> maybe we should hard code into the is.dom property to not include IE5/4
> windows (I think IE5 windows passes the test, but it's not DOM compliant)
No, it fails. Well my version does anyway. Anyone got IE5.5? IE5 Mac passes.
I believe the test should be written in such a way that if a browser
supports enough of the DOM, it should use the DOM code rather than it's own
proprietry way of doing things. It is possible that if we have a DOM branch
for a lot of the core API, things like Mac IE should use the DOM code,
rather than the IE code. It will probably work better (it's like the DOM
than IE5Win as far as I can tell). I don't have immediate access to a Mac,
so unfortunately I can't test my theory ;-)
> this way we can replace some of the is.def checks with is.dom ..
I think is.dom would be the way to go, when more browsers support the
standards they _should_ work with the existing dom code. I've never looked
at IE6 Beta, but perhaps it supports enough of DOM1 to pass the test.
Part of my blabbering is going on about different levels of dom support. My
reasoning behind this is, again Mac IE supports enough of DOM1 to do layer
creation DOM style, and finding inline layers etc. It doesn't support DOM2
events though. It should use the old IE way of doing things here. It
possible the next version of opera will do createElement and addChild, but
not addEventListener. DOM1 but not DOM2.
As these horrible 4 browsers slowly get replaced with browsers that support
DOM standards, we won't actually have to touch the code much, because the
new browers pass the DOM tests, and use the DOM code branches. I'm getting
into one of these "where should the API go now" kind of threads again aren't
I. The think is, we use DOM methods for a lot of it anyway, it doesn't need
to change much, just have solid comliance tests, and branching on is.dom
rather than is.ie or is.ns where possible.
Am I making sense?
> Pascal Bestebroer ([EMAIL PROTECTED])
> Software ontwikkelaar
> Oberon Informatiesystemen b.v.
> http://www.oibv.com
>
> > -----Oorspronkelijk bericht-----
> > Van: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED]]Namens Cameron Hart
> > Verzonden: dinsdag 13 maart 2001 15:32
> > Aan: [EMAIL PROTECTED]
> > Onderwerp: RE: [Dynapi-Dev] DynAPI X
> >
> >
> > I was just looking at the mouseevent code, and I've got a suggestion.
> > Instead is going (is.def&&!is.ie), why don't you say is.ns6
> > OR create a new
> > is.dom2 test or is.dom2e (e for events). I'm think NS6 is
> > probably the only
> > browser that does DOM2 events at the moment anyway, but
> > providing a dom2e
> > test would allow future DOM2 compliant browsers to run this code.
> >
> > I'm figuring there are different levels of the DOM, so we
> > should test for
> > these and use them where appropriate. At the moment (since I
> > just changed
> > it), the is.dom test test's for DOM1 stuff, IE5 Mac should
> > now pass this
> > test (but Opera won't). There is nothing in the DOM that lets
> > you ask the
> > browser, "are you dom1 compliant", so I'm just testing for
> > DOM stuff that is
> > used by the API.
> >
> >
> > _______________________________________________
> > Dynapi-Dev mailing list
> > [EMAIL PROTECTED]
> > http://lists.sourceforge.net/lists/listinfo/dynapi-dev
> >
>
>
> _______________________________________________
> Dynapi-Dev mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/dynapi-dev
_______________________________________________
Dynapi-Dev mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/dynapi-dev
