Alexander Lazareviæ <Alexander at Lazarevic.de> wrote:

>  
> Tim Rowledge schrieb:
> 
> | Nice start but we need to correlate event numbers to avoid clashes
> | with the Ffenestri host windows plugin...
> 
> That should be easy. Just say a number n and I will use n to n+2. Or
> I'll have a look at the simple to download Ffenestri sourcecode and
> find n for myself?
Sure - Johm McIntosh put it out on his iDisk I think so it should be
accessible. And it will be in VMMaker soon.

It would be even nicer one day to come up with a mechanism for the
loading of code to work out such stuff for itself. It's not like it
actually matters much what the values are so long as they don't clash.


> I don't think that
> 
> ~    http://laza.homeip.net/svn/squeak/platforms/win32/plugins/StickPlugin
> 
> is too cryptic.
Ah, well it is in my browser because it seems you're using xml and my
browser is xml illiterate... thus I saw nothing readable when clicking
on your link (It's just "]>"). The essence of cryptic, though not your
fault.

> ~    svn co
> http://laza.homeip.net/svn/squeak/platforms/win32/plugins/StickPlugin
> 
> and get all the latest code in one batch.
>
Nice idea _except_ I have no idea how that would screwup (or not?) my
platforms SVN tree I already have. As I recall from playing (fumbling)
with CVS it would probably overwrite important information. Does SVN do
better? Given that I'm in the middle of trying it out I really can't
risk it without reassurance.

Still, now I know the xml problem I can work around it manually this
time. From a really quick read it looks to me that the only thing
preventing this from working as an external plugin is a need to access
sqNextEventPut() form sqWin32Window.c - correct? We can probably fix
that.


tim
--
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Strange OpCodes: PDH: Page to Disk for the Hell of it