feat: allow passing args to object methods

[haltcase]

Add the ability to pass (string) arguments to object methods in a space delimited format. From what I understand, this would preclude strict adherence to Python's string formatting where spaces would be valid in keys when resolving against the object. ( and I know this has been requested or is potentially planned )

Despite that, I needed the functionality for myself and thought I'd propose it. If nothing else anyone interested can use my fork. I totally understand if it's not merged here.

Examples

( see here for a jsfiddle )

Assume all examples below use this object:

var obj = {
  getFirstName (other) {
    if (other) return 'Billy'
    return 'Alec'
  },
  lastName: 'Baldwin'
}

Prior to this PR, simple object method calls could be made:

'{getFirstName} {lastName} did the thing.'.format(obj)
// -> 'Alec Baldwin did the thing.'

The above will continue to work the same way, but this PR additionally allows for passing arguments to these functions:

'{getFirstName true} {lastName} did the thing.'.format(obj)

// -> 'Billy Baldwin did the thing.'

Note, however, that even though I'm passing true the arguments passed are always strings. The function's if only checks for a truthy value in this simple case but if you were expecting anything else you'd have to cast it yourself ( eg. using Boolean or Number constructors, or parseInt() ).

Given this is a string formatting library, this limitation doesn't seem like much of a limitation.

[haltcase]

I'm seeing the failure on one of the tests and will be looking into that.

  1) format invokes methods:
     TypeError: String.prototype.toLowerCase called on null or undefined
      at toLowerCase (native)
      at lookup (index.js:9:3699)
      at index.js:9:2366
      at String.replace (native)
      at index.js:9:1108
      at Context.<anonymous> (test/index.js:84:8)

[davidchambers]

Thanks for the pull request. I would, though, like to keep the API as close as possible to Python's.

It's worth noting that the results you desire are attainable with the current API:

> '{1} {lastName} did the thing.'.format(obj, obj.getFirstName(false))
'Alec Baldwin did the thing.'

> '{1} {lastName} did the thing.'.format(obj, obj.getFirstName(true))
'Billy Baldwin did the thing.'

[haltcase]

@davidchambers Not a problem, I assumed that would be the case. I can close out the PR if you want and maintain my own separate fork.

For my uses ( in a chat bot ), the template string is going to be set beforehand by end-users knowing they'll use things like {target}, {sender} etc. which will be replaced later. These are simple, but when adding something like {readfile ./path/to/file.txt} I think I need functionality like this PR allows.

edit: I'll ignore the lint issue then, but that's the only thing making it fail now. Ternary line!

[davidchambers]

I can close out the PR if you want and maintain my own separate fork.

It's a nice patch, but I think a fork is a good idea. One day I hope to complete and merge #2, so it's important not to drift from the Python API.

[haltcase]

👍 sounds good.