我们继续来 写 Tomcat 这次我们做优化，先看一下一个标准的http 协议
GET /servlet/com.yixin.HelloWorldServlet HTTP/1.1
Host: localhost:8080
Connection: keep-alive
sec-ch-ua: "Microsoft Edge";v="123", "Not:A-Brand";v="8", "Chromium";v="123"
sec-ch-ua-mobile: ?0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36 Edg/123.0.0.0
sec-ch-ua-platform: "Windows"
Accept: image/avif,image/webp,image/apng,image/svg+xml,image/*,*/*;q=0.8
我们这一版就要是把所需要的数据 取出来，放到map 集合里边，先看第一个 我们能获取 请求方式GET， 请求路径 以及 请求协议 HTTP/1.1 然后 空一行， 获取到 请求头。
如果根据 这个报文结构，让各位写一个 获取协议 应该不难吧， 什么 split（），indexof， 什么subString 整吧，但是 tomcat 源码中就不会这么干了。
来 我们看下 ，
HttpRequestLine 类
public class HttpRequestLine {
public static final int INITIAL_METHOD_SIZE = 8;
public static final int INITIAL_URI_SIZE = 128;
public static final int INITIAL_PROTOCOL_SIZE = 8;
public static final int MAX_METHOD_SIZE = 32;
public static final int MAX_URI_SIZE = 2048;
public static final int MAX_PROTOCOL_SIZE = 32;
//下面的属性对应于Http Request规范，即头行格式method uri protocol
//如：GET /hello.txt HTTP/1.1
//char[] 存储每段的字符串，对应的int值存储的是每段的结束位置
public char[] method;
public int methodEnd;
public char[] uri;
public int uriEnd;
public char[] protocol;
public int protocolEnd;
public HttpRequestLine() {
this(new char[INITIAL_METHOD_SIZE], 0, new char[INITIAL_URI_SIZE], 0,
new char[INITIAL_PROTOCOL_SIZE], 0);
}
public HttpRequestLine(char[] method, int methodEnd,
char[] uri, int uriEnd,
char[] protocol, int protocolEnd) {
this.method = method;
this.methodEnd = methodEnd;
this.uri = uri;
this.uriEnd = uriEnd;
this.protocol = protocol;
this.protocolEnd = protocolEnd;
}
public void recycle() {
methodEnd = 0;
uriEnd = 0;
protocolEnd = 0;
}
public int indexOf(char[] buf) {
return indexOf(buf, buf.length);
}
//这是主要的方法
/**
* 这个方法遵循的 从你查到的第一个字符串开始找
* 举个例子 在helloworld 里边找or
* 以第一个字符o在helloworld 开始遍历 找到符合的就返回一个pos位置
* 然后以pos 为起点 开始循环依次对比 看是否后边的满足 如果不满足 返回 pos++
* 继续开始这样找下去 一直遍历完所有helloworld
*/
public int indexOf(char[] buf, int end) {
char firstChar = buf[0];
int pos = 0; //pos是查找字符串buf在uri[]中的开始位置
while (pos < uriEnd) {
pos = indexOfChar(firstChar, pos); //首字符定位开始位置
if (pos == -1) {
return -1;
}
//uriEnd - pos 表示url的最后剩余是否小与查找字符串的长度
//如果后边都没有了，依旧没有查到 难就查不到了。
if ((uriEnd - pos) < end) {
return -1;
}
for (int i = 0; i < end; i++) { //从开始位置起逐个字符比对
if (uri[i + pos] != buf[i]) {
break;
}
if (i == (end - 1)) { //每个字符都相等，则匹配上了，返回开始位置
return pos;
}
}
pos++;
}
return -1;
}
public int indexOf(String str) {
return indexOf(str.toCharArray(), str.length());
}
//在uri[]中查找字符c的出现位置
public int indexOfChar(char c, int start) {
for (int i = start; i < uriEnd; i++) {
if (uri[i] == c) {
return i;
}
}
return -1;
}
}
这个比较简单，就是查找字符串位置
接下来 比较复杂了。
SocketInputStream 类
public class SocketInputStream extends InputStream {
private static final byte CR = (byte) '\r';
private static final byte LF = (byte) '\n';// 换行
private static final byte SP = (byte) ' ';
private static final byte HT = (byte) '\t';
private static final byte COLON = (byte) ':';
private static final int LC_OFFSET = 'A' - 'a';
protected byte buf[];
protected int count;
protected int pos;
protected InputStream is;
public SocketInputStream(InputStream is, int bufferSize) {
this.is = is;
buf = new byte[bufferSize];
}
//从输入流中解析出request line
public void readRequestLine(HttpRequestLine requestLine)
throws IOException {
int chr = 0;
//跳过空行
do {
try {
chr = read();
} catch (IOException e) {
}
} while ((chr == CR) || (chr == LF));
//第一个非空位置
pos--;
int maxRead = requestLine.method.length;
int readStart = pos;
int readCount = 0;
boolean space = false;
//解析第一段method，以空格结束 解析出 GET
while (!space) {
if (pos >= count) {
int val = read();
if (val == -1) {
throw new IOException("requestStream.readline.error");
}
pos = 0;
readStart = 0;
}
if (buf[pos] == SP) {
space = true;
}
requestLine.method[readCount] = (char) buf[pos];
readCount++;
pos++;
}
requestLine.methodEnd = readCount - 1; //method段的结束位置
maxRead = requestLine.uri.length;
readStart = pos;
readCount = 0;
space = false;
boolean eol = false;
//解析第二段uri，以空格结束 解析出/servlet/com.yixin.HelloWorldServlet
while (!space) {
if (pos >= count) {
int val = read();
if (val == -1)
throw new IOException("requestStream.readline.error");
pos = 0;
readStart = 0;
}
if (buf[pos] == SP) {
space = true;
}
requestLine.uri[readCount] = (char) buf[pos];
readCount++;
pos++;
}
requestLine.uriEnd = readCount - 1; //uri结束位置
maxRead = requestLine.protocol.length;
readStart = pos;
readCount = 0;
//解析第三段protocol，以eol结尾 解析出 HTTP/1.1
while (!eol) {
if (pos >= count) {
int val = read();
if (val == -1)
throw new IOException("requestStream.readline.error");
pos = 0;
readStart = 0;
}
if (buf[pos] == CR) {
// Skip CR.
} else if (buf[pos] == LF) {
eol = true;
} else {
requestLine.protocol[readCount] = (char) buf[pos];
readCount++;
}
pos++;
}
requestLine.protocolEnd = readCount;
}
public void readHeader(HttpHeader header)
throws IOException {
int chr = read();
if ((chr == CR) || (chr == LF)) { // Skipping CR
if (chr == CR)
read(); // Skipping LF
header.nameEnd = 0;
header.valueEnd = 0;
return;
} else {
pos--;
}
// 正在读取 header name
int maxRead = header.name.length;
int readStart = pos;
int readCount = 0;
boolean colon = false;
while (!colon) {
// 我们处于内部缓冲区的末尾
if (pos >= count) {
int val = read();
if (val == -1) {
throw new IOException("requestStream.readline.error");
}
pos = 0;
readStart = 0;
}
if (buf[pos] == COLON) {
colon = true;
}
char val = (char) buf[pos];
if ((val >= 'A') && (val <= 'Z')) {
val = (char) (val - LC_OFFSET);
}
header.name[readCount] = val;
readCount++;
pos++;
}
header.nameEnd = readCount - 1;
// 读取 header 值（可以跨越多行）
maxRead = header.value.length;
readStart = pos;
readCount = 0;
int crPos = -2;
boolean eol = false;
boolean validLine = true;
while (validLine) {
boolean space = true;
// 跳过空格
// 注意：仅删除前面的空格，后面的不删。
while (space) {
// 我们已经到了内部缓冲区的尽头
if (pos >= count) {
// 将内部缓冲区的一部分（或全部）复制到行缓冲区
int val = read();
if (val == -1)
throw new IOException("requestStream.readline.error");
pos = 0;
readStart = 0;
}
if ((buf[pos] == SP) || (buf[pos] == HT)) {
pos++;
} else {
space = false;
}
}
while (!eol) {
// 我们已经到了内部缓冲区的尽头
if (pos >= count) {
// 将内部缓冲区的一部分（或全部）复制到行缓冲区
int val = read();
if (val == -1)
throw new IOException("requestStream.readline.error");
pos = 0;
readStart = 0;
}
if (buf[pos] == CR) {
} else if (buf[pos] == LF) {
eol = true;
} else {
// FIXME：检查二进制转换是否正常
int ch = buf[pos] & 0xff;
header.value[readCount] = (char) ch;
readCount++;
}
pos++;
}
int nextChr = read();
//Microsoft Edge 因为有的vlaue 值中有空格的情况
if ((nextChr != SP) && (nextChr != HT)) {
pos--;
validLine = false;
} else {
eol = false;
header.value[readCount] = ' ';
readCount++;
}
}
header.valueEnd = readCount;
}
@Override
public int read() throws IOException {
if (pos >= count) {
fill();
if (pos >= count) {
return -1;
}
}
return buf[pos++] & 0xff;
}
public int available() throws IOException {
return (count - pos) + is.available();
}
public void close() throws IOException {
if (is == null) {
return;
}
is.close();
is = null;
buf = null;
}
protected void fill() throws IOException {
pos = 0;
count = 0;
int nRead = is.read(buf, 0, buf.length);
if (nRead > 0) {
count = nRead;
}
System.out.println(new String(buf));
}
}
总归就是 获取 http协议中的数据 ，readRequestLine 这个方法， 别看复杂就是获取 第一行数据，GET 请求路径，等
接下来就是获取 请求头数据放入到map 集合：
HttpHeader
public class HttpHeader {
public static final int INITIAL_NAME_SIZE = 64;
public static final int INITIAL_VALUE_SIZE = 512;
public static final int MAX_NAME_SIZE = 128;
public static final int MAX_VALUE_SIZE = 1024;
public char[] name;
public int nameEnd;
public char[] value;
public int valueEnd;
protected int hashCode = 0;
public HttpHeader() {
this(new char[INITIAL_NAME_SIZE], 0, new char[INITIAL_VALUE_SIZE], 0);
}
public HttpHeader(char[] name, int nameEnd, char[] value, int valueEnd) {
this.name = name;
this.nameEnd = nameEnd;
this.value = value;
this.valueEnd = valueEnd;
}
public HttpHeader(String name, String value) {
this.name = name.toLowerCase().toCharArray();
this.nameEnd = name.length();
this.value = value.toCharArray();
this.valueEnd = value.length();
}
public void recycle() {
nameEnd = 0;
valueEnd = 0;
hashCode = 0;
}
}
这是一个请求头的类
JxdRequest 这个类 需要 只显示 变更的部分：
private InputStream input;
private SocketInputStream sis;
private String uri;
InetAddress address;
int port;
protected HashMap<String, String> headers = new HashMap<>();
protected Map<String, String> parameters = new ConcurrentHashMap<>();
HttpRequestLine requestLine = new HttpRequestLine();
public void parse(Socket socket) {
try {
input = socket.getInputStream();
this.sis = new SocketInputStream(this.input, 2048);
parseConnection(socket);
this.sis.readRequestLine(requestLine);
parseHeaders();
} catch (IOException e) {
e.printStackTrace();
} catch (ServletException e) {
e.printStackTrace();
}
this.uri = new String(requestLine.uri, 0, requestLine.uriEnd);
}
private void parseConnection(Socket socket) {
address = socket.getInetAddress();
port = socket.getPort();
}
private void parseHeaders() throws IOException, ServletException {
while (true) {
HttpHeader header = new HttpHeader();
sis.readHeader(header);
//表示读取完毕
if (header.nameEnd == 0) {
if (header.valueEnd == 0) {
return;
} else {
throw new ServletException("httpProcessor.parseHeaders.colon");
}
}
String name = new String(header.name, 0, header.nameEnd);
String value = new String(header.value, 0, header.valueEnd);
// 设置相应的请求头
if (name.equals(DefaultHeaders.ACCEPT_LANGUAGE_NAME)) {
headers.put(name, value);
} else if (name.equals(DefaultHeaders.CONTENT_LENGTH_NAME)) {
headers.put(name, value);
} else if (name.equals(DefaultHeaders.CONTENT_TYPE_NAME)) {
headers.put(name, value);
} else if (name.equals(DefaultHeaders.HOST_NAME)) {
headers.put(name, value);
} else if (name.equals(DefaultHeaders.CONNECTION_NAME)) {
headers.put(name, value);
} else if (name.equals(DefaultHeaders.TRANSFER_ENCODING_NAME)) {
headers.put(name, value);
} else {
headers.put(name, value);
}
}
}
这个请求request 里边有各种方法 ，解析 i请求头，解析请求 路径 方式等，
其他的没有变化 ，然后 可以用上一节的main 方法 运行一下，结果是一样的。