为何Rust编写HTTP请求解析逻辑时循环内的状态变量未按预期更新?
问题描述
刚读完《Rust 编程语言》一书,为了自学Rust正在开发一个简单的HTTP服务器。当前编写了一个用于解析&[u8]字节切片并生成HttpRequest对象的函数,遇到异常:本应逐行解析请求头的循环,并没有更新用于跟踪解析状态的变量。
存在异常的解析代码
pub fn parse_request(buffer: &[u8]) -> Result<HttpRequest, &'static str> { let (mut line, mut curr_req_reader_pos) = next_req_line(buffer, 0); let (method, path, version) = parse_method_path_and_version(&line)?; let (line, curr_req_reader_pos) = next_req_line(buffer, curr_req_reader_pos); //eprintln!("--- got next line: {}, {} ---", curr_req_reader_pos, String::from_utf8_lossy(line)); let mut headers = vec![]; let mut lel = 0; while line.len() > 0 { //eprintln!("LOOP"); let header = parse_header(&line); eprintln!("--- parsed header: {} ---", String::from_utf8_lossy(line)); headers.push(header); let (line, curr_req_reader_pos) = next_req_line(buffer, curr_req_reader_pos); //eprintln!("--- got next line: {}, {} ---", curr_req_reader_pos, String::from_utf8_lossy(line)); let (line, curr_req_reader_pos) = next_req_line(buffer, curr_req_reader_pos); //eprintln!("--- got next line: {}, {} ---", curr_req_reader_pos, String::from_utf8_lossy(line)); //eprintln!("LOOP"); lel += 10; //break; } let (line, curr_req_reader_pos) = next_req_line(buffer, curr_req_reader_pos); //eprintln!("--- got next line: {}, {} ---", curr_req_reader_pos, String::from_utf8_lossy(line)); //eprintln!("{}", lel); let has_body; match method.as_ref() { "POST" | "PUT" | "PATCH" => has_body = true, _ => has_body = true }; let body; if has_body { let (line, curr_req_reader_pos) = next_req_line(buffer, curr_req_reader_pos); body = String::from_utf8_lossy(line).to_string(); } else { body = String::new(); } Ok(HttpRequest { method, path, version, headers, has_body, body }) }
配套测试用例
#[test] fn test_next_req_line() { let req = "GET / HTTP/1.1\r\nContent-Length: 3\r\n\r\n\r\n\r\nlel".as_bytes(); let (mut lel, mut pos) = next_req_line(req, 0); assert_eq!("GET / HTTP/1.1", String::from_utf8_lossy(lel)); assert!(lel.len() > 0); assert_eq!(16, pos); let (lel, pos) = next_req_line(req, pos); assert_eq!("Content-Length: 3", String::from_utf8_lossy(lel)); assert!(lel.len() > 0); assert_eq!(35, pos); let (lel, pos) = next_req_line(req, pos); assert_eq!("", String::from_utf8_lossy(lel)); assert!(lel.len() == 0); assert_eq!(37, pos); let (lel, pos) = next_req_line(req, pos); assert_eq!("", String::from_utf8_lossy(lel)); assert!(lel.len() == 0); let (lel, pos) = next_req_line(req, pos); assert_eq!("", String::from_utf8_lossy(lel)); assert!(lel.len() == 0); let (lel, pos) = next_req_line(req, pos); assert_eq!("lel", String::from_utf8_lossy(lel)); assert!(lel.len() == 3); } #[test] fn test_parse_request() { let mut request = String::from("GET / HTTP/1.1\r\n"); request.push_str("Content-Length: 3\r\n"); request.push_str("\r\n"); request.push_str("lel"); let request = request.as_bytes(); match parse_request(&request) { Ok(http_request) => { assert_eq!(http_request.method, "GET"); assert_eq!(http_request.path, "/"); assert_eq!(http_request.version, "HTTP/1.1"); }, Err(_) => { println!("Failed"); } } }
测试运行输出
Finished test [unoptimized + debuginfo] target(s) in 0.53s Running target/debug/deps/rweb-6bbc2a3130f7e3d9 running 4 tests test http::testing::test_next_req_line ... ok test http::testing::test_parse_header ... ok test http::testing::test_parse_method_path_and_version ... ok --- parsed method path version --- --- got next line: 35, Content-Length: 3 --- LOOP --- parsed header: Content-Length: 3 --- --- got next line: 37, --- --- got next line: 39, lel --- LOOP --- got next line: 37, --- 10 test http::testing::test_parse_request ... ok test result: ok. 4 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
异常表现
- 独立测试的各个解析方法功能完全正常,循环外的基础解析逻辑符合预期
while循环内部的状态跟踪变量line、curr_req_reader_pos没有按预期被更新,无法逐行收集请求头- 调试日志显示循环只执行了一次迭代就退出,和预期的逐行解析逻辑不符
问题根因
这是Rust 变量遮蔽(Shadowing) 规则带来的典型新手坑:
- 函数开头声明的
mut line、mut curr_req_reader_pos,是存在于函数全作用域的可变变量,本意是用来在循环迭代中持续更新解析进度 - 代码中所有需要更新这两个状态的位置,都使用了
let关键字做解构赋值。Rust语法规定,let关键字永远是在当前作用域创建全新的局部变量,不会去给外层同名变量赋值;这些新建的同名变量会在当前块内完全遮蔽外层的状态变量,块执行结束后就会被销毁,根本不会修改外层状态的值 while循环的退出判断读取的永远是外层没有被更新的状态变量,所以看起来就像循环完全没有更新变量一样
修复方法
- 把所有需要更新外层状态的位置,去掉
let关键字,直接给开头声明的可变变量赋值,不要创建新的遮蔽变量 - 清理逻辑冗余的问题:
- 循环内部连续调用了两次
next_req_line拉取下一行,会跳过有效请求头,根据HTTP解析规则,每次循环只需要调用一次拉取新行 - 最后的
has_body分支两个判断结果完全一致,不需要冗余的match逻辑
- 循环内部连续调用了两次
调整后的核心循环逻辑示例:
while line.len() > 0 { let header = parse_header(&line); eprintln!("--- parsed header: {} ---", String::from_utf8_lossy(line)); headers.push(header); // 去掉let,直接修改外层的可变状态变量 (line, curr_req_reader_pos) = next_req_line(buffer, curr_req_reader_pos); lel += 10; }
内容的提问来源于stack exchange,提问作者stacker
相关产品推荐
相关产品推荐

