The patch is in (I'm just never sure whether Bugzilla sends an e-mail to ccs on patch attachment):
https://bugzilla.mozilla.org/show_bug.cgi?id=414098 Thanks, Matthieu On Jan 25, 7:09 pm, Matthieu Riou <[EMAIL PROTECTED]> wrote: > On Jan 25, 10:02 am, Norris Boyd <[EMAIL PROTECTED]> wrote: > > > On Jan 25, 12:34 am, Matthieu Riou <[EMAIL PROTECTED]> wrote: > <snip> > > > I've started implementing this in Rhino and it's actually not such a > > > big change. I'd be happy to contribute it (just need to implement the > > > lookup methods as well) but would like to know your opinion before. > > > Would you accept this modification to Rhino? > > > Yes, thank you! Just create a bug and cc me on it. You can attach your > > patch. > > Will do. > > > > > > An extension to this (which I don't believe exists in SM) could be, > > > mostly for performance reasons, to support something like: > > > > a.__defineGetter__('[]', function() { print "masked!"; }); > > > > This would override all index accessors, whatever the index. For large > > > arrays it would be much better as you only have a single function > > > definition. > > > > So what do you think about this little proposal? > > > Make sense to me in general, but I want to move in concert with > > SpiderMonkey. Also, I haven't been tracking ES4 and we need to be > > careful that we don't add features that won't be forward compatible. > > Perhaps you could post to the mozilla.dev.tech.js-engine list and see > > what people think there of the change? > > Just posted the proposal there. I've also checked a bit ES4 and there > doesn't seem to be anything remotely similar (dynamic definition of > property accessors). There's even a bug filed by John Resig on the RI > about this (http://bugs.ecmascript.org/ticket/327). > > Thanks! > Matthieu > > > > > > Thanks, > > > Matthieu _______________________________________________ dev-tech-js-engine-rhino mailing list [email protected] https://lists.mozilla.org/listinfo/dev-tech-js-engine-rhino
