[whatwg] This step must be skipped if the form has no onreceived attribute

[Ian Hickson]

On Thu, 24 Jun 2004, Jim Ley wrote:
>>>
>>> Could you address the other points About re-serialisation of the DOM
>>> leading to something radically different from what was intended (the
>>> attribute should only be there if there was also an event listener
>>> after a re-serialisation and parse it wouldn't be there.)
>>
>> This is a general problem with event handlers -- any document using script
>> will have different behaviour if you remove all the event handlers.
>>
>> For example, if you take the following HTML4 fragment:
>>
>>   <form> <button> Test </button> </form>
>>
>> ...the behaviour will be radically different if, when the fragment was
>> serialise, the <form> element had an event listener that cancelled any
>> "submit" events.
>
> but that behaviour is "nothing"

The behaviour with script is whatever the script does, and without script
is a form submission. Hardly "nothing".


> which is bad, but it's better than the WF2 suggestion which would give
> you an attribute of onrecieved="chicken" which is likely to error

Eh? Where did the poultry come from. You only need to set the attribute,
not give it a value. Just set:

   onreceived=""

...and your event listener will be used if present, and the form
submission will continue as per normal (although with a slower,
non-incremental, page load).

I don't really understand what you are complaining about.


>> I fail to see how an issue tracker would improve matters in this
>> respect.
>
> Oh right, I do, since everything would have a "this is issue # X and we
> can easily see that it was addressed, and how it was addressed."

You can do that now. Look at the archives in Thread view.


>> the point above about reserialisation isn't relevant since, as
>> described above, it is an issue with event listeners as a whole, not
>> with the onreceived="" attribute).
>
> No, since we're forced in your WF2 to put an attribute in, which doesn't
> make sense unless the script exists but does exist then, normally with
> Event listeners if the script doesn't exist neither do any of the
> dependant attributes, with this method we have the dependant attributes,
> but no script to handle it.

I'm sorry, I have no idea what you're trying to say here. Could you give
an example?


>> If you think I skipped a comment by mistake, I urge you to reraise the
>> issue (as you did here, and as one would have to do with an issue
>> tracker).
>
> But I'll generally know when you don't respond to mine, I don't know
> when you don't respond to other peoples issues, and they may be highly
> relevant, but I can't know, if we have an issue tracker we can see
> that the open issues exist.

Every WF2-related e-mail above:

   http://listserver.dreamhost.com/pipermail/whatwg-whatwg.org/2004-June/date.html#210

...is, as far as I am concerned, a closed issue. (There are Web Apps
e-mails above there that are still open, but those are mainly listing
requirements, not issues.)

So if someone sent a WF2 e-mail and it is above that line, and I haven't
replied to it, and it is still relevant, then I missed it. I couldn't find
any such e-mail just now, but let me know if you find one.

The e-mails after that line are those that I am still replying to.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'