java、android可用的rtp封包解包h264
做直播，音视频通讯。经常需要通过rtp协议封装音视频数据来发送。网上找到的基本都是c或c++版本的，没有JAVA版本的。就算千辛万苦找到一篇java版本的，要么不能用，要么就是一些片段，要么有封包没解包。很是蛋疼，本人也是这样，刚开始不太熟悉rtp协议，不太明白怎么封包组包，痛苦了几天，终于搞出来了，分享给有需要的朋友，希望对你们有所帮助。
直接看代码吧。不多说了。
首先看看关键类：
package com.imsdk.socket.udp.codec;
import android.os.SystemClock;
import android.util.Log;
import java.io.ByteArrayInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.math.BigDecimal;
import java.util.Random;
import java.util.concurrent.Semaphore;
public class RtspPacketEncode {
private static final String TAG = "RtspPacketEncode";
//------------视频转换数据监听-----------
public interface H264ToRtpLinsener {
void h264ToRtpResponse(byte[] out, int len);
}
private H264ToRtpLinsener h264ToRtpLinsener;
//执行回调
private void exceuteH264ToRtpLinsener(byte[] out, int len) {
if (this.h264ToRtpLinsener != null) {
h264ToRtpLinsener.h264ToRtpResponse(out, len);
}
}
// -------视频--------
private int framerate = 10;
private byte[] sendbuf = new byte[1500];
private int packageSize = 1400;
private int seq_num = 0;
private int timestamp_increse = (int) (90000.0 / framerate);//framerate是帧率
private int ts_current = 0;
private int bytes = 0;
// -------视频END--------
public RtspPacketEncode(H264ToRtpLinsener h264ToRtpLinsener) {
this.h264ToRtpLinsener = h264ToRtpLinsener;
}
/**
* 一帧一帧的RTP封包
*
* @param r
* @return
*/
public void h264ToRtp(byte[] r, int h264len) throws Exception {
CalculateUtil.memset(sendbuf, 0, 1500);
sendbuf[1] = (byte) (sendbuf[1] | 96); // 负载类型号96,其值为：01100000
sendbuf[0] = (byte) (sendbuf[0] | 0x80); // 版本号,此版本固定为2
sendbuf[1] = (byte) (sendbuf[1] & 254); //标志位，由具体协议规定其值，其值为：01100000
sendbuf[11] = 10;//随机指定10，并在本RTP回话中全局唯一,java默认采用网络字节序号 不用转换（同源标识符的最后一个字节）
if (h264len <= packageSize) {
sendbuf[1] = (byte) (sendbuf[1] | 0x80); // 设置rtp M位为1，其值为：11100000，分包的最后一片，M位（第一位）为0，后7位是十进制的96，表示负载类型
sendbuf[3] = (byte) seq_num++;
System.arraycopy(CalculateUtil.intToByte(seq_num++), 0, sendbuf, 2, 2);//send[2]和send[3]为序列号，共两位
{
// java默认的网络字节序是大端字节序（无论在什么平台上），因为windows为小字节序，所以必须倒序
/**参考：
* http://blog.csdn.net/u011068702/article/details/51857557
* http://cpjsjxy.iteye.com/blog/1591261
*/
byte temp = 0;
temp = sendbuf[3];
sendbuf[3] = sendbuf[2];
sendbuf[2] = temp;
}
// FU-A HEADER, 并将这个HEADER填入sendbuf[12]
sendbuf[12] = (byte) (sendbuf[12] | ((byte) (r[0] & 0x80)) << 7);
sendbuf[12] = (byte) (sendbuf[12] | ((byte) ((r[0] & 0x60) >> 5)) << 5);
sendbuf[12] = (byte) (sendbuf[12] | ((byte) (r[0] & 0x1f)));
// 同理将sendbuf[13]赋给nalu_payload
//NALU头已经写到sendbuf[12]中，接下来则存放的是NAL的第一个字节之后的数据。所以从r的第二个字节开始复制
System.arraycopy(r, 1, sendbuf, 13, h264len - 1);
ts_current = ts_current + timestamp_increse;
System.arraycopy(CalculateUtil.intToByte(ts_current), 0, sendbuf, 4, 4);//序列号接下来是时间戳，4个字节，存储后也需要倒序
{
byte temp = 0;
temp = sendbuf[4];
sendbuf[4] = sendbuf[7];
sendbuf[7] = temp;
temp = sendbuf[5];
sendbuf[5] = sendbuf[6];
sendbuf[6] = temp;
}
bytes = h264len + 12;//获sendbuf的长度,为nalu的长度(包含nalu头但取出起始前缀,加上rtp_header固定长度12个字节)
//client.send(new DatagramPacket(sendbuf, bytes, addr, port/*9200*/));
//send(sendbuf,bytes);
exceuteH264ToRtpLinsener(sendbuf, bytes);
} else if (h264len > packageSize) {
int k = 0, l = 0;
k = h264len / packageSize;
l = h264len % packageSize;
int t = 0;
ts_current = ts_current + timestamp_increse;
System.arraycopy(CalculateUtil.intToByte(ts_current), 0, sendbuf, 4, 4);//时间戳，并且倒序
{
byte temp = 0;
temp = sendbuf[4];
sendbuf[4] = sendbuf[7];
sendbuf[7] = temp;
temp = sendbuf[5];
sendbuf[5] = sendbuf[6];
sendbuf[6] = temp;
}
while (t <= k) {
System.arraycopy(CalculateUtil.intToByte(seq_num++), 0, sendbuf, 2, 2);//序列号，并且倒序
{
byte temp = 0;
temp = sendbuf[3];
sendbuf[3] = sendbuf[2];
sendbuf[2] = temp;
}
if (t == 0) {//分包的第一片
sendbuf[1] = (byte) (sendbuf[1] & 0x7F);//其值为：01100000，不是最后一片，M位（第一位）设为0
//FU indicator，一个字节，紧接在RTP header之后，包括F,NRI，header
sendbuf[12] = (byte) (sendbuf[12] | ((byte) (r[0] & 0x80)) << 7);//禁止位，为0
sendbuf[12] = (byte) (sendbuf[12] | ((byte) ((r[0] & 0x60) >> 5)) << 5);//NRI，表示包的重要性
sendbuf[12] = (byte) (sendbuf[12] | (byte) (28));//TYPE，表示此FU-A包为什么类型，一般此处为28
//FU header，一个字节，S,E，R，TYPE
sendbuf[13] = (byte) (sendbuf[13] & 0xBF);//E=0，表示是否为最后一个包，是则为1
sendbuf[13] = (byte) (sendbuf[13] & 0xDF);//R=0，保留位，必须设置为0
sendbuf[13] = (byte) (sendbuf[13] | 0x80);//S=1，表示是否为第一个包，是则为1
sendbuf[13] = (byte) (sendbuf[13] | ((byte) (r[0] & 0x1f)));//TYPE，即NALU头对应的TYPE
//将除去NALU头剩下的NALU数据写入sendbuf的第14个字节之后。前14个字节包括：12字节的RTP Header，FU indicator，FU header
System.arraycopy(r, 1, sendbuf, 14, packageSize);
//client.send(new DatagramPacket(sendbuf, packageSize + 14, addr, port/*9200*/));
exceuteH264ToRtpLinsener(sendbuf, packageSize + 14);
t++;
} else if (t == k) {//分片的最后一片
sendbuf[1] = (byte) (sendbuf[1] | 0x80);
sendbuf[12] = (byte) (sendbuf[12] | ((byte) (r[0] & 0x80)) << 7);
sendbuf[12] = (byte) (sendbuf[12] | ((byte) ((r[0] & 0x60) >> 5)) << 5);
sendbuf[12] = (byte) (sendbuf[12] | (byte) (28));
sendbuf[13] = (byte) (sendbuf[13] & 0xDF); //R=0，保留位必须设为0
sendbuf[13] = (byte) (sendbuf[13] & 0x7F); //S=0，不是第一个包
sendbuf[13] = (byte) (sendbuf[13] | 0x40); //E=1，是最后一个包
sendbuf[13] = (byte) (sendbuf[13] | ((byte) (r[0] & 0x1f)));//NALU头对应的type
if (0 != l) {//如果不能整除，则有剩下的包，执行此代码。如果包大小恰好是1400的倍数，不执行此代码。
System.arraycopy(r, t * packageSize + 1, sendbuf, 14, l - 1);//l-1，不包含NALU头
bytes = l - 1 + 14; //bytes=l-1+14;
//client.send(new DatagramPacket(sendbuf, bytes, addr, port/*9200*/));
//send(sendbuf,bytes);
exceuteH264ToRtpLinsener(sendbuf, bytes);
}//pl
t++;
} else if (t < k && 0 != t) {//既不是第一片，又不是最后一片的包
sendbuf[1] = (byte) (sendbuf[1] & 0x7F); //M=0，其值为：01100000，不是最后一片，M位（第一位）设为0.
sendbuf[12] = (byte) (sendbuf[12] | ((byte) (r[0] & 0x80)) << 7);
sendbuf[12] = (byte) (sendbuf[12] | ((byte) ((r[0] & 0x60) >> 5)) << 5);
sendbuf[12] = (byte) (sendbuf[12] | (byte) (28));
sendbuf[13] = (byte) (sendbuf[13] & 0xDF); //R=0，保留位必须设为0
sendbuf[13] = (byte) (sendbuf[13] & 0x7F); //S=0，不是第一个包
sendbuf[13] = (byte) (sendbuf[13] & 0xBF); //E=0，不是最后一个包
sendbuf[13] = (byte) (sendbuf[13] | ((byte) (r[0] & 0x1f)));//NALU头对应的type
System.arraycopy(r, t * packageSize + 1, sendbuf, 14, packageSize);//不包含NALU头
//client.send(new DatagramPacket(sendbuf, packageSize + 14, addr, port/*9200*/));
//send(sendbuf,1414);
exceuteH264ToRtpLinsener(sendbuf, packageSize + 14);
t++;
}
}
}
}
}
计算类：
package com.imsdk.socket.udp.codec;
/**
* 计算类
*
* @author kokJuis
*/
public class CalculateUtil {
/**
* 注释：int到字节数组的转换！
*
* @param number
* @return
*/
public static byte[] intToByte(int number) {
int temp = number;
byte[] b = new byte[4];
for (int i = 0; i < b.length; i++) {
b[i] = new Integer(temp & 0xff).byteValue();// 将最低位保存在最低位
temp = temp >> 8; // 向右移8位
}
return b;
}
public static int byteToInt(byte b) {
//Java 总是把 byte 当做有符处理；我们可以通过将其和 0xFF 进行二进制与得到它的无符值
return b & 0xFF;
}
//byte 数组与 int 的相互转换
public static int byteArrayToInt(byte[] b) {
return b[3] & 0xFF |
(b[2] & 0xFF) << 8 |
(b[1] & 0xFF) << 16 |
(b[0] & 0xFF) << 24;
}
public static byte[] intToByteArray(int a) {
return new byte[] {
(byte) ((a >> 24) & 0xFF),
(byte) ((a >> 16) & 0xFF),
(byte) ((a >> 8) & 0xFF),
(byte) (a & 0xFF)
};
}
// 清空buf的值
public static void memset(byte[] buf, int value, int size) {
for (int i = 0; i < size; i++) {
buf[i] = (byte) value;
}
}
public static void dump(NALU_t n) {
System.out.println("len: " + n.len + " nal_unit_type:" + n.nal_unit_type);
}
// 判断是否为0x000001,如果是返回1
public static int FindStartCode2(byte[] Buf, int off) {
if (Buf[0 + off] != 0 || Buf[1 + off] != 0 || Buf[2 + off] != 1)
return 0;
else
return 1;
}
// 判断是否为0x00000001,如果是返回1
public static int FindStartCode3(byte[] Buf, int off) {
if (Buf[0 + off] != 0 || Buf[1 + off] != 0 || Buf[2 + off] != 0 || Buf[3 + off] != 1)
return 0;
else
return 1;
}
}
使用的话，实现监听就可以了：
@Override
public void h264ToRtpResponse(byte[] out, int len) {
//h264转rtp监听
if (out != null) {
Log.v(TAG, "---发送数据---" + len);
netSendTask.pushBuf(out, len);
}
}
rtspPacketEncode.h264ToRtp(h264, ret);组包类：
package com.imsdk.socket.udp.codec;
public class RtspPacketDecode {
private byte[] h264Buffer;
private int h264Len = 0;
private int h264Pos = 0;
private static final byte[] start_code = {0, 0, 0, 1}; // h264 start code
//传入视频的分辨率
public RtspPacketDecode(int width, int height) {
h264Buffer = new byte[getYuvBuffer(width, height)];
}
/**
* RTP解包H264
*
* @param rtpData
* @return
*/
public byte[] rtp2h264(byte[] rtpData, int rtpLen) {
int fu_header_len = 12; // FU-Header长度为12字节
int extension = (rtpData[0] & (1 << 4)); // X: 扩展为是否为1
if (extension > 0) {
// 计算扩展头的长度
int extLen = (rtpData[12] << 24) + (rtpData[13] << 16) + (rtpData[14] << 8) + rtpData[15];
fu_header_len += (extLen + 1) * 4;
}
// 解析FU-indicator
byte indicatorType = (byte) (CalculateUtil.byteToInt(rtpData[fu_header_len]) & 0x1f); // 取出low 5 bit 则为FU-indicator type
byte nri = (byte) ((CalculateUtil.byteToInt(rtpData[fu_header_len]) >> 5) & 0x03); // 取出h2bit and h3bit
byte f = (byte) (CalculateUtil.byteToInt(rtpData[fu_header_len]) >> 7); // 取出h1bit
byte h264_nal_header;
byte fu_header;
if (indicatorType == 28) { // FU-A
fu_header = rtpData[fu_header_len + 1];
byte s = (byte) (rtpData[fu_header_len + 1] & 0x80);
byte e = (byte) (rtpData[fu_header_len + 1] & 0x40);
if (e == 64) { // end of fu-a
//ZOLogUtil.d("RtpParser", "end of fu-a.....;;;");
byte[] temp = new byte[rtpLen - (fu_header_len + 2)];
System.arraycopy(rtpData, fu_header_len + 2, temp, 0, temp.length);
writeData2Buffer(temp, temp.length);
if (h264Pos >= 0) {
h264Pos = -1;
if (h264Len > 0) {
byte[] h264Data = new byte[h264Len];
System.arraycopy(h264Buffer, 0, h264Data, 0, h264Len);
h264Len = 0;
return h264Data;
}
}
} else if (s == -128) { // start of fu-a
h264Pos = 0; // 指针归0
writeData2Buffer(start_code, 4); // 写入H264起始码
h264_nal_header = (byte) ((fu_header & 0x1f) | (nri << 5) | (f << 7));
writeData2Buffer(new byte[]{h264_nal_header}, 1);
byte[] temp = new byte[rtpLen - (fu_header_len + 2)];
System.arraycopy(rtpData, fu_header_len + 2, temp, 0, temp.length); // 负载数据
writeData2Buffer(temp, temp.length);
} else {
byte[] temp = new byte[rtpLen - (fu_header_len + 2)];
System.arraycopy(rtpData, fu_header_len + 2, temp, 0, temp.length);
writeData2Buffer(temp, temp.length);
}
} else { // nalu
h264Pos = 0;
writeData2Buffer(start_code, 4);
byte[] temp = new byte[rtpLen - fu_header_len];
System.arraycopy(rtpData, fu_header_len, temp, 0, temp.length);
writeData2Buffer(temp, temp.length);
if (h264Pos >= 0) {
h264Pos = -1;
if (h264Len > 0) {
byte[] h264Data = new byte[h264Len];
System.arraycopy(h264Buffer, 0, h264Data, 0, h264Len);
h264Len = 0;
return h264Data;
}
}
}
return null;
}
private void writeData2Buffer(byte[] data, int len) {
if (h264Pos >= 0) {
System.arraycopy(data, 0, h264Buffer, h264Pos, len);
h264Pos += len;
h264Len += len;
}
}
//计算h264大小
public int getYuvBuffer(int width, int height) {
// stride = ALIGN(width, 16)
int stride = (int) Math.ceil(width / 16.0) * 16;
// y_size = stride * height
int y_size = stride * height;
// c_stride = ALIGN(stride/2, 16)
int c_stride = (int) Math.ceil(width / 32.0) * 16;
// c_size = c_stride * height/2
int c_size = c_stride * height / 2;
// size = y_size + c_size * 2
return y_size + c_size * 2;
}
}
使用：
byte[] tmp = rtspPacketDecode.rtp2h264(out,len);