总的来说…

想要获取调用 require() 时加载的确切的文件名，请使用 require.resolve() 函数。

综上所述，以下用伪代码的形式来表述 require.resolve 中的高级算法是如何工作的：

  1. require(X) from module at path Y
  2. 1. If X is a core module,
  3.    a. return the core module
  4.    b. STOP
  5. 2. If X begins with './' or '/' or '../'
  6.    a. LOAD_AS_FILE(Y + X)
  7.    b. LOAD_AS_DIRECTORY(Y + X)
  8. 3. LOAD_NODE_MODULES(X, dirname(Y))
  9. 4. THROW "not found"
  11. LOAD_AS_FILE(X)
  12. 1. If X is a file, load X as JavaScript text.  STOP
  13. 2. If X.js is a file, load X.js as JavaScript text.  STOP
  14. 3. If X.json is a file, parse X.json to a JavaScript Object.  STOP
  15. 4. If X.node is a file, load X.node as binary addon.  STOP
  17. LOAD_AS_DIRECTORY(X)
  18. 1. If X/package.json is a file,
  19.    a. Parse X/package.json, and look for "main" field.
  20.    b. let M = X + (json main field)
  21.    c. LOAD_AS_FILE(M)
  22. 2. If X/index.js is a file, load X/index.js as JavaScript text.  STOP
  23. 3. If X/index.json is a file, parse X/index.json to a JavaScript object. STOP
  24. 4. If X/index.node is a file, load X/index.node as binary addon.  STOP
  26. LOAD_NODE_MODULES(X, START)
  27. 1. let DIRS=NODE_MODULES_PATHS(START)
  28. 2. for each DIR in DIRS:
  29.    a. LOAD_AS_FILE(DIR/X)
  30.    b. LOAD_AS_DIRECTORY(DIR/X)
  32. NODE_MODULES_PATHS(START)
  33. 1. let PARTS = path split(START)
  34. 2. let I = count of PARTS - 1
  35. 3. let DIRS = []
  36. 4. while I >= 0,
  37.    a. if PARTS[I] = "node_modules" CONTINUE
  38.    b. DIR = path join(PARTS[0 .. I] + "node_modules")
  39.    c. DIRS = DIRS + DIR
  40.    d. let I = I - 1
  41. 5. return DIRS