Make carousel ignore non-items in carousel-inner

[sdhull]

Seems to me that it should actively ensure it only tries to transition
to things with the item class.

[fat]

hm not keen on this – dont want to add a class dependency if we don't have to

[sdhull]

@fat kind of a bizarre objection considering that Carousel is already thoroughly dependent on the item class so this is not really technically adding a class dependency.

Moreover dismissing a contribution prior to discussion doesn't really encourage collaborators to continue offering help.

[rwaldron]

hm not keen on this – dont want to add a class dependency if we don't have to

Worst reason ever.

...But seriously, this is egregiously irresponsible.

[sdhull]

lol @rwaldron thanks. I agree. Go ahead and check out foundation/foundation-sites#4079 -- this is my similar PR for foundation. It was shortly after this "interaction" with @fat that I decided none of my projects going forward would utilize twbs. To be fair, my PR on foundation was described a bit better. Still doesn't excuse dismissing useful contributions for non-reasons without even discussing first IMO.

Come check out Foundation! I'm a fan so far. 😉

[rwaldron]

We're using Ember and twbs for a client project here @bocoup and @fat's arrogance in this thread is getting in the way of writing a clean implementation. Thanks for that Jacob!

[tkellen]

Chiming in to agree with @rwaldron / @sdhull. This one line change makes a ton of sense and eliminates a slew of possible unexpected behaviors.

[mdo]

@rwaldron "Egregiously irresponsible", "worst reason ever", and "arrogance" seems rather out of line to me. Get off your high horse and realize that folks make can make mistakes. Perhaps we overlooked something, or thought this was another issue or PR. Or there was some other misunderstanding. Or we didn't properly communicate our thoughts on the matter. How does your shitty attitude help anything?

@sdhull Indeed, the different in PRs is striking. It may have made difference here, but who knows. Appreciate you cross linking that for comparison and additional context.

As to addressing this problem, why does this change need to happen? The only children of a carousel should be it's items. What else would come between them that we'd need to scope things to the class name here?

[tkellen]

Template engines that leverage metamorph insert non item children into carousels.

Here is one of numerous issues discussing methods of working around the manifestation of it in Ember:
http://stackoverflow.com/questions/20876938/bootstrap-carousel-not-working-impacted-by-ember-metamorph-identifiers

[sdhull]

@mdo if contributions are dismissed out of hand without an opportunity for discussion, the project is going to miss a lot of good stuff. I understand discussion is hard, however I always thought that it was a part of open source. Sometimes good things are hard.

[tkellen]

Rather than getting into a debate about what could or should have happened here, could we please revisit the contribution itself? It's a small, useful change (with a test case no less!). What needs to be done to see it merged?

[mdo]

@tkellen Regarding metamorph, is there no way around this? Why would there be content inserted that isn't the content you want to be inserted? That aspect sounds like a problem with metamorph and not Bootstrap.

Beyond the JS, what impact does this have on the CSS? Not the proposed changes here, but the fact that a third party is inserting unwanted markup? Sounds like a messy situation.

[mdo]

@sdhull Yup, and generally we try our best to do and enable just that. However, seven months ago, we were knee deep in shipping v3 and didn't have time to spend on every issue and PR. That's not an excuse, but it's important to be mindful of the context and timing of an issue.

[tkellen]

@mdo Unfortunately, no. The script tags that get interspersed are used for two-way binding. From where I am sitting, making the selector slightly more specific seems like a perfect solution. Short of that, some kind of configurable option would be ideal. Thoughts?

[fat]

@sdhull

Moreover dismissing a contribution prior to discussion doesn't really encourage collaborators to continue offering help.

totally agree. that's my bad.

i'll try to get a version of this patch in today. Also, I'll work on coming across less dismissive… didn't mean for that to be the case.

[fat]

also, fwiw  – in the future please be mindful of the coding style… single quotes, semicolons, etc.

thanks!