Node.js源碼解析-require背後

Node.js源碼解析-require背後

歡迎來個人博客閱讀：《Node.js源碼解析-require背後》

在編寫 Node.js 應用的過程當中，咱們或多或少的都寫過相似 const xxx = require('xxx') 的代碼，其做用是引入模塊。不知你們有沒有想過，這段代碼是如何肯定咱們要引入的模塊？又是以怎樣的上下文來執行模塊代碼的呢？

讓咱們來翻開 Node.js 源碼，先找到 lib/module.js 中的 Module.prototype.require() 函數

// lib/module.js

Module.prototype.require = function(path) {
  assert(path, 'missing path');
  assert(typeof path === 'string', 'path must be a string');
  return Module._load(path, this, /* isMain */ false);
};

// Check the cache for the requested file.
// 1. If a module already exists in the cache: return its exports object.
// 2. If the module is native: call `NativeModule.require()` with the
//    filename and return the result.
// 3. Otherwise, create a new module for the file and save it to the cache.
//    Then have it load  the file contents before returning its exports
//    object.
Module._load = function(request, parent, isMain) {
  if (parent) {
    debug('Module._load REQUEST %s parent: %s', request, parent.id);
  }

  var filename = Module._resolveFilename(request, parent, isMain);

  var cachedModule = Module._cache[filename];
  if (cachedModule) {
    return cachedModule.exports;
  }

  if (NativeModule.nonInternalExists(filename)) {
    debug('load native module %s', request);
    return NativeModule.require(filename);
  }

  var module = new Module(filename, parent);

  if (isMain) {
    process.mainModule = module;
    module.id = '.';
  }

  Module._cache[filename] = module;

  tryModuleLoad(module, filename);

  return module.exports;
};

Module.prototype.require() 對傳入的 path 簡單斷言後調用 Module._load() 來導入模塊

Module._load() 的執行思路以下:

• 查找模塊: Module._resolveFilename()

• 存在緩存: 返回 cachedModule.exports

• 是內置模塊: 見 NativeModule.require()

• 加載模塊: tryModuleLoad()

所以，Module.prototype.require() 的源碼能夠分爲兩大塊: 查找模塊和加載模塊

查找模塊

查找模塊的關鍵在於定位模塊的具體路徑，這個功能由 Module._resolveFilename() 函數實現

// lib/module.js

Module._resolveFilename = function(request, parent, isMain) {
  if (NativeModule.nonInternalExists(request)) {
    return request;
  }

  // 可能存在要加載模塊的目錄
  var paths = Module._resolveLookupPaths(request, parent, true);

  // 具體的模塊路徑
  var filename = Module._findPath(request, paths, isMain);
  if (!filename) {
    var err = new Error(`Cannot find module '${request}'`);
    err.code = 'MODULE_NOT_FOUND';
    throw err;
  }
  return filename;
};

Module._resolveFilename() 函數先調用 Module._resolveLookupPaths() 計算出模塊可能存在的目錄，而後調用 Module._findPath() 獲得模塊路徑

Module._resolveLookupPaths

經過調用 Module._resolveLookupPaths() 函數能夠計算出模塊可能存在的目錄，在調用時存在三種狀況:

• 直接 require('xxx'): 須要遞歸查詢路徑上的 node_modules 目錄和全局 node_modules 目錄

• 經過 Module.runMain() 或 --eval 參數: 返回執行命令行的目錄

• 使用相對 / 絕對路徑導入: 這時，直接返回父模塊目錄便可

// lib/module.js

Module._resolveLookupPaths = function(request, parent, newReturn) {
  if (NativeModule.nonInternalExists(request)) {
    debug('looking for %j in []', request);
    return (newReturn ? null : [request, []]);
  }

  // 相似 require('xxx')
  if (request.length < 2 ||
      request.charCodeAt(0) !== 46 ||       // 非 . 開頭
      (request.charCodeAt(1) !== 46         // 非 .. 開頭
        && request.charCodeAt(1) !== 47)) { // 非 / 開頭
    var paths = modulePaths;
    if (parent) {
      if (!parent.paths)
        paths = parent.paths = [];
      else
        paths = parent.paths.concat(paths);
    }

    // Maintain backwards compat with certain broken uses of require('.')
    // by putting the module's directory in front of the lookup paths.
    if (request === '.') {
      if (parent && parent.filename) {
        paths.unshift(path.dirname(parent.filename));
      } else {
        paths.unshift(path.resolve(request));
      }
    }

    debug('looking for %j in %j', request, paths);
    return (newReturn ? (paths.length > 0 ? paths : null) : [request, paths]);
  }

  // 執行 Module.runMain() 進入，此時 request 是絕對路徑

  // with --eval, parent.id is not set and parent.filename is null
  if (!parent || !parent.id || !parent.filename) {
    // make require('./path/to/foo') work - normally the path is taken
    // from realpath(__filename) but with eval there is no filename
    var mainPaths = ['.'].concat(Module._nodeModulePaths('.'), modulePaths);

    debug('looking for %j in %j', request, mainPaths);
    return (newReturn ? mainPaths : [request, mainPaths]);
  }

  // ...

  var parentDir = [path.dirname(parent.filename)];
  debug('looking for %j in %j', id, parentDir);
  return (newReturn ? parentDir : [id, parentDir]);
};

Module._findPath

使用 Module._resolveLookupPaths() 函數找到模塊可能存在的目錄後，調用 Module._findPath() 函數，遞歸查找模塊

Module._findPath() 函數在查找模塊時，存在如下幾種狀況:

• require 的是文件 ==> .js / .json / .node

• require 的文件夾 ==> 找 index.js / index.json / index.node

• require 的包 ==> 找 package.json

// lib/module.js

Module._findPath = function(request, paths, isMain) {
  if (path.isAbsolute(request)) {
    paths = [''];
  } else if (!paths || paths.length === 0) {
    return false;
  }

  // 計算 cacheKey
  // 對於同一模塊，每一個目錄的 cacheKey 不一樣
  var cacheKey = request + '\x00' +
                (paths.length === 1 ? paths[0] : paths.join('\x00'));
  var entry = Module._pathCache[cacheKey];
  // 有緩存就走緩存
  if (entry)
    return entry;

  var exts;
  var trailingSlash = request.length > 0 &&
                      request.charCodeAt(request.length - 1) === 47; // /

  // For each path
  for (var i = 0; i < paths.length; i++) {
    const curPath = paths[i];
    // 不存在就跳過
    if (curPath && stat(curPath) < 1) continue;
    var basePath = path.resolve(curPath, request);
    var filename;

    var rc = stat(basePath);
    if (!trailingSlash) {
      if (rc === 0) {  // 文件
        if (preserveSymlinks && !isMain) {
          filename = path.resolve(basePath);
        } else {
          filename = toRealPath(basePath);
        }
      } else if (rc === 1) {  // 目錄或 package
        if (exts === undefined)
          exts = Object.keys(Module._extensions);
        // 若是是目錄，則 filename 爲 false
        filename = tryPackage(basePath, exts, isMain);
      }

      // 對應是文件但未給出文件後綴的狀況
      if (!filename) {
        // try it with each of the extensions
        if (exts === undefined)
          exts = Object.keys(Module._extensions);
        filename = tryExtensions(basePath, exts, isMain);
      }
    }

    if (!filename && rc === 1) {  // 目錄或 package
      if (exts === undefined)
        exts = Object.keys(Module._extensions);
      // 若是是目錄，則 filename 爲 false
      filename = tryPackage(basePath, exts, isMain);
    }

    if (!filename && rc === 1) {  // 目錄
      // try it with each of the extensions at "index"
      if (exts === undefined)
        exts = Object.keys(Module._extensions);
      filename = tryExtensions(path.resolve(basePath, 'index'), exts, isMain);
    }

    if (filename) {
      // Warn once if '.' resolved outside the module dir
      if (request === '.' && i > 0) {
        if (!warned) {
          warned = true;
          process.emitWarning(
            'warning: require(\'.\') resolved outside the package ' +
            'directory. This functionality is deprecated and will be removed ' +
            'soon.',
            'DeprecationWarning', 'DEP0019');
        }
      }

      Module._pathCache[cacheKey] = filename;
      return filename;
    }
  }
  return false;
};

上面的 for 循環內有許多重複代碼，能夠優化爲:

var exts = Object.keys(Module._extensions);
  var trailingSlash = request.length > 0 &&
                      request.charCodeAt(request.length - 1) === 47; // /

  for (var i = 0; i < paths.length; i++) {
    const curPath = paths[i];
    // 不存在就跳過
    if (curPath && stat(curPath) < 1) continue;
    var basePath = path.resolve(curPath, request);
    var filename;

    var rc = stat(basePath);
    if (!trailingSlash && rc !== 1) {  // 文件或啥也沒有
      if (rc === 0) {  // 文件
        if (preserveSymlinks && !isMain) {
          filename = path.resolve(basePath);
        } else {
          filename = toRealPath(basePath);
        }
      } else {  // 對應是文件但未給出文件後綴的狀況
        filename = tryExtensions(basePath, exts, isMain);
      }
    }

    if(!filename && rc === 1){  // 目錄或 package
      filename = tryPackage(basePath, exts, isMain);

      // 若是是目錄，則 filename 爲 false
      if (!filename) {
        // try it with each of the extensions at "index"
        filename = tryExtensions(path.resolve(basePath, 'index'), exts, isMain);
      }
    }

    if (filename) {
      // Warn once if '.' resolved outside the module dir
      if (request === '.' && i > 0) {
        if (!warned) {
          warned = true;
          process.emitWarning(
            'warning: require(\'.\') resolved outside the package ' +
            'directory. This functionality is deprecated and will be removed ' +
            'soon.',
            'DeprecationWarning', 'DEP0019');
        }
      }

      Module._pathCache[cacheKey] = filename;
      return filename;
    }
  }

Module._findPath() 函數內部依賴下面幾個方法:

// lib/module.js

function tryPackage(requestPath, exts, isMain) {
  // 讀取目錄下的 package.json 並返回 package.main，沒有則返回 false
  var pkg = readPackage(requestPath);

  if (!pkg) return false;

  var filename = path.resolve(requestPath, pkg);
  return tryFile(filename, isMain) || // package.main 有後綴
         tryExtensions(filename, exts, isMain) || // package.main 沒有後綴
         tryExtensions(path.resolve(filename, 'index'), exts, isMain); // package.main 不存在則默認加載 index.js / index.json / index.node
}

function readPackage(requestPath) {
  const entry = packageMainCache[requestPath];
  if (entry)
    return entry;

  const jsonPath = path.resolve(requestPath, 'package.json');
  const json = internalModuleReadFile(path._makeLong(jsonPath));

  // 沒有 package.json，說明不是一個包
  if (json === undefined) {
    return false;
  }

  try {
    var pkg = packageMainCache[requestPath] = JSON.parse(json).main;
  } catch (e) {
    e.path = jsonPath;
    e.message = 'Error parsing ' + jsonPath + ': ' + e.message;
    throw e;
  }
  return pkg;
}

// given a path check a the file exists with any of the set extensions
function tryExtensions(p, exts, isMain) {
  for (var i = 0; i < exts.length; i++) {
    const filename = tryFile(p + exts[i], isMain);

    if (filename) {
      return filename;
    }
  }
  return false;
}

// check if the file exists and is not a directory
// if using --preserve-symlinks and isMain is false,
// keep symlinks intact, otherwise resolve to the
// absolute realpath.
function tryFile(requestPath, isMain) {
  const rc = stat(requestPath);
  if (preserveSymlinks && !isMain) {
    return rc === 0 && path.resolve(requestPath);
  }
  return rc === 0 && toRealPath(requestPath);
}

function toRealPath(requestPath) {
  return fs.realpathSync(requestPath, {
    [internalFS.realpathCacheKey]: realpathCache
  });
}

經過以上步驟，咱們找到了對應的模塊文件，下面開始加載模塊

加載模塊

經過 Module._resolveFilename() 函數找到具體的模塊文件路徑後，就能夠開始加載模塊了

Module._load() 調用 tryModuleLoad() 函數來加載模塊。tryModuleLoad() 則將 module.load() 函數（ 真正加載模塊的函數 ）包裹在一個 try-finally 塊中

// lib/module.js

function tryModuleLoad(module, filename) {
  var threw = true;
  try {
    module.load(filename);
    threw = false;
  } finally {
    if (threw) {
      delete Module._cache[filename];
    }
  }
}

Module.prototype.load()

對於 Module.prototype.load() 函數來講，模塊文件有三種類型:

• .js: 讀取文件而後調用 module._compile() 編譯執行，這是默認狀況

• .json: 做爲 json 文件讀取

• .node: 直接執行編譯後的二進制文件

// lib/module.js

// Given a file name, pass it to the proper extension handler.
Module.prototype.load = function(filename) {
  debug('load %j for module %j', filename, this.id);

  assert(!this.loaded);
  this.filename = filename;
  // 全局 node_modules 和文件路徑上全部的 node_modules
  this.paths = Module._nodeModulePaths(path.dirname(filename));

  // 經過文件擴展名肯定加載方式，默認採用 .js 的方式加載
  var extension = path.extname(filename) || '.js';
  if (!Module._extensions[extension]) extension = '.js';
  Module._extensions[extension](this, filename);
  this.loaded = true;
};

// Native extension for .js
Module._extensions['.js'] = function(module, filename) {
  var content = fs.readFileSync(filename, 'utf8');
  module._compile(internalModule.stripBOM(content), filename);
};

// Native extension for .json
Module._extensions['.json'] = function(module, filename) {
  var content = fs.readFileSync(filename, 'utf8');
  try {
    module.exports = JSON.parse(internalModule.stripBOM(content));
  } catch (err) {
    err.message = filename + ': ' + err.message;
    throw err;
  }
};

//Native extension for .node
Module._extensions['.node'] = function(module, filename) {
  return process.dlopen(module, path._makeLong(filename));
};

// Run the file contents in the correct scope or sandbox. Expose
// the correct helper variables (require, module, exports) to
// the file.
// Returns exception, if any.
Module.prototype._compile = function(content, filename) {

  content = internalModule.stripShebang(content);

  // 使用下面這個結構包裹
  // (function (exports, require, module, __filename, __dirname) {
  //   ...
  // });
  var wrapper = Module.wrap(content);

  // 在沙箱內生成函數對象
  var compiledWrapper = vm.runInThisContext(wrapper, {
    filename: filename,
    lineOffset: 0,
    displayErrors: true
  });

  // ...

  var dirname = path.dirname(filename);
  var require = internalModule.makeRequireFunction(this);
  var depth = internalModule.requireDepth;
  if (depth === 0) stat.cache = new Map();
  var result;
  if (inspectorWrapper) {
    result = inspectorWrapper(compiledWrapper, this.exports, this.exports,
                              require, this, filename, dirname);
  } else {
    result = compiledWrapper.call(this.exports, this.exports, require, this,
                                  filename, dirname);
  }
  if (depth === 0) stat.cache = null;
  return result;
};

從上面的代碼能夠看出，執行模塊時，模塊中的 require，不是 module.js 中的 require，而是由 internalModule.makeRequireFunction() 生成的一個新的函數對象，其隱藏了 module 的實現細節，方便使用

// lib/internal/module.js

// Invoke with makeRequireFunction(module) where |module| is the Module object
// to use as the context for the require() function.
function makeRequireFunction(mod) {
  const Module = mod.constructor;

  function require(path) {
    try {
      exports.requireDepth += 1;
      return mod.require(path);
    } finally {
      exports.requireDepth -= 1;
    }
  }

  function resolve(request) {
    return Module._resolveFilename(request, mod);
  }

  require.resolve = resolve;

  require.main = process.mainModule;

  // Enable support to add extra extension types.
  require.extensions = Module._extensions;

  require.cache = Module._cache;

  return require;
}

各個模塊文件中 require 的區別:

• 內置模塊（ module.js / fs.js 等 ）: 對應 NativeModule.require 函數，僅供 node 內部使用

• 第三方模塊: 對應 internalModule.makeRequireFunction() 函數的執行結果，底層依賴 Module.prototype.require()，遵循 CommonJS 規範

總結

當執行 const xxx = require('xxx') 這段代碼時

1. 先根據 'xxx' 和模塊所在目錄得出被 require 的模塊可能存在的目錄 - Module._resolveLookupPaths

2. 再根據 'xxx' 和 1 的結果得出被 require 的模塊的文件路徑 - Module._findPath

3. 而後根據其拓展名肯定加載方式 - Module.prototype.load

4. 最後將 module.exports 導出

參考：

• https://github.com/nodejs/node/blob/master/lib/module.js

• https://github.com/nodejs/node/blob/master/lib/internal/module.js