[mcclim-devel] help needed with the beagle backend
Error in process Listener(3): Cannot IMPORT "SET-PORT-KEYBOARD-FOCUS" from package "CLIMI" While executing: CCL::OPERATION-ON-ALL-SPECS Type :GO to continue, :POP to abort. If continued: Ignore attempt to IMPORT "SET-PORT-KEYBOARD-FOCUS" from
Hi there. I have been trying to get the beagle back-end for mcclim running. I am using use openmcl 0.14.2-p1, mac os x 10.3.8, and a fresh cvs of mcclim taken this morning. The clx back-end runs nicely -thanks to the splendid effort of the mcclim team. My first attempt consisted of executing the statement (asdf:operate 'asdf:load-op 'beagle). Even when repeated in an OpenMCL Listener, the result was: package "CLIMI" Type :? for other options. 1 > I then examined the documentation in the Backends/beagle directiory of the mcclim (cvs) distribution. The load files essentially perform the following steps inside the OpenMCL Listener window: (load "/opt/local/share/openmcl/0.14.2-p1/tools/asdf") (pushnew "/Users/tom/.asdf-install-openmcl-dir/systems/" asdf:*central-registry*) (asdf:operate 'asdf:load-op :asdf-install) (asdf:operate 'asdf:load-op 'clx) (asdf:operate 'asdf:load-op :mcclim) (asdf:operate 'asdf:load-op :clim-examples) (asdf:operate 'asdf:load-op :clim-listener) (asdf:operate 'asdf:load-op :beagle) This is equivalent to my current ~/.openmclrc file, except that all of this is stated to be done inside the OpenMCL Listener, and the final statement. Upon pasting the above sequence of statements into an OpenMCL Listener the same error results. Does anyone know if this error is a genuine bug, or is there something else missing in the loading sequence I describe above. I am aware that the beagle back-end is experimental, but my aim is to be able to reproduce some of the screen-shots of mcclim using the beagle back-end which available on the web. Thanks.
Tom Winchester <tom.winchester@internode.on.net> writes:
I have been trying to get the beagle back-end for mcclim running. I am using use openmcl 0.14.2-p1, mac os x 10.3.8, and a fresh cvs of mcclim taken this morning. The clx back-end runs nicely -thanks to the splendid effort of the mcclim team.
My first attempt consisted of executing the statement (asdf:operate 'asdf:load-op 'beagle). Even when repeated in an OpenMCL Listener, the result was:
Error in process Listener(3): Cannot IMPORT "SET-PORT-KEYBOARD-FOCUS" from package "CLIMI"
I believe that the backend protocol changed recently, and this symbol was renamed to %set-port-keyboard-focus. Try changing uses of it in the beagle backend (and adding a timestamp keyword argument to the method definition). Cheers, Christophe
On 28/02/2005, at 5:35 PM, Christophe Rhodes wrote:
Error in process Listener(3): Cannot IMPORT "SET-PORT-KEYBOARD-FOCUS" from package "CLIMI"
I believe that the backend protocol changed recently, and this symbol was renamed to %set-port-keyboard-focus. Try changing uses of it in the beagle backend (and adding a timestamp keyword argument to the method definition).
Thanks for the tip. I have done this (events.lisp, and package.lisp), and the listener comes up as stated. It also has the problems discussed eleswhere, but that is another story. Regards.
Tom Winchester <tom.winchester@internode.on.net> writes:
On 28/02/2005, at 5:35 PM, Christophe Rhodes wrote:
Error in process Listener(3): Cannot IMPORT "SET-PORT-KEYBOARD-FOCUS" from package "CLIMI"
I believe that the backend protocol changed recently, and this symbol was renamed to %set-port-keyboard-focus. Try changing uses of it in the beagle backend (and adding a timestamp keyword argument to the method definition).
Thanks for the tip.
I have done this (events.lisp, and package.lisp), and the listener comes up as stated. It also has the problems discussed eleswhere, but that is another story.
Since you've tested it, could you send a patch to this list? As you may be aware, the mcclim developers are attempting to put a release together for this weekend, and I'm pretty sure that elementary build fixes are in scope. Cheers, Christophe
Hi there. I'm sending to the list the patches that Christopher has prodded me into generating. They are the output of diff. The patches attached when applied to the files in Backends/beagle/events.lisp, and Backends/beagle/package.lisp work in the sense that on my system (mac os x 10.3.8, openmcl from darwinports) the nerw files compile, and produce clim windows similar to ones thrown up by a google search. They were tested by pasting the following lines into an OpenMCL Listener (use (require "cocoa") in a command line openmcl, or launch the .app version): ;;;; Load asdf. #-:asdf (load "/opt/local/share/openmcl/0.14.2-p1/tools/asdf") (pushnew "/Users/tom/.asdf-install-openmcl-dir/systems/" asdf:*central-registry*) #-:asdf-install (asdf:operate 'asdf:load-op :asdf-install) ;;;; Load mcclim. (asdf:operate 'asdf:load-op :mcclim) (asdf:operate 'asdf:load-op :clim-examples) (asdf:operate 'asdf:load-op :clim-listener) (asdf:operate 'asdf:load-op :clouseau) ;;;; Load the beagle backend (asdf:operate 'asdf:load-op :beagle) (setf climi::*default-server-path* :beagle) The listener is launched by (clim-listener:run-listener :new-process t). I include this description of the invocation as the method in the README file in the beagle directory has hard coded paths (mentioned in a page google threw up), and clearly precedes the recent introduction of the mcclim.asd file. Of course, as it stands the beagle backend is not that functional, but that is another story. Regards. On 01/03/2005, at 5:52 PM, Christophe Rhodes wrote:
Tom Winchester <tom.winchester@internode.on.net> writes:
On 28/02/2005, at 5:35 PM, Christophe Rhodes wrote:
Error in process Listener(3): Cannot IMPORT "SET-PORT-KEYBOARD-FOCUS" from package "CLIMI"
I believe that the backend protocol changed recently, and this symbol was renamed to %set-port-keyboard-focus. Try changing uses of it in the beagle backend (and adding a timestamp keyword argument to the method definition).
Thanks for the tip.
I have done this (events.lisp, and package.lisp), and the listener comes up as stated. It also has the problems discussed eleswhere, but that is another story.
Since you've tested it, could you send a patch to this list? As you may be aware, the mcclim developers are attempting to put a release together for this weekend, and I'm pretty sure that elementary build fixes are in scope.
Cheers,
Christophe
On 2005-03-01, Tom Winchester <tom.winchester@internode.on.net> wrote:
I'm sending to the list the patches that Christopher has prodded me into generating. They are the output of diff.
Thanks, I committed the two patches. Unfortunately, I lack the Mac to test them, so testing from other McCLIM users would be appreciated (-: Cheers, -- Andreas Fuchs, <asf@boinkor.net>, asf@jabber.at, antifuchs
participants (3)
-
Andreas Fuchs -
Christophe Rhodes -
Tom Winchester