[mcclim-devel] Re: [mcclim-cvs] CVS update: mcclim/dialog.lisp

[rpgoldman at real-time.com]

>>>>> "mgr" == Max-Gerd Retzlaff <m.retzlaff at gmx.net> writes:

    mgr> Hello
    mgr> On Thu, Aug 25, 2005 at 06:53:49PM -0500, rpgoldman at real-time.com wrote:
    >> >>>>> "mgr" == Max-Gerd Retzlaff <m.retzlaff at gmx.net> writes:
    >> 
    mgr> And the McCLIM user's manuel should contain in its (not yet existing)
    mgr> paragraph on OPEN-WINDOW-STREAM saying that one probably wants to call
    mgr> OPEN-WINDOW-STREAM with :input-buffer as (climi::frame-event-queue
    mgr> *application-frame*)..
    >> 
    >> This raises the obvious question of whether this should somehow be
    >> made the default value of that argument.  Something like 

    mgr> Well, that would be against the CLIM II specification. As already
    mgr> quoted in my last mail it says:

    mgr>    If input-buffer is supplied, it is an input buffer or
    mgr>    event queue to use for the resulting window. Programmers
    mgr>    will generally supply this when they want the new window
    mgr>    to share its input buffer with an existing
    mgr>    application. The default is to create a new input buffer.

Um... but I thought that your earlier message said that the default
behavior dictated by the spec didn't work.  Maybe I didn't fully
understand.  I guess I had thought a spec violation was better than a
spec-obeying busted behavior...

If the spec-dictated behavior really *is* busted, would it be
reasonable to hoist an error on defaulting the argument, and suggest
that the user do the spec-defying but workable thing?

Best,
R