Redundant ".call" call

[killmenot]

Hi @robtweed

Please take a look:

ewd-document-store/lib/proto/value.js

Line 43 in 36d15aa

this._set.call(this, value);

I think this._set.call(this, value); can be replaced to this._set(value) here because in getter we use this context normally

[robtweed]

You're probably right that this is unnecessary, but is there a risk of de-stabilising anything elsewhere by changing it? this_set is an "internal" method, not intended for external / widespread use - but we'd need to locate every situation where it's invoked throughout ewd-document-store before changing this invocation

[killmenot]

Initially, I missed cases to check the passed context to different function. However, when I was working on QEWD's unit tests I found that I should test it as well. I wrote some helpers that allows me to test such scenarios. Now, I'm updated other libs that contains call/apply

Back to this snippet.

Initially, I updated the code that tests it to:

it('should set value', function () {
  documentNode.value = 'bar';
  expect(documentNode._set).toHaveBeenCalledWithContext(documentNode, 'bar');
});

The code above makes sure that when _set called the passed context is correct (because we use .call()). Then, I took a look at code again and updated it to this._set(value); . The test passed again.

It's not a big issue for me to have this._set(value); or this._set.call(this, value); . I believe you a owner should use the way you think is more correct based on many different sources of information. My goal is to make sure that lib is covered with test well :)

P.S.
If we keep using .call I need to update should set value test case