Any instance of the string 'in' in a collapse's class attribute breaks Collapse

[lcannon]

Presumably this would be fixed by changing the regex to match /\bin\b/ instead of /in/ in the toggle function.

[thednp]

Do you have a test scenario with the issue please?

[lcannon]

Sure. The situation that brought this up was a collapse

<button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".navbar-warning-collapse">

that targeted this

<div class="navbar-collapse collapse navbar-warning-collapse">
    ...
</div>

Because /in/ matches against 'navbar-warning-collapse' the toggle method always tries to close the collapse.

[lcannon]

Thanks for the library, by the way, was exactly what I was looking for!

[thednp]

Does the \bin RegExp work with IE8? I will probably have to update all code to this, I didn't expect other cases could be touched.

[thewisenerd]

this issue still isn't fixed.

1. \bin: http://regexr.com/3euta
2. \bin\b: http://regexr.com/3eutm
3. (^|\s)in($|\s): http://regexr.com/3eutg

1. matches words like navbar-inverse-collapse and inverse-class
2. matches words like navbar-in-example
3. checks if string begins in or there is a space before in, and ensures that either string ends with in or a space comes after.

edit: cc @thednp @lcannon

[thednp]

I think we will have a new hasClass(element,className) utility to work with a proper regex.

[thewisenerd]

@thednp cool!

at this point i have to manually patch javascript because i have a div with class name navbar-inverse-collapse :P

hoping this gets patched soon.

[dgrammatiko]

Why not use classList.contains('class-name') and polyfill ie?

[thednp]

To keep polyfill dependency to as low as possible. I noticed some users haven't even heard of them.

[dgrammatiko]

@thednp according to Microsoft: https://www.microsoft.com/en-us/WindowsForBusiness/End-of-IE-support
EOL is in 3 weeks for IE 9, 10.
I guess people could either include ft polyfill service: https://polyfill.io/v2/docs/
or your version pumped with classList.
For me is better to have a clean ES5 code (and polyfill whatever missing) than try to make it "one fits all". But that's my 2c...

[RyanZim]

I think we're all looking forward to bootstrap v4; then we won't have to support IE8.

[dgrammatiko]

@RyanZim BS v4 just dropped IE 9: twbs/bootstrap#21387

[thednp]

yes, our v4 version will be IE9+ of course.

[RyanZim]

@DGT41 Thanks for pointing that out. If bootstrap is dropping IE 8 & 9, we should too IMO @thednp.

[thednp]

@RyanZim we only need a proper Event.prototype + dispatcher polyfill for IE9+, so why just not, the same polyfill is required for IE10+ and I believe also for Ms Edge. (for future BSN4 I mean)

[RyanZim]

@thednp BS v4's css will be totally broken in IE9, why support IE9 for the JS?

[thednp]

https://v4-alpha.getbootstrap.com/migration/

Dropped IE8 and iOS 6 support. v4 is now only IE9+ and iOS 7+. For sites needing either of those, use v3.
Added official support for Android v5.0 Lollipop’s Browser and WebView. Earlier versions of the Android Browser and WebView remain only unofficially supported.

But we are talking about the alpha, we don't know what's on the table with the RC/stable version.

[RyanZim]

Hosted docs are out of date with the alpha development branch. Actual latest docs are here: https://github.com/twbs/bootstrap/blob/v4-dev/docs/migration.md#browser-support.

Dropped IE8, IE9, and iOS 6 support. v4 is now only IE10+ and iOS 7+. For sites needing either of those, use v3.

[thednp]

Ah great, so people will have to fork or look for other frameworks ;)

Off topic: I have had some ideas on hot to make a better HTML5 framework with plain JS some grids and few more components.

[dgrammatiko]

If you don't mind I will create some custom components based on your code (@thednp )

[thednp]

Why should I? As long as the MIT license is all the way, we're all happy :)

[dgrammatiko]

Off topic: I have had some ideas on hot to make a better HTML5 framework with plain JS

@thednp there is https://frend.co but doesn't fulfilled most cases...

[thednp]

Indeed, that's a bit too "one purpose" to me.

In my mind I think of a good typography + an improved&simplified bsn + some grid + some content light design.

[thednp]

I was thinking about it and I wish to have a final note in regards to @RyanZim & @DGT41 remarks: our bsn here is for BS3 and will continue to support older browsers, just as the original script. The future version BSN will be also like BS4 for browser support.

@thewisenerd I started implementing a hasClass(element,classNAME) utility in all scripts, I'll commit the new code later today.

Happy New Year!!

[dgrammatiko]

Happy new year @thednp 🎉

Reasonable decision, I like it!