Re: [STUMP] Re: Bug, or something, Stumpwm crashing when creating a new window

[Rohan Nicholls]

On Fri, Sep 5, 2008 at 2:05 PM, Rohan Nicholls
<address@hidden> wrote:
> On Thu, Sep 4, 2008 at 7:13 AM, Shawn <address@hidden> wrote:
>> "Rohan Nicholls" <address@hidden> writes:
>>
>>>> The value NIL is not of type STUMPWM::SCREEN.
>>>> 12: (STUMPWM::SCREEN-ADD-MAPPED-WINDOW NIL #<XLIB:WINDOW :0 C00009>)
>>>> 13: (STUMPWM::ADD-WINDOW NIL #<XLIB:WINDOW :0 C00009>)
>>>> 14: (STUMPWM::PROCESS-MAPPED-WINDOW NIL #<XLIB:WINDOW :0 C00009>)
>>>> 15: ((LABELS #:G145)
>>>>     :PARENT
>>>>     #<XLIB:WINDOW :0 69>
>>>>     :SEND-EVENT-P
>>>>     NIL
>>>>     :WINDOW
>>>>     #<XLIB:WINDOW :0 C00009>)
>>>> 16: (STUMPWM::HANDLE-EVENT)[:EXTERNAL]
>>
>> The problem seems to be that find-screen is returning nil and not the
>> window's screen. I suspect if you add some debug output to print the
>> value returned by find-screen in the :map-request event handler, you
>> will find this to be so.
>>
>
> I have set up a logging thingy that dumps the information
> to a file, indeed, the find-screen is returning null.  But I
> thought the other information might be useful that is
> being used in the function.  When I play with the machine
> tonight,  I will have opened a bunch of windows, and can
> mail what showed up.  If there is something specific you
> need to know about, let me know, and I'll try and add it
> to the debugging output.

And here is the debugging output:

":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 E00006> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 80B615> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 400011> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 400011> | screen: NIL| win: #S(window \"address@hidden" #x400011) |
wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 400011> | screen: NIL| win: #S(window \"address@hidden" #x400011) |
wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 40006C> | screen: NIL| win: NIL | wwin: NIL
"":map-request screen find: parent: #<WINDOW :0 69> | window: #<WINDOW
:0 40006C> | screen: NIL| win: #S(window \"address@hidden" #x40006C) |
wwin: NIL
"
It was basically the values of everything I could find at the time the
the event was triggered.

The "win: NIL"  above the emacs windows were urxvt's, and firefox
being fired up.  I don't know why they would not register, but it was
only with the emacsclient windows that the logging message (it was
echoed in the screen, when it was being written) showed up above the
newly created window, with the other apps it was under the new
window.

I hope this helps.

Rohan