Hi,
At present, it's only possible to have one backend loaded at once, because STANDARD-POINTER is defined per-backend. This can't be right, can it? On the other hand, the attached is my 'solution' (for the CLX backend; I haven't tried fiddling with Beagle, but this also works for my in-development Null backend), and I'm not wild about it.
Still. Any opinions?
Cheers,
Christophe
csr21@cam.ac.uk wrote:
Hi, At present, it's only possible to have one backend loaded at once, because STANDARD-POINTER is defined per-backend. This can't be right, can it? On the other hand, the attached is my 'solution' (for the CLX
I recall finding this a little odd when I started Beagle; but I (obviously) didn't fix it back then (or since).
backend; I haven't tried fiddling with Beagle, but this also works for my in-development Null backend), and I'm not wild about it. Still. Any opinions?
I can't see any problems with this wrt Beagle. My only opinion is that 'pointer-cursor' should perhaps be a slot in standard-pointer, or just a vanilla GF with a 'warn: unimplemented for type standard-pointer' and the back ends add methods accordingly.
-Duncan
Cheers, Christophe _______________________________________________ mcclim-devel mailing list mcclim-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/mcclim-devel
On Wednesday, June 15, 2005, at 10:18 am, Christophe Rhodes wrote:
Hi,
At present, it's only possible to have one backend loaded at once,
As an aside; I was able to load and execute Listeners running in both the CLX and Beagle backends without observing problems prior to this change, so it's certainly possible to have multiple backends loaded at once. OTOH maybe I was just lucky.
-Duncan
because STANDARD-POINTER is defined per-backend. This can't be right, can it? On the other hand, the attached is my 'solution' (for the CLX backend; I haven't tried fiddling with Beagle, but this also works for my in-development Null backend), and I'm not wild about it.
Still. Any opinions?
<pointer.diff> Cheers,
Christophe _______________________________________________ mcclim-devel mailing list mcclim-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/mcclim-devel
Duncan Rose duncan@robotcat.demon.co.uk writes:
On Wednesday, June 15, 2005, at 10:18 am, Christophe Rhodes wrote:
Hi,
At present, it's only possible to have one backend loaded at once,
As an aside; I was able to load and execute Listeners running in both the CLX and Beagle backends without observing problems prior to this change, so it's certainly possible to have multiple backends loaded at once. OTOH maybe I was just lucky.
I think it's (partly?) because your beagle synthesize-pointer-motion-event method is not specialized; in addition, I was misleading or insufficiently specific in my description. The failure mode I was seeing was that an attempt to use a CLX-backend menu would give an error message about no applicable methods for synthesize-pointer-motion-event given a STANDARD-POINTER.
Thanks,
Christophe
Christophe Rhodes csr21@cam.ac.uk writes:
At present, it's only possible to have one backend loaded at once, because STANDARD-POINTER is defined per-backend. This can't be right, can it? On the other hand, the attached is my 'solution' (for the CLX backend; I haven't tried fiddling with Beagle, but this also works for my in-development Null backend), and I'm not wild about it.
Still. Any opinions?
I've merged this, including a blind modification to the relevant beagle code. I hope I haven't broken anything in the process.
Cheers,
Christophe
On Thursday, June 16, 2005, at 10:26 am, Christophe Rhodes wrote:
Christophe Rhodes csr21@cam.ac.uk writes:
At present, it's only possible to have one backend loaded at once, because STANDARD-POINTER is defined per-backend. This can't be right, can it? On the other hand, the attached is my 'solution' (for the CLX backend; I haven't tried fiddling with Beagle, but this also works for my in-development Null backend), and I'm not wild about it.
Still. Any opinions?
I've merged this, including a blind modification to the relevant beagle code. I hope I haven't broken anything in the process.
Beagle appears to be fine; thanks for making the change.
-Duncan
Cheers,
Christophe _______________________________________________ mcclim-devel mailing list mcclim-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/mcclim-devel
-
Christophe Rhodes -
Duncan Rose -
duncan＠robotcat.demon.co.uk