Update notes.xsl

[AurelioDeRosa]

Fix issue #586.

[arthurvr]

The data-doesnt-accept-undefined note does also appear on the jQuery.data post. Of course, jQuery.data doesn't return the jQuery object is is called on.

[AurelioDeRosa]

@arthurvr In this case I vote for splitting the two notes as they have a different behavior. If everyone agrees I'll modify the PR to include an additional note specific for jQuery.data().

[arthurvr]

Seems fine.

[dmethvin]

Seems like we need to split the two, I agree. How do you guys feel about undefined not actually setting the value?

recognised

Have we standardised on UK English? Oh I mean, standardized. 😄 I guess I never noticed before.

[AurelioDeRosa]

To be honest I feel more like .data('thing', undefined) should act as .removeData('thing'). We had a similar discussion in #523.

P.S. British English vs American English: their differences will kill me.

[dmethvin]

I had forgotten about #523. It's tough because I think a good argument can be made either way. I agree with you that removing the data item feels best, but it would certainly be a breaking change as @gibson042 pointed out. It's hard to know whether a lot of plugins and developers depend on the current no-op behavior.

[AurelioDeRosa]

@dmethvin What if we introduce this BC in jQuery 3.0?

[dmethvin]

If very few people depend on this (or those people don't upgrade) then it would be nice to eliminate this kind of strange exception. If it turns out that several high-profile popular plugins break because we change this, then it would be a real pain to do and it would prevent people from upgrading to 3.0. I don't know which applies. 😄

[AurelioDeRosa]

@dmethvin we're all in your hands :)

[dmethvin]

I'll leave the hard decisions to @timmywil now! Sucker!

[arthurvr]

@AurelioDeRosa Besides from the discussion above, this PR needs some updates. Would you like to do them? I'd love to land this.

[AurelioDeRosa]

@arthurvr I was waiting for a feedback by @timmywil.

[timmywil]

Sorry I missed this. I think this can be landed. As far as #523, I think we'll need to gauge how much it will break. Personally, I always saw undefined as it's own value and thought it was most intuitive for .data('something', undefined) to act similarly to assigning a property on an object to the value undefined, and .removeData('something') being the equivalent of using the delete operator. Otherwise, determining whether something was set to undefined or whether it was never set is impossible. Regardless, we can thrash on that issue there.

[arthurvr]

I can def. agree with @timmywil.

[AurelioDeRosa]

So, I understand the conclusion is "Let's leave things as they are in regard of undefined and split the note in two notes". I'll update my PR accordingly.

[arthurvr]

Thank you, @AurelioDeRosa!

[arthurvr]

I'll update my PR accordingly.

@AurelioDeRosa You've got the chance to update this PR? Would love to land it.

[arthurvr]

ping @AurelioDeRosa

[AurelioDeRosa]

PR updated @arthurvr

[arthurvr]

recognised

We use American English. Could you fix that on the while in the jquery-data-doesnt-accept-undefined note?