setInterleaveDuration(uint32_t duration);
    int32_t getTimeScale() const { return mTimeScale; }

    status_t setGeoData(int latitudex10000, int longitudex10000);
    void setStartTimeOffsetMs(int ms) { mStartTimeOffsetMs = ms; }
    int32_t getStartTimeOffsetMs() const { return mStartTimeOffsetMs; }

protected:
    virtual ~MPEG4Writer();

private:
    class Track;

    int  mFd;
    status_t mInitCheck;
    bool mIsRealTimeRecording;
    bool mUse4ByteNalLength;
    bool mUse32BitOffset;
    bool mIsFileSizeLimitExplicitlyRequested;
    bool mPaused;
    bool mStarted;  // Writer thread + track threads started successfully
    bool mWriterThreadStarted;  // Only writer thread started successfully
    off64_t mOffset;
    off_t mMdatOffset;
    uint8_t *mMoovBoxBuffer;
    off64_t mMoovBoxBufferOffset;
    bool  mWriteMoovBoxToMemory;
    off64_t mFreeBoxOffset;
    bool mStreamableFile;
    off64_t mEstimatedMoovBoxSize;
    uint32_t mInterleaveDurationUs;
    int32_t mTimeScale;
    int64_t mStartTimestampUs;
    int mLatitudex10000;
    int mLongitudex10000;
    bool mAreGeoTagsAvailable;
    int32_t mStartTimeOffsetMs;

    Mutex mLock;

    List mTracks;

    List mBoxes;

    void setStartTimestampUs(int64_t timeUs);
    int64_t getStartTimestampUs();  // Not const
    status_t startTracks(MetaData *params);
    size_t numTracks();
    int64_t estimateMoovBoxSize(int32_t bitRate);

    struct Chunk {
        Track               *mTrack;        // Owner
        int64_t             mTimeStampUs;   // Timestamp of the 1st sample
        List mSamples;       // Sample data

        // Convenient constructor
        Chunk(): mTrack(NULL), mTimeStampUs(0) {}

        Chunk(Track *track, int64_t timeUs, List samples)
            : mTrack(track), mTimeStampUs(timeUs), mSamples(samples) {
        }

    };
    struct ChunkInfo {
        Track               *mTrack;        // Owner
        List         mChunks;        // Remaining chunks to be written

        // Previous chunk timestamp that has been written
        int64_t mPrevChunkTimestampUs;

        // Max time interval between neighboring chunks
        int64_t mMaxInterChunkDurUs;

    };

    bool            mIsFirstChunk;
    volatile bool   mDone;                  // Writer thread is done?
    pthread_t       mThread;                // Thread id for the writer
    List mChunkInfos;            // Chunk infos
    Condition       mChunkReadyCondition;   // Signal that chunks are available

    // Writer thread handling
    status_t startWriterThread();
    void stopWriterThread();
    static void *ThreadWrapper(void *me);
    void threadFunc();

    // Buffer a single chunk to be written out later.
    void bufferChunk(const Chunk& chunk);

    // Write all buffered chunks from all tracks
    void writeAllChunks();

    // Retrieve the proper chunk to write if there is one
    // Return true if a chunk is found; otherwise, return false.
    bool findChunkToWrite(Chunk *chunk);

    // Actually write the given chunk to the file.
    void writeChunkToFile(Chunk* chunk);

    // Adjust other track media clock (presumably wall clock)
    // based on audio track media clock with the drift time.
    int64_t mDriftTimeUs;
    void setDriftTimeUs(int64_t driftTimeUs);
    int64_t getDriftTimeUs();

    // Return whether the nal length is 4 bytes or 2 bytes
    // Only makes sense for H.264/AVC
    bool useNalLengthFour();

    // Return whether the writer is used for real time recording.
    // In real time recording mode, new samples will be allowed to buffered into
    // chunks in higher priority thread, even though the file writer has not
    // drained the chunks yet.
    // By default, real time recording is on.
    bool isRealTimeRecording() const;

    void lock();
    void unlock();

    // Acquire lock before calling these methods
    off64_t addSample_l(MediaBuffer *buffer);
    off64_t addLengthPrefixedSample_l(MediaBuffer *buffer);

    bool exceedsFileSizeLimit();
    bool use32BitFileOffset() const;
    bool exceedsFileDurationLimit();
    bool isFileStreamable() const;
    void trackProgressStatus(size_t trackId, int64_t timeUs, status_t err = OK);
    void writeCompositionMatrix(int32_t degrees);
    void writeMvhdBox(int64_t durationUs);
    void writeMoovBox(int64_t durationUs);
    void writeFtypBox(MetaData *param);
    void writeUdtaBox();
    void writeGeoDataBox();
    void writeLatitude(int degreex10000);
    void writeLongitude(int degreex10000);
    void sendSessionSummary();
    void release();
    status_t reset();

    static uint32_t getMpeg4Time();

    MPEG4Writer(const MPEG4Writer &);
    MPEG4Writer &operator=(const MPEG4Writer &);
};

}  // namespace android

#endif  // MPEG4_WRITER_H_

MediaWriter.h源码：

/*
 * Copyright (C) 2010 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#ifndef MEDIA_WRITER_H_

#define MEDIA_WRITER_H_

#include 
#include 

namespace android {

struct MediaSource;
struct MetaData;

struct MediaWriter : public RefBase {
    MediaWriter()
        : mMaxFileSizeLimitBytes(0),
          mMaxFileDurationLimitUs(0) {
    }

    virtual status_t addSource(const sp &source) = 0;
    virtual bool reachedEOS() = 0;
    virtual status_t start(MetaData *params = NULL) = 0;
    virtual status_t stop() = 0;
    virtual status_t pause() = 0;

    virtual void setMaxFileSize(int64_t bytes) { mMaxFileSizeLimitBytes = bytes; }
    virtual void setMaxFileDuration(int64_t durationUs) { mMaxFileDurationLimitUs = durationUs; }
    virtual void setListener(const sp& listener) {
        mListener = listener;
    }

    virtual status_t dump(int fd, const Vector& args) {
        return OK;
    }

protected:
    virtual ~MediaWriter() {}
    int64_t mMaxFileSizeLimitBytes;
    int64_t mMaxFileDurationLimitUs;
    sp mListener;

    void notify(int msg, int ext1, int ext2) {
        if (mListener != NULL) {
            mListener->notify(msg, ext1, ext2);
        }
    }
private:
    MediaWriter(const MediaWriter &);
    MediaWriter &operator=(const MediaWriter &);
};

}  // namespace android

#endif  // MEDIA_WRITER_H_

回过头再来看MediaMuxer.cpp源码

/*
 * Copyright 2013, The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

//#define LOG_NDEBUG 0
#define LOG_TAG "MediaMuxer"
#include 

#include 

#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 

namespace android {

MediaMuxer::MediaMuxer(const char *path, OutputFormat format)
    : mState(UNINITIALIZED) {
    if (format == OUTPUT_FORMAT_MPEG_4) {
        mWriter = new MPEG4Writer(path);
        mFileMeta = new MetaData;
        mState = INITIALIZED;
    }

}

MediaMuxer::MediaMuxer(int fd, OutputFormat format)
    : mState(UNINITIALIZED) {
    if (format == OUTPUT_FORMAT_MPEG_4) {
        mWriter = new MPEG4Writer(fd);
        mFileMeta = new MetaData;
        mState = INITIALIZED;
    }
}

MediaMuxer::~MediaMuxer() {
    Mutex::Autolock autoLock(mMuxerLock);

    // Clean up all the internal resources.
    mFileMeta.clear();
    mWriter.clear();
    mTrackList.clear();
}

ssize_t MediaMuxer::addTrack(const sp &format) {
    Mutex::Autolock autoLock(mMuxerLock);

    if (format.get() == NULL) {
        ALOGE("addTrack() get a null format");
        return -EINVAL;
    }

    if (mState != INITIALIZED) {
        ALOGE("addTrack() must be called after constructor and before start().");
        return INVALID_OPERATION;
    }

    sp trackMeta = new MetaData;
    convertMessageToMetaData(format, trackMeta);

    sp newTrack = new MediaAdapter(trackMeta);
    status_t result = mWriter->addSource(newTrack);
    if (result == OK) {
        return mTrackList.add(newTrack);
    }
    return -1;
}

status_t MediaMuxer::setOrientationHint(int degrees) {
    Mutex::Autolock autoLock(mMuxerLock);
    if (mState != INITIALIZED) {
        ALOGE("setOrientationHint() must be called before start().");
        return INVALID_OPERATION;
    }

    if (degrees != 0 && degrees != 90 && degrees != 180 && degrees != 270) {
        ALOGE("setOrientationHint() get invalid degrees");
        return -EINVAL;
    }

    mFileMeta->setInt32(kKeyRotation, degrees);
    return OK;
}

status_t MediaMuxer::setLocation(int latitude, int longitude) {
    Mutex::Autolock autoLock(mMuxerLock);
    if (mState != INITIALIZED) {
        ALOGE("setLocation() must be called before start().");
        return INVALID_OPERATION;
    }
    ALOGV("Setting location: latitude = %d, longitude = %d", latitude, longitude);
    return mWriter->setGeoData(latitude, longitude);
}

status_t MediaMuxer::start() {
    Mutex::Autolock autoLock(mMuxerLock);
    if (mState == INITIALIZED) {
        mState = STARTED;
        mFileMeta->setInt32(kKeyRealTimeRecording, false);
        return mWriter->start(mFileMeta.get());
    } else {
        ALOGE("start() is called in invalid state %d", mState);
        return INVALID_OPERATION;
    }
}

status_t MediaMuxer::stop() {
    Mutex::Autolock autoLock(mMuxerLock);

    if (mState == STARTED) {
        mState = STOPPED;
        for (size_t i = 0; i < mTrackList.size(); i++) {
            if (mTrackList[i]->stop() != OK) {
                return INVALID_OPERATION;
            }
        }
        return mWriter->stop();
    } else {
        ALOGE("stop() is called in invalid state %d", mState);
        return INVALID_OPERATION;
    }
}

status_t MediaMuxer::writeSampleData(const sp &buffer, size_t trackIndex,
                                     int64_t timeUs, uint32_t flags) {
    Mutex::Autolock autoLock(mMuxerLock);

    if (buffer.get() == NULL) {
        ALOGE("WriteSampleData() get an NULL buffer.");
        return -EINVAL;
    }

    if (mState != STARTED) {
        ALOGE("WriteSampleData() is called in invalid state %d", mState);
        return INVALID_OPERATION;
    }

    if (trackIndex >= mTrackList.size()) {
        ALOGE("WriteSampleData() get an invalid index %d", trackIndex);
        return -EINVAL;
    }

    MediaBuffer* mediaBuffer = new MediaBuffer(buffer);

    mediaBuffer->add_ref(); // Released in MediaAdapter::signalBufferReturned().
    mediaBuffer->set_range(buffer->offset(), buffer->size());

    sp sampleMetaData = mediaBuffer->meta_data();
    sampleMetaData->setInt64(kKeyTime, timeUs);
    // Just set the kKeyDecodingTime as the presentation time for now.
    sampleMetaData->setInt64(kKeyDecodingTime, timeUs);

    if (flags & MediaCodec::BUFFER_FLAG_SYNCFRAME) {
        sampleMetaData->setInt32(kKeyIsSyncFrame, true);
    }

    sp currentTrack = mTrackList[trackIndex];
    // This pushBuffer will wait until the mediaBuffer is consumed.
    return currentTrack->pushBuffer(mediaBuffer);
}

}  // namespace android

在代码中通过以下代码创建一个合成器：

    const char *outputFileName = "/sdcard/muxeroutput.mp4";
    sp muxer = new MediaMuxer(outputFileName,MediaMuxer::OUTPUT_FORMAT_MPEG_4);

要正常使用Muxer还必须在实例化后、start前添加需要合成的音视频信息也就是需要调用：

MediaMuxer::addTrack(const sp &format)

在实际项目中代码如下：

if (haveVideo) {
    ALOGV("muxer: adding video track.");
    outputVideoTrack = muxer->addTrack(encoderOutputVideoFormat);
}
if (haveAudio) {
    ALOGV("muxer: adding audio track.");
    outputAudioTrack = muxer->addTrack(encoderOutputAudioFormat);
}
ALOGV("muxer: starting");
muxer->start();
muxing = true;

从MediaMuxer.cpp中可以看到addTrack实际上是把相应的音视频信息转成成可配置到MediaWriter中数据，并添加到一个mTrackList数组中.

ssize_t MediaMuxer::addTrack(const sp &format) {
    Mutex::Autolock autoLock(mMuxerLock);

    if (format.get() == NULL) {
        ALOGE("addTrack() get a null format");
        return -EINVAL;
    }

    if (mState != INITIALIZED) {
        ALOGE("addTrack() must be called after constructor and before start().");
        return INVALID_OPERATION;
    }

    sp trackMeta = new MetaData;
    convertMessageToMetaData(format, trackMeta);

    sp newTrack = new MediaAdapter(trackMeta);
    status_t result = mWriter->addSource(newTrack);
    if (result == OK) {
        return mTrackList.add(newTrack);
    }
    return -1;
}

MediaWriter在之前的代码中已经提到过，而MediaAdapter又是什么？

/*
 * Copyright (C) 2013 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#ifndef MEDIA_ADAPTER_H
#define MEDIA_ADAPTER_H

#include 
#include 
#include 
#include 
#include 

namespace android {

// Convert the MediaMuxer's push model into MPEG4Writer's pull model.
// Used only by the MediaMuxer for now.
struct MediaAdapter : public MediaSource, public MediaBufferObserver {
public:
    // MetaData is used to set the format and returned at getFormat.
    MediaAdapter(const sp &meta);
    virtual ~MediaAdapter();
    /////////////////////////////////////////////////
    // Inherited functions from MediaSource
    /////////////////////////////////////////////////

    virtual status_t start(MetaData *params = NULL);
    virtual status_t stop();
    virtual sp getFormat();
    virtual status_t read(
            MediaBuffer **buffer, const ReadOptions *options = NULL);

    /////////////////////////////////////////////////
    // Inherited functions from MediaBufferObserver
    /////////////////////////////////////////////////

    virtual void signalBufferReturned(MediaBuffer *buffer);

    /////////////////////////////////////////////////
    // Non-inherited functions:
    /////////////////////////////////////////////////

    // pushBuffer() will wait for the read() finish, and read() will have a
    // deep copy, such that after pushBuffer return, the buffer can be re-used.
    status_t pushBuffer(MediaBuffer *buffer);

private:
    Mutex mAdapterLock;
    // Make sure the read() wait for the incoming buffer.
    Condition mBufferReadCond;
    // Make sure the pushBuffer() wait for the current buffer consumed.
    Condition mBufferReturnedCond;

    MediaBuffer *mCurrentMediaBuffer;

    bool mStarted;
    sp mOutputFormat;

    DISALLOW_EVIL_CONSTRUCTORS(MediaAdapter);
};

}  // namespace android

#endif  // MEDIA_ADAPTER_H

继续看MediaSource：

/*
 * Copyright (C) 2009 The Android Open Source Project
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#ifndef MEDIA_SOURCE_H_

#define MEDIA_SOURCE_H_

#include 

#include 
#include 
#include 

namespace android {

class MediaBuffer;
class MetaData;

struct MediaSource : public virtual RefBase {
    MediaSource();

    // To be called before any other methods on this object, except
    // getFormat().
    virtual status_t start(MetaData *params = NULL) = 0;

    // Any blocking read call returns immediately with a result of NO_INIT.
    // It is an error to call any methods other than start after this call
    // returns. Any buffers the object may be holding onto at the time of
    // the stop() call are released.
    // Also, it is imperative that any buffers output by this object and
    // held onto by callers be released before a call to stop() !!!
    virtual status_t stop() = 0;

    // Returns the format of the data output by this media source.
    virtual sp getFormat() = 0;

    struct ReadOptions;

    // Returns a new buffer of data. Call blocks until a
    // buffer is available, an error is encountered of the end of the stream
    // is reached.
    // End of stream is signalled by a result of ERROR_END_OF_STREAM.
    // A result of INFO_FORMAT_CHANGED indicates that the format of this
    // MediaSource has changed mid-stream, the client can continue reading
    // but should be prepared for buffers of the new configuration.
    virtual status_t read(
            MediaBuffer **buffer, const ReadOptions *options = NULL) = 0;

    // Options that modify read() behaviour. The default is to
    // a) not request a seek
    // b) not be late, i.e. lateness_us = 0
    struct ReadOptions {
        enum SeekMode {
            SEEK_PREVIOUS_SYNC,
            SEEK_NEXT_SYNC,
            SEEK_CLOSEST_SYNC,
            SEEK_CLOSEST,
        };

        ReadOptions();

        // Reset everything back to defaults.
        void reset();

        void setSeekTo(int64_t time_us, SeekMode mode = SEEK_CLOSEST_SYNC);
        void clearSeekTo();
        bool getSeekTo(int64_t *time_us, SeekMode *mode) const;

        void setLateBy(int64_t lateness_us);
        int64_t getLateBy() const;

    private:
        enum Options {
            kSeekTo_Option      = 1,
        };

        uint32_t mOptions;
        int64_t mSeekTimeUs;
        SeekMode mSeekMode;
        int64_t mLatenessUs;
    };

    // Causes this source to suspend pulling data from its upstream source
    // until a subsequent read-with-seek. Currently only supported by
    // OMXCodec.
    virtual status_t pause() {
        return ERROR_UNSUPPORTED;
    }

    // The consumer of this media source requests that the given buffers
    // are to be returned exclusively in response to read calls.
    // This will be called after a successful start() and before the
    // first read() call.
    // Callee assumes ownership of the buffers if no error is returned.
    virtual status_t setBuffers(const Vector &buffers) {
        return ERROR_UNSUPPORTED;
    }

protected:
    virtual ~MediaSource();

private:
    MediaSource(const MediaSource &);
    MediaSource &operator=(const MediaSource &);
};

}  // namespace android

#endif  // MEDIA_SOURCE_H_

看到这里就明白了MediaSource就是一个RefBase指针，MediaAdapter继承于MediaSource，那么在MediaMuxer::addTrack时，调用到MPEG4Writer::addSource：

status_t MPEG4Writer::addSource(const sp &source) {
    Mutex::Autolock l(mLock);
    if (mStarted) {
        ALOGE("Attempt to add source AFTER recording is started");
        return UNKNOWN_ERROR;
    }

    // At most 2 tracks can be supported.
    if (mTracks.size() >= 2) {
        ALOGE("Too many tracks (%d) to add", mTracks.size());
        return ERROR_UNSUPPORTED;
    }

    CHECK(source.get() != NULL);

    // A track of type other than video or audio is not supported.
    const char *mime;
    source->getFormat()->findCString(kKeyMIMEType, &mime);
    bool isAudio = !strncasecmp(mime, "audio/", 6);
    bool isVideo = !strncasecmp(mime, "video/", 6);
    if (!isAudio && !isVideo) {
        ALOGE("Track (%s) other than video or audio is not supported",
            mime);
        return ERROR_UNSUPPORTED;
    }

    // At this point, we know the track to be added is either
    // video or audio. Thus, we only need to check whether it
    // is an audio track or not (if it is not, then it must be
    // a video track).

    // No more than one video or one audio track is supported.
    for (List::iterator it = mTracks.begin();
         it != mTracks.end(); ++it) {
        if ((*it)->isAudio() == isAudio) {
            ALOGE("%s track already exists", isAudio? "Audio": "Video");
            return ERROR_UNSUPPORTED;
        }
    }

    // This is the first track of either audio or video.
    // Go ahead to add the track.
    Track *track = new Track(this, source, 1 + mTracks.size());
    mTracks.push_back(track);

    return OK;
}

MediaMuxer在配置好音视频信息后的start处理：

status_t MediaMuxer::start() {
    Mutex::Autolock autoLock(mMuxerLock);
    if (mState == INITIALIZED) {
        mState = STARTED;
        mFileMeta->setInt32(kKeyRealTimeRecording, false);
        return mWriter->start(mFileMeta.get());
    } else {
        ALOGE("start() is called in invalid state %d", mState);
        return INVALID_OPERATION;
    }
}

status_t MPEG4Writer::start(MetaData *param) {
    if (mInitCheck != OK) {
        return UNKNOWN_ERROR;
    }

    if (mMaxFileSizeLimitBytes != 0) {
        mIsFileSizeLimitExplicitlyRequested = true;
    }

    int32_t use64BitOffset;
    if (param &&
        param->findInt32(kKey64BitFileOffset, &use64BitOffset) &&
        use64BitOffset) {
        mUse32BitOffset = false;
    }

    if (mUse32BitOffset) {
        // Implicit 32 bit file size limit
        if (mMaxFileSizeLimitBytes == 0) {
            mMaxFileSizeLimitBytes = kMax32BitFileSize;
        }

        // If file size is set to be larger than the 32 bit file
        // size limit, treat it as an error.
        if (mMaxFileSizeLimitBytes > kMax32BitFileSize) {
            ALOGW("32-bit file size limit (%lld bytes) too big. "
                 "It is changed to %lld bytes",
                mMaxFileSizeLimitBytes, kMax32BitFileSize);
            mMaxFileSizeLimitBytes = kMax32BitFileSize;
        }
    }

    int32_t use2ByteNalLength;
    if (param &&
        param->findInt32(kKey2ByteNalLength, &use2ByteNalLength) &&
        use2ByteNalLength) {
        mUse4ByteNalLength = false;
    }

    int32_t isRealTimeRecording;
    if (param && param->findInt32(kKeyRealTimeRecording, &isRealTimeRecording)) {
        mIsRealTimeRecording = isRealTimeRecording;
    }

    mStartTimestampUs = -1;

    if (mStarted) {
        if (mPaused) {
            mPaused = false;
            return startTracks(param);
        }
        return OK;
    }

    if (!param ||
        !param->findInt32(kKeyTimeScale, &mTimeScale)) {
        mTimeScale = 1000;
    }
    CHECK_GT(mTimeScale, 0);
    ALOGV("movie time scale: %d", mTimeScale);

    /*
     * When the requested file size limit is small, the priority
     * is to meet the file size limit requirement, rather than
     * to make the file streamable. mStreamableFile does not tell
     * whether the actual recorded file is streamable or not.
     */
    mStreamableFile =
        (mMaxFileSizeLimitBytes != 0 &&
         mMaxFileSizeLimitBytes >= kMinStreamableFileSizeInBytes);

   
    mWriteMoovBoxToMemory = false;
    mMoovBoxBuffer = NULL;
    mMoovBoxBufferOffset = 0;

    writeFtypBox(param);

    mFreeBoxOffset = mOffset;

    if (mEstimatedMoovBoxSize == 0) {
        int32_t bitRate = -1;
        if (param) {
            param->findInt32(kKeyBitRate, &bitRate);
        }
        mEstimatedMoovBoxSize = estimateMoovBoxSize(bitRate);
    }
    CHECK_GE(mEstimatedMoovBoxSize, 8);
    if (mStreamableFile) {
        // Reserve a 'free' box only for streamable file
        lseek64(mFd, mFreeBoxOffset, SEEK_SET);
        writeInt32(mEstimatedMoovBoxSize);
        write("free", 4);
        mMdatOffset = mFreeBoxOffset + mEstimatedMoovBoxSize;
    } else {
        mMdatOffset = mOffset;
    }

    mOffset = mMdatOffset;
    lseek64(mFd, mMdatOffset, SEEK_SET);
    if (mUse32BitOffset) {
        write("????mdat", 8);
    } else {
        write("\x00\x00\x00\x01mdat????????", 16);
    }

    status_t err = startWriterThread();
    if (err != OK) {
        return err;
    }

    err = startTracks(param);
    if (err != OK) {
        return err;
    }

    mStarted = true;
    return OK;
}

status_t MPEG4Writer::startWriterThread() {
    ALOGV("startWriterThread");

    mDone = false;
    mIsFirstChunk = true;
    mDriftTimeUs = 0;
    for (List::iterator it = mTracks.begin();
         it != mTracks.end(); ++it) {
        ChunkInfo info;
        info.mTrack = *it;
        info.mPrevChunkTimestampUs = 0;
        info.mMaxInterChunkDurUs = 0;
        mChunkInfos.push_back(info);
    }

    pthread_attr_t attr;
    pthread_attr_init(&attr);
    pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_JOINABLE);
    pthread_create(&mThread, &attr, ThreadWrapper, this);
    pthread_attr_destroy(&attr);
    mWriterThreadStarted = true;
    return OK;
}

在start过程中，开启了write的线程。
在项目中，我是这样写入音视频数据：

if(trackIndex == videoTrack){
  if(presentationTimeUs >= mLastVideoSampleTime){
    useOriTime = true;
  }
  if (size >= 0 && outputVideoTrack != -1) {
    if(useOriTime){
      mLastVideoSampleTime = presentationTimeUs;
      err = muxer->writeSampleData(buffer,outputVideoTrack,mLastVideoSampleTime, flags);
      CHECK_EQ(err, (status_t)OK);
    }else{
      ALOGV("%s encoder loss one buffer.",trackIndex==videoTrack?"video":"audio");
    }
                                
  }
}else if(trackIndex == audioTrack){
  if(presentationTimeUs >= mLastAudioSampleTime){
    useOriTime = true;
  }
  if (size >= 0 && outputAudioTrack != -1) {
    if(useOriTime){
      mLastAudioSampleTime = presentationTimeUs;
      err = muxer->writeSampleData(buffer,outputAudioTrack,mLastAudioSampleTime, flags);
      CHECK_EQ(err, (status_t)OK);
    }else{
      ALOGV("%s encoder loss one   buffer.",trackIndex==videoTrack?"video":"audio");
    }
  }
}

通过muxer->writeSampleData写入音视频ES。

3.结束语

Android kitkat版本在应用层只支持对Mp4文件的合成，要支持TS等其他封装格式，就可以在libstagefright的MediaMuxer源码模块作处理，通过上面代码的分析基本已经可以只需要改几行代码就可以支持TS的封装了。本文分析到这里就结束了，感谢持续关注！

你可能感兴趣的:(MediaMuxer源码分析)

  • JVM源码分析之堆外内存完全解读 HeapDump性能社区
    概述广义的堆外内存说到堆外内存，那大家肯定想到堆内内存，这也是我们大家接触最多的，我们在jvm参数里通常设置-Xmx来指定我们的堆的最大值，不过这还不是我们理解的Java堆，-Xmx的值是新生代和老生代的和的最大值，我们在jvm参数里通常还会加一个参数-XX:MaxPermSize来指定持久代的最大值，那么我们认识的Java堆的最大值其实是-Xmx和-XX:MaxPermSize的总和，在分代算法
  • Golang Channel PandaSkr golang
    Channel解析1.Channel源码分析1.1Channel数据结构typehchanstruct{qcountuint//channel的元素数量dataqsizuint//channel循环队列长度bufunsafe.Pointer//指向循环队列的指针elemsizeuint16//元素大小closeduint32//channel是否关闭0-未关闭elemtype*_type//元素类
  • Java集合类框架源码分析 之 RoleList源码解析 【6】 yunzhonghefei Java集合类源码分析RoleList源码解析
    该类继承于ArrayList，针对Role进行了一些扩展。其他方法和ArrayList中基本相同，源码不做针对性分析：看一下类简介：/***代表了一个roles的列表，作为方法setRoles()的参数，去创建一个关联关系，并且尝试在同一个关系中设置多个角色。*ARoleListrepresentsalistofroles(Roleobjects).Itisusedas*parameterwhen
  • 【鸿蒙OH-v5.0源码分析之 Linux Kernel 部分】004 - Kernel 启动引导代码head.S 源码逐行分析 "小夜猫&小懒虫&小财迷"的男人 鸿蒙OH-v5.0源码分析之Uboot+Kernel部分harmonyoslinux华为
    【鸿蒙OH-v5.0源码分析之LinuxKernel部分】004-Kernel启动引导代码head.S源码逐行分析系列文章汇总：《鸿蒙OH-v5.0源码分析之Uboot+Kernel部分】000-文章链接汇总》本文链接：《【鸿蒙OH-v5.0源码分析之LinuxKernel部分】004-Kernel启动引导代码head.S源码逐行分析》head.S主要工作如下：保存内核启动参数,无效化处理器缓存(
  • Java高并发编程详解系列-深入理解Thread构造 nihui123 高并发Java高并发Java高并发
    上篇分享中主要是对线程的基本概念和基本操作做了一个分享，同时提出了两种常用的创建多线程的方法，当然在后期的分享中也会提及到更多的创建线程的方式，到后期的分享的时候再说。这次主要是深入的理解一下Thread的构造函数，通过构造函数对于Thread有一个更加深入的了解。这里首先提供一个JDK1.6的ThreadAPI截图线程命名规范  从源码分析可以看到在Thread类中默认提供了线程的命名方式，这个
  • 【QT教程】QT6硬件图形界面编程 QT硬件编程 QT性能优化QT原理源码QT界面美化 qtqt6.3qt5c++QT教程
    QT6硬件图形界面编程使用AI技术辅助生成QT界面美化视频课程QT性能优化视频课程QT原理与源码分析视频课程QTQMLC++扩展开发视频课程免费QT视频课程您可以看免费1000+个QT技术视频免费QT视频课程QT统计图和QT数据可视化视频免费看免费QT视频课程QT性能优化视频免费看免费QT视频课程QT界面美化视频免费看1QT6硬件图形界面编程概述1.1QT6硬件图形界面编程简介1.1.1QT6硬件
  • Jetpack LiveData源码分析 xiangxiongfly915 #AndroidJetpack系列JetpackLiveData源码分析
    文章目录JetpackLiveData源码分析前提源码分析注册订阅流程LiveData#observe()LifecycleRegistry#addObserver()ObserverWithState#dispatchEvent()LiveData#removeObserver()LiveData$LifecycleBoundObserver类LifecycleBoundObserver#sho
  • ExoPlayer架构详解与源码分析（17）——TrackSelector 山雨楼 ExoPlayer架构android音视频ExoPlayerMedia3源码
    系列文章目录ExoPlayer架构详解与源码分析（1）——前言ExoPlayer架构详解与源码分析（2）——PlayerExoPlayer架构详解与源码分析（3）——TimelineExoPlayer架构详解与源码分析（4）——整体架构ExoPlayer架构详解与源码分析（5）——MediaSourceExoPlayer架构详解与源码分析（6）——MediaPeriodExoPlayer架构详解与
  • ExoPlayer架构详解与源码分析（12）——Cache 山雨楼 ExoPlayer架构android音视频ExoPlayerMedia3源码
    系列文章目录ExoPlayer架构详解与源码分析（1）——前言ExoPlayer架构详解与源码分析（2）——PlayerExoPlayer架构详解与源码分析（3）——TimelineExoPlayer架构详解与源码分析（4）——整体架构ExoPlayer架构详解与源码分析（5）——MediaSourceExoPlayer架构详解与源码分析（6）——MediaPeriodExoPlayer架构详解与
  • ExoPlayer架构详解与源码分析（7）——SampleQueue(4) 2401_83740189 程序员架构
    long[]newSourceIds=newlong[newCapacity];long[]newOffsets=newlong[newCapacity];long[]newTimesUs=newlong[newCapacity];int[]newFlags=newint[newCapacity];int[]newSizes=newint[newCapacity];CryptoData[]newC
  • 【HarmonyOS】- 常见算法简单写法 数的羊都睡了 HarmonyOSArkTS鸿蒙
    文章目录知识回顾前言源码分析1.冒泡排序2.二分法查找拓展知识时间、空间复杂度总结知识回顾前言常见算法简单写法源码分析1.冒泡排序functionbubbleSort(arr:number[]):number[]{constn=arr.length;for(leti=0;iarr[j+1]){//交换元素consttemp=arr[j];arr[j]=arr[j+1];arr[j+1]=temp;
  • [ docker-ce源码分析系列 ] 修改resolv.conf文件被还原的原因 nangonghen dockerdocker
    1概述：1.1环境版本信息如下：a、操作系统：centos7.6，amd64b、服务器docker版本：v18.09.22resolv.conf文件被还原的现象：容器中的/etc/resolv.conf文件，是由宿主机/var/lib/docker/containers/xxxx/resolv.conf文件挂载。在dockerrestart容器之前，手动修改了/var/lib/docker/con
  • dispatch_once源码分析 福伟_Y
    GCD里的单例函数dispatch_once是我们经常会用到的，今天我们来稍做深入分析一下。GCD的源码都在libdispatch.dylib库里，这个库在libSystem_initializer被初始化，可理解为在dyld里被加载和初始化的(之前的文章有分析过)。dispatch_once作为单例的使用入口，通过分析得到它是一个宏定义，_dispatch_once函数在libdispatch.
  • jQuery Easyui 源码分析之combo组件 90后北京程序员 前端-easyuieasyui之combobox
    /***jQueryEasyUI1.3.1*该源码完全由压缩码翻译而来，并非网络上放出的源码，请勿索要。*/(function($){functionsetSize(target,width){varopts=$.data(target,"combo").options;varcombo=$.data(target,"combo").combo;varpanel=$.data(target,"co
  • html5carousel图片轮播,全面解析Bootstrap中Carousel轮播的使用方法 RemusrickCat
    本文实例为大家全面的解析了Bootstrap中Carousel的使用方法，供大家参考，具体内容如下源码文件：Carousel.scssCarousel.js实现原理：隐藏所有要显示的元素，然后指定当前要显示的为block，宽、高自适应源码分析：1、Html结构：主要分为以四个部分1.1、容器：最外层div，需要一个data-ride=”carousel”来指定为轮播放插件，并且提供一个Id，方便圆
  • Java集合-----List接口及其实现类：ArrayList、LinkedList、Vector Colourful． Java集合java集合
    文章目录List接口概述List接口的常用方法List接口的实现类ArrayList源码分析类继承关系ArrayList中的属性：ArrayList构造函数：ArrayList中常用方法添加操作：add()删除操作：remove()获取元素：get()ArrayList是如何实现序列化的？ArrayList的总结LinkedList源码分析类继承关系类成员属性类构造器LinkedList的List
  • React Native通讯原理 zbl_zbl androidReactNativ
    之前写过一篇文章ReactNativeAndroid源码分析，在此文章的基础上分析和总结下RN与Native的通讯流程。本文基于Android代码分析，iOS实现原理类似。1.通讯框架图先来解析下各个模块的角色与作用：Java层，这块的实现在ReactAndroid中-ReactContext:Android上下文子类，包含一个CatalystInstance实例，用于获取NativeModule
  • [linux 驱动]增加一个文件节点控制led灯亮灭 嵌入式成长家 linux内核的系统实战linux驱动linux驱动led灯驱动
    目录1修改设备树2修改驱动3驱动源码3.1驱动源码3.2设备树节点3.3驱动源码分析3.3.1##解释3.3.2class_create解释3.3.3class_create_file解释3.3.4of_get_named_gpio_flags解释3.3.5devm_gpio_request解释3.3.6platform_driver_register解释3.3.7platform_driver_
  • 【QT教程】QT6对话框编程 QT性能优化QT原理源码QT界面美化 qtqt6.3qt5c++QT教程
    QT6对话框编程使用AI技术辅助生成QT界面美化视频课程QT性能优化视频课程QT原理与源码分析视频课程QTQMLC++扩展开发视频课程免费QT视频课程您可以看免费1000+个QT技术视频免费QT视频课程QT统计图和QT数据可视化视频免费看免费QT视频课程QT性能优化视频免费看免费QT视频课程QT界面美化视频免费看1QT6对话框编程基础1.1QT6对话框简介1.1.1QT6对话框简介QT6对话框简介
  • 鸿蒙轻内核M核源码分析系列十二 事件Event OpenHarmony_小贾 OpenHarmonyHarmonyOS鸿蒙开发harmonyosopenharmony鸿蒙内核鸿蒙开发移动开发嵌入式硬件驱动开发
    事件（Event）是一种任务间通信的机制，可用于任务间的同步。多任务环境下，任务之间往往需要同步操作，一个等待即是一个同步。事件可以提供一对多、多对多的同步操作。本文通过分析鸿蒙轻内核事件模块的源码，深入掌握事件的使用。本文中所涉及的源码，以OpenHarmonyLiteOS-M内核为例，均可以在开源站点https://gitee.com/openharmony/kernel_liteos_m获取
  • 鸿蒙轻内核M核源码分析系列四 中断Hwi OpenHarmony_小贾 鸿蒙开发OpenHarmonyHarmonyOSharmonyos单片机OpenHarmony嵌入式硬件鸿蒙开发移动开发鸿蒙内核
    在鸿蒙轻内核源码分析系列前几篇文章中，剖析了重要的数据结构。本文，我们讲述一下中断，会给读者介绍中断的概念，鸿蒙轻内核的中断模块的源代码。本文中所涉及的源码，以OpenHarmonyLiteOS-M内核为例。1、中断概念介绍中断是指出现需要时，CPU暂停执行当前程序，转而执行新程序的过程。当外设需要CPU时，将通过产生中断信号使CPU立即中断当前任务来响应中断请求。在剖析中断源代码之前，下面介绍些
  • 鸿蒙轻内核M核源码分析系列五 时间管理 OpenHarmony_小贾 HarmonyOSOpenHarmony鸿蒙开发harmonyosopenharmony鸿蒙开发NAPI鸿蒙内核移动开发嵌入式
    在鸿蒙轻内核源码分析上一篇文章中，我们剖析了中断的源码，简单提到了Tick中断。本文会继续分析Tick和时间相关的源码，给读者介绍鸿蒙轻内核的时间管理模块。本文中所涉及的源码，以OpenHarmonyLiteOS-M内核为例，均可以在开源站点https://gitee.com/openharmony/kernel_liteos_m获取。时间管理模块以系统时钟为基础，可以分为2部分，一部分是SysT
  • 鸿蒙轻内核M核源码分析系列六 任务及任务调度（1）任务栈 OpenHarmony_小贾 鸿蒙开发OpenHarmonyHarmonyOSHarmonyOSopenharmony鸿蒙开发移动开发鸿蒙内核驱动开发嵌入式硬件
    继续分析鸿蒙轻内核源码，我们本文开始要分析下任务及任务调度模块。首先，我们介绍下任务栈的基础概念。任务栈是高地址向低地址生长的递减栈，栈指针指向即将入栈的元素位置。初始化后未使用过的栈空间初始化的内容为宏OS_TASK_STACK_INIT代表的数值0xCACACACA，栈顶初始化为宏OS_TASK_MAGIC_WORD代表的数值0xCCCCCCCC。一个任务栈的示意图如下，其中，栈底指针是栈的最
  • 鸿蒙轻内核M核源码分析系列三 数据结构-任务排序链表 OpenHarmony_小贾 HarmonyOSOpenHarmony鸿蒙开发数据结构harmonyos移动开发OpenHarmony鸿蒙内核鸿蒙开发嵌入式硬件
    在鸿蒙轻内核源码分析系列一和系列二，我们分析了双向循环链表、优先级就绪队列的源码。本文会继续给读者介绍鸿蒙轻内核源码中重要的数据结构：任务排序链表TaskSortLinkAttr。鸿蒙轻内核的任务排序链表，用于任务延迟到期/超时唤醒等业务场景，是一个非常重要、非常基础的数据结构。本文中所涉及的源码，以OpenHarmonyLiteOS-M内核为例。1任务排序链表我们先看下任务排序链接的数据结构。任
  • 鸿蒙轻内核A核源码分析系列七 进程管理 (2) OpenHarmony_小贾 HarmonyOS鸿蒙开发OpenHarmonyharmonyosOpenHarmony移动开发驱动开发鸿蒙内核LiteOS-A内核进程通信
    本文先熟悉下进程管理的文件kernel\base\core\los_process.c中的内部接口，读读代码，做些记录。1、LiteOS-A内核进程全局变量⑴是进程池，存放各个进程控制块LosProcessCB的信息。⑵处开始的g_freeProcess是空闲进程链表，挂载各个空闲进程控制块；g_processRecycleList是待回收进程控制块链表，挂载各个等待回收的进程控制块。⑶处开始的g
  • 鸿蒙原生开发——轻内核A核源码分析系列三 物理内存（2） OpenHarmony_小贾 鸿蒙开发HarmonyOSOpenHarmonyharmonyosopenharmony移动开发程序人生鸿蒙开发
    3.1.2.3函数OsVmPhysLargeAlloc当执行到这个函数时，说明空闲链表上的单个内存页节点的大小已经不能满足要求，超过了第9个链表上的内存页节点的大小了。⑴处计算需要申请的内存大小。⑵从最大的链表上进行遍历每一个内存页节点。⑶根据每个内存页的开始内存地址，计算需要的内存的结束地址，如果超过内存段的大小，则继续遍历下一个内存页节点。⑷处此时paStart表示当前内存页的结束地址，接下来
  • JsonCpp源码分析——Reader 哎呦，帅小伙哦 #jsoncppjson
    1、与Writer模块功能相反，可以将Reader理解成一个反序列化的工具，Writer的作用主要是将Value对象转成string或者流式的结构，Reader的作用主要是将流式的结构转成Value类型的对象。Reader类的主要职责有3个，解析JSON字符串：将JSON格式的字符串读取并解析成相应的C++数据结构。处理不同的数据类型，支持解析JSON对象、数组、字符串、数字、布尔值和null。处
  • vue源码分析-挂载流程和模板编译 yyzzabc123 vue.js
    前面几节我们从newVue创建实例开始，介绍了创建实例时执行初始化流程中的重要两步，配置选项的资源合并,以及响应式系统的核心思想，数据代理。在合并章节，我们对Vue丰富的选项合并策略有了基本的认知，在数据代理章节我们又对代理拦截的意义和使用场景有了深入的认识。按照Vue源码的设计思路，初始化过程还会进行很多操作，例如组件之间创建关联，初始化事件中心，初始化数据并建立响应式系统等，并最终将模板和数据
  • 鸿蒙内核解析,鸿蒙内核源码分析(内存概念篇)|解读鸿蒙源码 刘轩鸿 鸿蒙内核解析
    提示：本文基于开源鸿蒙内核分析，官方源码【kernel_liteos_a】官方文档【docs】参考文档【HuaweiLiteOS】本文作者：鸿蒙内核发烧友，用生活场景讲故事的方式去解构内核，一窥究竟，让神秘的内核栩栩如生，浮现眼前。博文全部原创，持续更新，敬请关注。内容仅代表个人观点，错误之处，欢迎大家指正完善。本系列全部文章进入鸿蒙源码分析(总目录)查看目录最难讲的章节坦白讲内存是整个系列里面最
  • 鸿蒙轻内核A核源码分析系列七 进程管理 (3) OpenHarmony_小贾 OpenHarmony鸿蒙开发HarmonyOSharmonyos嵌入式硬件OpenHarmony鸿蒙嵌入式鸿蒙开发鸿蒙内核进程关联
    本文记录下进程相关的初始化函数，如OsSystemProcessCreate、OsProcessInit、OsProcessCreateInit、OsUserInitProcess、OsDeInitPCB、OsUserInitProcessStart等。1、LiteOS-A内核进程创建初始化通用函数先看看一些内部函数，不管是初始化用户态进程还是内核态进程，都会使用这些函数，包含进程控制块初始化函数
  • Dom 周华华 JavaScripthtml
    <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml&q
  • 【Spark九十六】RDD API之combineByKey bit1129 spark
    1. combineByKey函数的运行机制   RDD提供了很多针对元素类型为(K,V)的API，这些API封装在PairRDDFunctions类中，通过Scala隐式转换使用。这些API实现上是借助于combineByKey实现的。combineByKey函数本身也是RDD开放给Spark开发人员使用的API之一   首先看一下combineByKey的方法说明：
  • msyql设置密码报错：ERROR 1372 (HY000): 解决方法详解 daizj mysql设置密码
    MySql给用户设置权限同时指定访问密码时，会提示如下错误： ERROR 1372 (HY000): Password hash should be a 41-digit hexadecimal number；   问题原因：你输入的密码是明文。不允许这么输入。   解决办法：用select password('你想输入的密码');查询出你的密码对应的字符串， 然后
  • 路漫漫其修远兮 吾将上下而求索 周凡杨 学习 思索
    王国维在他的《人间词话》中曾经概括了为学的三种境界古今之成大事业、大学问者，罔不经过三种之境界。“昨夜西风凋碧树。独上高楼，望尽天涯路。”此第一境界也。“衣带渐宽终不悔，为伊消得人憔悴。”此第二境界也。“众里寻他千百度，蓦然回首，那人却在灯火阑珊处。”此第三境界也。学习技术，这也是你必须经历的三种境界。第一层境界是说，学习的路是漫漫的，你必须做好充分的思想准备，如果半途而废还不如不要开始。这里，注
  • Hadoop(二)对话单的操作 朱辉辉33 hadoop
    Debug： 1、 A = LOAD '/user/hue/task.txt' USING PigStorage(' ') AS (col1,col2,col3); DUMP A; //输出结果前几行示例： (>ggsnPDPRecord(21),,) (-->recordType(0),,) (-->networkInitiation(1),,)
  • web报表工具FineReport常用函数的用法总结（日期和时间函数） 老A不折腾 finereport报表工具web开发
    web报表工具FineReport常用函数的用法总结（日期和时间函数）   说明：凡函数中以日期作为参数因子的，其中日期的形式都必须是yy/mm/dd。而且必须用英文环境下双引号(" ")引用。   DATE DATE(year,month,day):返回一个表示某一特定日期的系列数。 Year:代表年，可为一到四位数。 Month:代表月份。
  • c++ 宏定义中的##操作符 墙头上一根草 C++
    #与##在宏定义中的--宏展开 #include <stdio.h> #define f(a,b) a##b #define g(a)   #a #define h(a) g(a) int main() {       &nbs
  • 分析Spring源代码之，DI的实现 aijuans springDI现源代码
    (转) 分析Spring源代码之，DI的实现 2012/1/3 by tony                 接着上次的讲，以下这个sample [java]  view plain copy print
  • for循环的进化 alxw4616 JavaScript
    // for循环的进化 // 菜鸟 for (var i = 0; i < Things.length ; i++) { // Things[i] } // 老鸟 for (var i = 0, len = Things.length; i < len; i++) { // Things[i] } // 大师 for (var i = Things.le
  • 网络编程Socket和ServerSocket简单的使用 百合不是茶 网络编程基础IP地址端口
      网络编程;TCP/IP协议   网络:实现计算机之间的信息共享,数据资源的交换   协议:数据交换需要遵守的一种协议,按照约定的数据格式等写出去   端口:用于计算机之间的通信      每运行一个程序，系统会分配一个编号给该程序，作为和外界交换数据的唯一标识 0~65535   查看被使用的
  • JDK1.5 生产消费者 bijian1013 javathread生产消费者java多线程
    ArrayBlockingQueue：        一个由数组支持的有界阻塞队列。此队列按 FIFO（先进先出）原则对元素进行排序。队列的头部 是在队列中存在时间最长的元素。队列的尾部 是在队列中存在时间最短的元素。新元素插入到队列的尾部，队列检索操作则是从队列头部开始获得元素。 ArrayBlockingQueue的常用方法：
  • JAVA版身份证获取性别、出生日期及年龄 bijian1013 java性别出生日期年龄
            工作中需要根据身份证获取性别、出生日期及年龄，且要还要支持15位长度的身份证号码，网上搜索了一下，经过测试好像多少存在点问题，干脆自已写一个。 CertificateNo.java package com.bijian.study; import java.util.Calendar; import
  • 【Java范型六】范型与枚举 bit1129 java
    首先，枚举类型的定义不能带有类型参数，所以，不能把枚举类型定义为范型枚举类，例如下面的枚举类定义是有编译错的   public enum EnumGenerics<T> { //编译错，提示枚举不能带有范型参数 OK, ERROR; public <T> T get(T type) { return null;
  • 【Nginx五】Nginx常用日志格式含义 bit1129 nginx
    1. log_format 1.1 log_format指令用于指定日志的格式，格式：   log_format name(格式名称) type(格式样式)   1.2 如下是一个常用的Nginx日志格式：   log_format main '[$time_local]|$request_time|$status|$body_bytes
  • Lua 语言 15 分钟快速入门 ronin47 lua 基础
    - - 单行注释 - - [[      [多行注释] - - ]]   - - - - - - - - - - - 1. 变量 & 控制流 - - - - - - - - - - num = 23 - - 数字都是双精度 str = 'aspythonstring'
  • java-35.求一个矩阵中最大的二维矩阵 ( 元素和最大 ) bylijinnan java
    the idea is from: http://blog.csdn.net/zhanxinhang/article/details/6731134 public class MaxSubMatrix { /**see http://blog.csdn.net/zhanxinhang/article/details/6731134 * Q35 求一个矩阵中最大的二维
  • mongoDB文档型数据库特点 开窍的石头 mongoDB文档型数据库特点
    MongoDD: 文档型数据库存储的是Bson文档-->json的二进制 特点：内部是执行引擎是js解释器，把文档转成Bson结构，在查询时转换成js对象。 mongoDB传统型数据库对比    传统类型数据库：结构化数据，定好了表结构后每一个内容符合表结构的。也就是说每一行每一列的数据都是一样的    文档型数据库：不用定好数据结构，
  • [毕业季节]欢迎广大毕业生加入JAVA程序员的行列 comsci java
        一年一度的毕业季来临了。。。。。。。。      正在投简历的学弟学妹们。。。如果觉得学校推荐的单位和公司不适合自己的兴趣和专业，可以考虑来我们软件行业，做一名职业程序员。。。      软件行业的开发工具中，对初学者最友好的就是JAVA语言了，网络上不仅仅有大量的
  • PHP操作Excel – PHPExcel 基本用法详解 cuiyadll PHPExcel
    导出excel属性设置//Include classrequire_once('Classes/PHPExcel.php');require_once('Classes/PHPExcel/Writer/Excel2007.php');$objPHPExcel = new PHPExcel();//Set properties 设置文件属性$objPHPExcel->getProperties
  • IBM Webshpere MQ Client User Issue (MCAUSER) darrenzhu IBMjmsuserMQMCAUSER
    IBM MQ JMS Client去连接远端MQ Server的时候，需要提供User和Password吗？ 答案是根据情况而定，取决于所定义的Channel里面的属性Message channel agent user identifier (MCAUSER)的设置。 http://stackoverflow.com/questions/20209429/how-mca-user-i
  • 网线的接法 dcj3sjt126com
    一、PC连HUB (直连线)A端：（标准568B）：白橙，橙，白绿，蓝，白蓝，绿，白棕，棕。 B端：（标准568B）：白橙，橙，白绿，蓝，白蓝，绿，白棕，棕。 二、PC连PC （交叉线）A端：(568A)： 白绿，绿，白橙，蓝，白蓝，橙，白棕，棕； B端：（标准568B）：白橙，橙，白绿，蓝，白蓝，绿，白棕，棕。 三、HUB连HUB&nb
  • Vimium插件让键盘党像操作Vim一样操作Chrome dcj3sjt126com chromevim
    什么是键盘党？ 键盘党是指尽可能将所有电脑操作用键盘来完成，而不去动鼠标的人。鼠标应该说是新手们的最爱，很直观，指哪点哪，很听话！不过常常使用电脑的人，如果一直使用鼠标的话，手会发酸，因为操作鼠标的时候，手臂不是在一个自然的状态，臂肌会处于绷紧状态。而使用键盘则双手是放松状态，只有手指在动。而且尽量少的从鼠标移动到键盘来回操作，也省不少事。 在chrome里安装 vimium 插件
  • MongoDB查询（2）——数组查询[六] eksliang mongodbMongoDB查询数组
    MongoDB查询数组 转载请出自出处：http://eksliang.iteye.com/blog/2177292 一、概述  MongoDB查询数组与查询标量值是一样的，例如，有一个水果列表，如下所示： > db.food.find() { "_id" : "001", "fruits" : [ "苹
  • cordova读写文件（1） gundumw100 JavaScriptCordova
    使用cordova可以很方便的在手机sdcard中读写文件。 首先需要安装cordova插件：file 命令为： cordova plugin add org.apache.cordova.file 然后就可以读写文件了，这里我先是写入一个文件，具体的JS代码为： var datas=null;//datas need write var directory=&
  • HTML5 FormData 进行文件jquery ajax 上传 到又拍云 ileson jqueryAjaxhtml5FormData
    html5 新东西：FormData  可以提交二进制数据。 页面test.html <!DOCTYPE> <html> <head> <title> formdata file jquery ajax upload</title> </head> <body> <
  • swift appearanceWhenContainedIn:(version1.2 xcode6.4) 啸笑天 version
      swift1.2中没有oc中对应的方法： + (instancetype)appearanceWhenContainedIn:(Class <UIAppearanceContainer>)ContainerClass, ... NS_REQUIRES_NIL_TERMINATION;  解决方法： 在swift项目中新建oc类如下： #import &
  • java实现SMTP邮件服务器 macroli java编程
    电子邮件传递可以由多种协议来实现。目前，在Internet 网上最流行的三种电子邮件协议是SMTP、POP3 和 IMAP，下面分别简单介绍。 　　◆ SMTP 协议 　　简单邮件传输协议(Simple Mail Transfer Protocol,SMTP)是一个运行在TCP/IP之上的协议，用它发送和接收电子邮件。SMTP 服务器在默认端口25上监听。SMTP客户使用一组简单的、基于文本的
  • mongodb group by having where 查询sql qiaolevip 每天进步一点点学习永无止境mongo纵观千象
    SELECT cust_id, SUM(price) as total FROM orders WHERE status = 'A' GROUP BY cust_id HAVING total > 250 db.orders.aggregate( [ { $match: { status: 'A' } }, { $group: {
  • Struts2 Pojo（六） Luob. POJOstrust2
    注意：附件中有完整案例 1.采用POJO对象的方法进行赋值和传值 2.web配置 <?xml version="1.0" encoding="UTF-8"?> <web-app version="2.5" xmlns="http://java.sun.com/xml/ns/javaee&q
  • struts2步骤 wuai struts
    1、添加jar包 2、在web.xml中配置过滤器 <filter>        <filter-name>struts2</filter-name>        <filter-class>org.apache.st
按字母分类： ABCDEFGHIJKLMNOPQRSTUVWXYZ其他