Query regarding coding paradigm used in this library.

Why in the code string concatenation is used to create the function? I can see, this approach is being followed in almost every file. Why shouldn’t it be written as a normal function? For example like this in HookCodeFactory .

switch(this.options.type) { case "sync": return new Function(this.args(), "\"use strict\";\n" + this.header() + this.content({ onError: err => throw ${err};\n, onResult: result => return ${result};\n, onDone: () => "", rethrowIfPossible: true }));

[ooflorent]

Sure! Let’s consider the following functions; they are really simple hook factory implementations:

const f = (...fns) => (...args) => {
  for (const fn of fns) {
    fn(...args)
  }
}

const g = (a, b, c) => (x, y) => {
  a(x, y)
  b(x, y)
  c(x, y)
}

Both functions achieve the same thing but g is more specific. While f accepts any number of callbacks and applies any number of arguments on them, g takes 3 callbacks and calls them using 2 arguments. Because g is specialised for this use case, it will perform much better than f.

Keep in mind that tapable is used by webpack. The created hooks can hold dozens of callbacks with multiple arguments depending on the event. Specialised functions like g could not realistically be used to cover all callsites. This is the reason why hooks are generated: this way it is possible to define a specialised version where loops are unrolled to improve performances.

[ooflorent]

Here is a good explanation: What’s up with monomorphism?