jessyink-team team mailing list archive

[Hannes Hochreiner <hannes@xxxxxxxxxxxxxx>]

Hi Egor,

On 7 Jun 2009, at 12:00, Egor Zindy wrote:

> Hey Team,
>
> I had a look at key events for the past hour, and across browsers,
> things look really messed-up.
>
> http://ejohn.org/blog/keypress-in-safari-31/
> http://unixpapa.com/js/key.html
>
> I'm not even sure what I took back from reading all this. Was webkit's
> decision of not sending special characters via keypress justified?
> keypress is bad? use keydown instead? how can I find the ascii value
> of the key then... all very confusing. Like the example given, nothing
> garanties (with keydown) that when you press SHIFT + 4, it'll be the $
> sign character you expect.
>
> Some people do this:
> http://dev.rubyonrails.org/attachment/ticket/9666/event_keypress.diff
> if (eventName == 'keypress' && Prototype.Browser.WebKit) return  
> "keydown";
>
> But then some people say you shouldn't.
> http://github.com/sstephenson/prototype/blob/ 
> bfd5353cbfa8b344d959d28c4e0cc86cda61cbc4/src/event.js
>
> The above from prototypejs seems to be the best solution by the way...
> if I'm not mistaken.

Hmm, that's really a quite messed up situation. I am not completely  
sure what the best solution is. I quickly browsed the links and I  
agree with you, the prototype approach might be the best solution.  
Also, considering the experience the guys working on prototype must  
have, I think it's a save option to pick.

> Anything we can lift from there? I'll keep working at it as proper
> (cross-browser) key handling in JessyInk is pretty much critical...
> especially with the "custom keys" targeted for release 1.3........

Completely with you on that. Cross-browser key handling is vital for  
JessyInk. If you want, I can try and work on the system that lets the  
user assign the keys (based on your extension and what we discussed  
so far), so you can really dig into the event handling. Another  
solution is to fix the event handling with the next release and leave  
the customisation for 1.4.

Cheers,
Hannes