Hi John,
I pushed a patch that might fix your problem. Now the WM_SYSCOMMAND
message is parsed for window minimize, maximize, and restore, and the
window drawing is turned off when it's minimized.
To test, please bootstrap with -exclude=io because file monitors are
broken on windows right now.
./factor -i=boot.x86.32.image -exclude=io
I think this patch doesn't cover two cases: windows key + d, windows
key + m. There doesn't seem to be a WM_SYSCOMMAND on these events,
so it probably arrives in another message?
You might also consider upgrading your video card driver. Which video
card/driver do you have?
Thanks,
Doug
On Apr 7, 2008, at 8:24 PM, John Pallister wrote:
> Righto, the call stack is now:
>
> [ call -> stop ]
> [
> ui-notify-flag get lower-flag [ update-ui ] ui-try
> -> update-ui-loop
> ]
> [ swap >c call -> c> drop ]
> [
>> r datastack r>
> swap slip ->
>> r datastack r>
> 2dup [ length ] compare sgn {
> { -1 [ trim-datastacks nip relative-underflow ] }
> { 0 [ 2drop ] }
> { 1 [ trim-datastacks drop relative-overflow ] }
> } case
> ]
> [ >r call -> r> ]
> [ iterate-step -> iterate-next (each-integer) ]
> [ swap >r 2dup 2slip -> r> swap ]
> [ >r >r call -> r> r> ]
> [ >r associate >n r> call -> ndrop ]
> [ swap >c call -> c> drop ]
> (draw-world)
> flush-gl-context
> (win32-error)
> throw
> [
> self error-thread set-global
> continuation -> error-continuation set-global rethrow
> ]
>
> This is the same as the one I reported previously (apart
> from 'flush-gl-context' of course).
>
> The data stack is also the same:
>
> 258
> "The wait operation timed out."
>
> The retain stack I reported incorrectly last time, it's
> actually:
>
> { }
> 1
> [ ~vector~ nth-unsafe dup update-hand draw-world ]
> 0
>
> Hope this helps,
>
> John :^P
>
> Slava Pestov wrote:
>> Hi,
>>
>> It sounds like of the following two methods are failing:
>>
>> M: windows-ui-backend select-gl-context ( handle -- )
>> [ win-hDC ] keep win-hRC wglMakeCurrent win32-error=0/f ;
>>
>> M: windows-ui-backend flush-gl-context ( handle -- )
>> win-hDC SwapBuffers win32-error=0/f ;
>>
>> To find out which one, paste the following in your listener before
>> minimizing:
>>
>> M: windows-ui-backend select-gl-context ( handle -- )
>> "entering select-gl-context" show
>> [ win-hDC ] keep win-hRC wglMakeCurrent win32-error=0/f
>> "leaving select-gl-context" show ;
>>
>> M: windows-ui-backend flush-gl-context ( handle -- )
>> "entering flush-gl-context" show
>> win-hDC SwapBuffers win32-error=0/f
>> "leaving flush-gl-context" show ;
>>
>> Hit all restarts as they come up. Then try minimizing the window, and
>> look at the traceback window again. The 'show' calls have the effect
>> of inhibiting tail call optimization so you'll see which function
>> called win32-error=0/f.
>>
>> Slava
>>
>> On Apr 7, 2008, at 7:22 PM, John Pallister wrote:
>>
>>> Hello list,
>>>
>>> I have downloaded this latest build, and it gives me the
>>> same error I reported last week.
>>>
>>> Initially the IDE seems to work well, but when I minimise
>>> the window I get an "Error" window that says:
>>>
>>> An error occurred while drawing the world T{ world ~track~ t
>>> f f "Factor workspace" ~model~ ~hashtable~....
>>> This world has been deactivated to prevent cascading errors.
>>> The wait operation timed out.
>>>
>>> After that, it's game over. This is on XP SP2 with Visual
>>> C++ Express 2008 installed (FWIW). Offhand, I can't think of
>>> anything else that might be interfering with Factor. It runs
>>> e.g. the Second Life client OK.
>>>
>>> In other news, this new Windows build has a blank console
>>> window that was absent from the Cygwin build.
>>>
>>> Cheers,
>>>
>>> John :^P
>>>
>>> Slava Pestov wrote:
>>>> Hi all,
>>>>
>>>> An automatically built Windows/x86 binary was just uploaded:
>>>>
>>>> http://factorcode.org/downloads/winnt-x86-32/
>>>>
>>>> As usual, this means all libraries now load and all unit tests
>>>> pass.
>>>>
>>>> Slava
>
> --
> John Pallister
> [EMAIL PROTECTED]
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
> Register now and save $200. Hurry, offer ends at 11:59 p.m.,
> Monday, April 7! Use priority code J8TLD2.
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Factor-talk mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/factor-talk
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Factor-talk mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/factor-talk
