- mcclim-devel - mailman3.common-lisp.net

[mcclim-devel] repaint-sheet, +everywhere+ and output recording
by Clemens Fruhwirth 02 Jan '06

02 Jan '06

(repaint-sheet instance-inheriting-from-output-recording-stream +everywhere+) does not work. The reason is %handle-repaint in recording.lisp. It rewrites region to a bounding-rectangle regardless the type of the region. Hence, %handle-repaint needs to check whether region is region-equal to +nowhere+ or +everywhere+, the only unbounded region types that do not participate in the bounding rectangle protocol. Please comment on my patch, because I personally don't like it. It seems a bit ackward to make explicit checks for these two regions, probably there are better solutions. If you need something to reproduce the bug, try http://paste.lisp.org/display/15284#3 - uncomment standard-output-recording-stream from the long list of superclasses that imitate those of application-pane to reproduce the error. -- Fruhwirth Clemens - http://clemens.endorphin.org for robots: sp4mtrap(a)endorphin.org

1 0

[mcclim-devel] [patch] fix typos and postscript backend
by Tim Daly Jr. 30 Dec '05

30 Dec '05

Hi guys, This is a small patch to fix some typos and add output recording for NEW-PAGE in the postscript backend. When I tried to run CLIM-DEMO::POSTSCRIPT-TEST, it produced an invalid postscript file. Ghostscript failed to load it, complaining of a dictionary stack underflow. The reason was that the output from NEW-PAGE was at the top of the file, and not embedded between pages as it was meant to be. This is because WITH-OUTPUT-TO-POSTSCRIPT-STREAM uses output recording, (Perhaps this was introduced to make EPS work? Just a guess.), yet NEW-PAGE draws immediately instead of being recorded. I attempted to add output recording for NEW-PAGE. This patch fixes the postscript test mostly -- page 7 is translated to the lower right corner for some reason, but it's at least syntactically correct. I have no understanding of this stuff, so please check whether I've done something boneheaded. Thanks, Tim

2 1

[mcclim-devel] [patch] fix apparent typos
by tim＠tenkan.org 30 Dec '05

30 Dec '05

In my travels through the looking-pane, I bumped into these: -- -Tim

2 1

[mcclim-devel] Improved patch for with-output-as-gadget
by Paul Werkowski 30 Dec '05

30 Dec '05

Here is an improved patch that fixes problems in an earlier post with incorrect cursor position when incremental redisplay was active. It is still required that calls to with-output-as-gadget be wrapped in (updating-output ( ... :cache-value <some-constant>)) to avoid problems if the output-record is ever erased. Happy New Year! Paul

1 0

with-output-as-gadget (was Re: [mcclim-devel] Missing realizer in call to MAKE-PANE-1)
by pw＠snoopy.mv.com 28 Dec '05

28 Dec '05

I wrote: > I now cannot get a radio-box to appear at all, even > at the frame definition eg > > (:panes (rb (with-radio-box () :a :b :c))) > > so maybe something has regressed in that area? > No, it seems that radio/check boxes want only strings in the :choices list. I'm not sure the spec limits the items to strings - at least Lispworks CLIM is more inclusive. Anyhow, there still seems to be a problem in the with-output-as-gadget setup. I end up in the debugger with some layout function calling sheet-native-transformation on the parent of the check-box-pane which is NIL at this time. Anyone got a clue as who is responsible for providing a parent for this thing? Paul

3 7

[mcclim-devel] Re: [mcclim-cvs] CVS update: mcclim/dialog.lisp
by Paul Werkowski 26 Dec '05

26 Dec '05

> In method > #<STANDARD-METHOD LISP::PCL-CLOSE (CLIM-INTERNALS::WINDOW-STREAM) {2846A855}>: > No next method for arguments > (#<CLIM-INTERNALS::WINDOW-STREAM "(Unnamed Pane)" {58A56285}> NIL). > [Condition of type PCL::NO-NEXT-METHOD-ERROR] |That does not happen on sbcl. The last line of climi::close in |panes.lisp, the call (call-next-method) is the problem, as currently |there is no such next method. The following patch helps but I'll not |commit it as there might be such a method in the future. See The solution here is (when (next-method-p)(call-next-method))Paul

1 0

[mcclim-devel] with-output-as-gadget, again
by Paul Werkowski 24 Dec '05

24 Dec '05

Max, When you were researching this, did you notice that there is no problem with gadgets that have direct drawing methods, such as label-pane and toggle-button-pane. I wonder if there is missing replay-output-record method for the classes that have children (check-box radio-box scroller, etc). Paul

1 0

Re: with-output-as-gadget (was Re: [mcclim-devel] Missing realizer in
by pw＠snoopy.mv.com 23 Dec '05

23 Dec '05

> Anyhow, there still seems to be a problem in the > with-output-as-gadget setup. I end up in the debugger > with some layout function calling sheet-native-transformation > on the parent of the check-box-pane which is NIL > at this time. Anyone got a clue as who is responsible > for providing a parent for this thing? The problem here is caused by the call to allocate-space in the initialize-instance :after on gadget-output-record. Not calling that results in no errors and a check-box gadget being displayed. I have no idea if this is an appropriate fix but hopefully this will provide food for thought for those that are in the know. Paul

1 0

Re: [mcclim-devel] Missing realizer in call to MAKE-PANE-1
by pw＠snoopy.mv.com 23 Dec '05

23 Dec '05

> Hmm, well you've got me there (sort of). I'm pretty sure that as I reduced > code from the real program that first showed that problem to the minimal > code in my email I did have with-look-and-feel-realization in play with > no apparent difference. I probably had some cockpit trouble at that point. > > Still no cigar though. Now I get a display visible, but no gadget. > > So, back to the debugging pit. Stay tuned. > It seems that if I change from a radio-box gadget to a label-pane things work as they should. I now cannot get a radio-box to appear at all, even at the frame definition eg (:panes (rb (with-radio-box () :a :b :c))) so maybe something has regressed in that area? Paul

1 0

[mcclim-devel] Reading in a list of objects
by Troels Henriksen 23 Dec '05

23 Dec '05

I have defined a presentation-type, column, as well as the appropriate present and accept methods. I need to accept a list of column objects from the user, but I am a bit stumped on how to do this. Do I have to define a new "columns" type, or does McCLIM include facilities to automatically read in sequences of objects from the user? -- \ Troels "Athas" Henriksen /\ sigkill.dk/blog (Danish)

2 1