gzip解压strpython_HTTP响应gzip+chunked分段压缩流的解压缩问题

Accept-Encoding: gzip

// 响应头不会压缩
HTTP/1.1 200 OK
Server: Apache-Coyote/1.1
Content-Encoding: gzip
Content-Type: text/html;charset&＃61;UTF-8
Date: Wed, 20 Feb 2019 06:19:04 GMT// 响应体会进压缩
xxxxxxxxxx

public static byte[] compress(String str, String encoding) {if (str &＃61;&＃61; null || str.length() &＃61;&＃61; 0) {return null;}ByteArrayOutputStream out &＃61; new ByteArrayOutputStream();GZIPOutputStream gzip;try {gzip &＃61; new GZIPOutputStream(out);gzip.write(str.getBytes(encoding));//将字符串转为字节数组&＃xff0c;对字节数组进行压缩gzip.close();} catch (IOException e) {}return out.toByteArray();//返回压缩后的字节流}

Content-Encoding: gzip
Content-Length: 13131

HTTP/1.1 200 OK
Server: Apache-Coyote/1.1
Content-Encoding: gzip
Content-Type: text/html;charset&＃61;UTF-8
Transfer-Encoding: chunked // 分段的数据就会返回这个响应头
Date: Wed, 20 Feb 2019 06:19:04 GMTa3 // 16进制
xxxxx
5d9f
xxxxx
0 // 以 0 为结尾

public class SegmentInputStream extends InputStream {private InputStream mInputStream; //需要处理的字节流private HashMap mHeaders; //响应头private boolean mChunked; //分段的标识private boolean mIsData; private boolean mEnd; //读取到末尾的标志 即读取到长度为 0private long mReadLength &＃61; 0L;//当前读取到的长度private long mSegmentLength &＃61; -1L; //分段时&＃xff0c;每一段的长度public final boolean DEBUG &＃61; true;public SegmentInputStream(InputStream inputStream) throws IOException {mInputStream &＃61; inputStream;mHeaders &＃61; new HashMap<>();mChunked &＃61; false;mIsData &＃61; false;mEnd &＃61; false;parseHeaders(); //在构造函数的时候就先将响应头解析&＃xff0c;因为其没有压缩}public HashMap getHeaders() {return mHeaders;}//重写read 方法&＃xff0c;每次读的时候跳过分段的16 进制数字&＃64;Overridepublic int read() throws IOException {return !mChunked ? mInputStream.read() : readChunked();}private int readChunked() throws IOException {if (mEnd) {return -1;}int byteCode;if (mIsData) {byteCode &＃61; mInputStream.read();mReadLength&＃43;&＃43;;if (mReadLength &＃61;&＃61; mSegmentLength) {mIsData &＃61; false;mReadLength &＃61; 0L;mSegmentLength &＃61; -1L;}} // <<数据的部分读取完毕else {int endTag &＃61; 0;//回车字符标识 一个 /n/r 就是一个回车byte[] buffer &＃61; new byte[1];ArrayList bytes &＃61; new ArrayList<>();while ((byteCode &＃61; mInputStream.read()) !&＃61; -1) {buffer[0] &＃61; (byte) byteCode;// 因为read(x,x,x)// 最后会调用read 所以是一个递归&＃xff0c;会栈溢出if (buffer[0] !&＃61; &＃39;r&＃39; && buffer[0] !&＃61; &＃39;n&＃39;) {bytes.add(buffer[0]);endTag &＃61; 0;} else {/* (buffer[0] &＃61;&＃61; &＃39;n&＃39; || buffer[0] &＃61;&＃61; &＃39;r&＃39;)*/endTag&＃43;&＃43;;if (endTag &＃61;&＃61; 2 && bytes.size() !&＃61; 0) {//四个字符就是有两个回车符&＃xff0c;响应头就终止byte[] resultByte &＃61; new byte[bytes.size()];for (int i &＃61; 0; i bytes &＃61; new ArrayList<>();while (read(buffer, 0, 1) !&＃61; -1) { //bytes.add(buffer[0]);if (buffer[0] &＃61;&＃61; &＃39;n&＃39; || buffer[0] &＃61;&＃61; &＃39;r&＃39;) {enterCount&＃43;&＃43;;if (enterCount &＃61;&＃61; 4) { //四个字符就是有两个回车符&＃xff0c;响应头就终止break;}} else {enterCount &＃61; 0;}}byte[] resultByte &＃61; new byte[bytes.size()];for (int i &＃61; 0; i