Re: Alternative and fast option to some very useful jQuery features?

From	Ben Bacarisse <ben.usenet@bsb.me.uk>
Newsgroups	comp.lang.javascript
Subject	Re: Alternative and fast option to some very useful jQuery features?
Date	2017-03-21 01:33 +0000
Organization	A noiseless patient Spider
Message-ID	<871strpg5j.fsf@bsb.me.uk> (permalink)
References	(3 earlier) <08927503-fc48-4ab2-835f-21a1d0904e9b@googlegroups.com> <87h92psfkr.fsf@bsb.me.uk> <b1217b03-6f5f-4bde-a5a0-614f7d6117a4@googlegroups.com> <878to0qlgx.fsf@bsb.me.uk> <40b0b1df-ba67-4f60-aed4-b66aa822dbe8@googlegroups.com>

Show all headers | View raw

justaguy <lichunshen84@gmail.com> writes:

> On Monday, March 20, 2017 at 6:41:24 AM UTC-4, Ben Bacarisse wrote:
>> justaguy writes:
>> 
>> > On Sunday, March 19, 2017 at 6:53:34 AM UTC-4, Ben Bacarisse wrote:
>> <snip>
>> >> Sure, but the idea is that the browser will simply suggest possible
>> >> items from the list when the user types.  Support for this is not
>> >> universal, of course.
>> >
>> > Quick update.  I tried it with Chrome (ver 56) and Firefox (ver 52).
>> > And I pass the selected value to a js function for further process
>> > with Event of "onInput", both seems to work.  So far Chrome hasn't
>> > crashed but FF just crashed.  Fyi, it's not my js function's problem,
>> > several persons including myself has been testing it for about a week,
>> > browser never crashed.
>> 
>> What does "crash" mean in this case?
>
> Browser: FF (ver 52)
> Err msg:
> "
> Gah. Your tab just crashed.
>
> We can help you!
>
> Choose Restore This Tab to reload page content.
> ...
> "
>
> Condition:
> I changed the event from onInput to onSelect.  I then changed it back,
> however, it still exhibited the same behavior (maybe because of
> cache).  Today, I changed it to onSelect, expectedly it crashed again,
> I changed it back to onInput but it did not crash this time.
>
>> 
>> > It raised concern in me if HTML5 "list" INPUT attribute is ready for
>> > production...
>> 
>> Does it go wrong when there are no scripts at all?  
>
> Well. I have a few js functions.  But as mentioned in my previous,
> these functions have been working perfectly fine for some time.
> Hence, they are not likely the culprit.

I suspect that list attributes have worked perfectly for some time too
(I've used them a lot and not seen an issues so far) so this is not a
compelling argument.  The reason to remove the scripts is to get the
smallest error case you can.

>>If not, why do you
>> think it's the list attribute?  If it does, have you filed a bug?  This
>> would be a significant help to the community.
> Do we still need to file a bug?

I'd argue that nothing should cause that message to appear so it's a bug
one way or another, but if it's due to using nothing but the list
attribute then it's a very important one.

If your test case involves a complex document with lots of scripts, it's
going to take much more effort to track down than a small test case.  It
would be helpful if you can gradually reduce the complexity of the
document until you get as small a test case as you can that still
triggers the crash.  I know this is non-trivial, but it helps everyone.

-- 
Ben.

Thread

csiph-web