> Great, this is the best thing. It seems that the attribute
> pageYOffset is undefined in IE. I don't really know what it is
> supposed to do and it seems to be only used to update the
> window.status, so maybe we can just avoid it if the browser is IE
> (using some test like /'if (navigator.appVersion.indexOf("MSIE")!
> =-1)'/ ) or try to find some equivalent to this property in IE (I
> have tried the following link, but without much success: http://
> msdn2.microsoft.com/en-us/library/ms535873.aspx).
We'd have to find some equivalent. The window.status is misleading
-- That's the protocol by which the page communicates with WOLips (I
listen for window status change events in WOLips and parse them to
find out what the browser wants to do). This fix is necessarily for
outline view not sucking :) Basically when you make a change to your
component, we regenerate HTML for the view. Without something
equivalent to this code, the scrollbar would reset every reload.
This fix records where you are looking before each reload, reloads,
then repositions the view to where you left off, so you can't tell
that the entire document is reloading each time. I'll probably skip
it temporarily on Windows just so you can use it at all, but we will
need a proper fix eventually. Incidentally Pascal pointed out that
that document looks like ass on FireFox, so if anyone is using
Eclipse on Linux (which is FFX by default in the SWT Browser
component), it's basically unusable.
ms
This archive was generated by hypermail 2.0.0 : Fri Oct 12 2007 - 15:16:09 EDT