摘要

在详解http报文一文中，详细介绍了http报文的文本结构。那么作为服务端，web容器是如何解析http报文的呢？本文以jetty和undertow容器为例，来解析web容器是如何处理http报文的。

在前文中我们从概览中可以了解到，http报文其实就是一定规则的字符串，那么解析它们，就是解析字符串，看看是否满足http协议约定的规则。

start-line: 起始行,描述请求或响应的基本信息

*( header-field CRLF ): 头

CRLF

[message-body]: 消息body，实际传输的数据

jetty

以下代码都是jetty9.4.12版本

如何解析这么长的字符串呢，jetty是通过状态机来实现的。具体可以看下org.eclipse.jetty.http.HttpParse类

 public enum State
    {
        START,
        METHOD,
        
![](https://img2018.cnblogs.com/blog/1147363/201910/1147363-20191009220439773-204646534.png),
        SPACE1,
        STATUS,
        URI,
        SPACE2,
        REQUEST_VERSION,
        REASON,
        PROXY,
        HEADER,
        CONTENT,
        EOF_CONTENT,
        CHUNKED_CONTENT,
        CHUNK_SIZE,
        CHUNK_PARAMS,
        CHUNK,
        TRAILER,
        END,
        CLOSE,  // The associated stream/endpoint should be closed
        CLOSED  // The associated stream/endpoint is at EOF
    }

总共分成了21种状态，然后进行状态间的流转。在parseNext方法中分别对起始行 -> header -> body content分别解析

public boolean parseNext(ByteBuffer buffer)
    {
        try
        {
            // Start a request/response
            if (_state==State.START)
            {
                // 快速判断
                if (quickStart(buffer))
                    return true;
            }

            // Request/response line 转换
            if (_state.ordinal()>= State.START.ordinal() && _state.ordinal()<State.HEADER.ordinal())
            {
                if (parseLine(buffer))
                    return true;
            }

            // headers转换
            if (_state== State.HEADER)
            {
                if (parseFields(buffer))
                    return true;
            }

            // content转换
            if (_state.ordinal()>= State.CONTENT.ordinal() && _state.ordinal()<State.TRAILER.ordinal())
            {
                // Handle HEAD response
                if (_responseStatus>0 && _headResponse)
                {
                    setState(State.END);
                    return handleContentMessage();
                }
                else
                {
                    if (parseContent(buffer))
                        return true;
                }
            }
         
        return false;
    }

整体流程

整体有三条路径

1. 开始 -> start-line -> header -> 结束
2. 开始 -> start-line -> header -> content -> 结束
3. 开始 -> start-line -> header -> chunk-content -> 结束
   

起始行

start-line = request-line(请求起始行）/（响应起始行）status-line

1. 请求报文解析状态迁移
   请求行：START -> METHOD -> SPACE1 -> URI -> SPACE2 -> REQUEST_VERSION

2. 响应报文解析状态迁移
   响应行：START -> RESPONSE_VERSION -> SPACE1 -> STATUS -> SPACE2 -> REASON

header 头

HEADER 的状态只有一种了，在jetty的老版本中还区分了HEADER_IN_NAM, HEADER_VALUE, HEADER_IN_VALUE等，9.4中都去除了。为了提高匹配效率，jetty使用了Trie树快速匹配header头。

static
    {
        CACHE.put(new HttpField(HttpHeader.CONNECTION,HttpHeaderValue.CLOSE));
        CACHE.put(new HttpField(HttpHeader.CONNECTION,HttpHeaderValue.KEEP_ALIVE));
      // 以下省略了很多了通用header头

content

请求体：

1. CONTENT -> END，这种是普通的带Content-Length头的报文，HttpParser一直运行CONTENT状态，直到最后ContentLength达到了指定的数量，则进入END状态
2. chunked分块传输的数据
   CHUNKED_CONTENT -> CHUNK_SIZE -> CHUNK -> CHUNK_END -> END

undertow

undertow是另一种web容器，它的处理方式与jetty有什么不同呢
状态机种类不一样了，io.undertow.util.HttpString.ParseState

    public static final int VERB = 0;
    public static final int PATH = 1;
    public static final int PATH_PARAMETERS = 2;
    public static final int QUERY_PARAMETERS = 3;
    public static final int VERSION = 4;
    public static final int AFTER_VERSION = 5;
    public static final int HEADER = 6;
    public static final int HEADER_VALUE = 7;
    public static final int PARSE_COMPLETE = 8;

具体处理流程在HttpRequestParser抽象类中

public void handle(ByteBuffer buffer, final ParseState currentState, final HttpServerExchange builder) throws BadRequestException {
        if (currentState.state == ParseState.VERB) {
            //fast path, we assume that it will parse fully so we avoid all the if statements

            // 快速处理GET
            final int position = buffer.position();
            if (buffer.remaining() > 3
                    && buffer.get(position) == 'G'
                    && buffer.get(position + 1) == 'E'
                    && buffer.get(position + 2) == 'T'
                    && buffer.get(position + 3) == ' ') {
                buffer.position(position + 4);
                builder.setRequestMethod(Methods.GET);
                currentState.state = ParseState.PATH;
            } else {
                try {
                    handleHttpVerb(buffer, currentState, builder);
                } catch (IllegalArgumentException e) {
                    throw new BadRequestException(e);
                }
            }
            // 处理path
            handlePath(buffer, currentState, builder);
           // 处理版本
            if (failed) {
                handleHttpVersion(buffer, currentState, builder);
                handleAfterVersion(buffer, currentState);
            }
            // 处理header
            while (currentState.state != ParseState.PARSE_COMPLETE && buffer.hasRemaining()) {
                handleHeader(buffer, currentState, builder);
                if (currentState.state == ParseState.HEADER_VALUE) {
                    handleHeaderValue(buffer, currentState, builder);
                }
            }
            return;
        }
        handleStateful(buffer, currentState, builder);
    }

与jetty不同的是对content的处理，在header处理完以后，将数据放到io.undertow.server.HttpServerExchange,然后根据类型，有不同的content读取方式，比如处理固定长度的，FixedLengthStreamSourceConduit。

关注公众号【方丈的寺院】，第一时间收到文章的更新，与方丈一起开始技术修行之路

参考

http://www.blogjava.net/DLevin/archive/2014/04/19/411673.html

https://www.ph0ly.com/2018/10/06/jetty/connection/http-parser/

https://webtide.com/http-trailers-in-jetty/

http://undertow.io/undertow-docs/undertow-docs-2.0.0/