World: r3wp

it's a RebGUi issue

there's a bug in my widget and i have to stuff one character in it, 
delete it and then insert the characters I want

I don't think you can insert a del

or backspace ...

you can't generate events in normal view R2.

though you can call a face's feel if you want.

just build an object with the same fields as an event and it will 
work  :-)

is there a windows key stroke recorder?

yeah it exists.  just google it.

hmm...maybe something else.  just control the mouse and the keyboard 
buffer

I thought Gregg did some Api work with windows... can't find it by 
googling.

Leaving all RebGUI specialities aside, can't you just patch the widgets 
FEEL/ENGAGE to allow objects being passed to it? If there is a point 
where you're in control to FOCUS the face in question, that may be 
possible.

I can't actually get focus to the face either :(

My old send-keys stuff should be around somewhere. If not, let me 
know and I'll zap it to you.

did it not rely on the console being opened to work?

Gregg, please send it to me

Graham, sent.

Thanks .. looks like send-keys solves my problem too :)  I needed 
to send a backspace to a rebgui field to initialize it and this works

for those who don't know how to change the title of a window in R2....

rebol []

win: view/offset/new layout [

 button 400x20"change window name" [ win/text: random "ogiue goiueng 
 oesigneso ginue sogine sgioun " win/changes: [text] show win]
] 200x200

do-events

The following code cause an error if you click on the window (XP 
SP2) :

view layout [box effect [ draw [
    pen none navy
    line-pattern 2 7 2 3
]]]

Does others could confirm the problem ?

it just quits here in 2.7.7.

It's known bug. At least Maxim was talking about it with Cyphre in 
OSX group a few days ago.

Tested with View 2.7.7 & 2.7.8 on Win7/32-bit, it quits too here.

I always get :

Instruction at "0x7c920cce" use the memory address "0x40080000". 
The memory can't be "read".

(rougthly traduce from the French message !)

yes, its a BUG in AGG, using more than 2 values in line pattern will 
cause the crash.  its the oldest AGG bug I know of.


spoke with cyphre, its possible that this will be fixed in next release 
of R2, which shoudn't take more than a few weeks IMHO.

How does one capture whether a window is maximized or not?

I can see a maximize and a minimize event, but no restore event. 
And face/changes is cleared, when the event handler reads it. Not 
very useful.

face changes is used for the internal show to optimise if/what it 
should redraw. as such its a "write only" property, reading from 
it makes no real sense.


for example if you set changes to [offset] it will move the window 
MUCH faster than if you change offset & show without it.

I have a problem. Basically I need to know how to interupt a function 
that is looping. 

So this is what happens: The user presses "start" button which calls 
the printing routine. I want to be able to allow him to stop the 
process if he wishes by pressing a "stop" button. The issue is the 
face is only recording my attempts to stop. Any thoughts?

I usually set a flag with a button and try to read the flag in the 
loop. I'm not sure if it's easy to get an immediately responding 
button, though.

perhaps if you wait a short period prior to reading the flag in the 
loop.

Yes, I've tried to do that but it doesn't seem to work.

I'll put a wait and see

OK, I'm getting close.

Yup, the wait does the trick. Thanks Henrik.

ok, cool

yes, the wait allows the event system to act like an async system. 
 


just be careful though, you can get lost in concurrency and get apps 
which close all windows but don't actually close the rebol process, 
which ends up in an event deadlock.    :-)

Yes, I actually saw that in my tests as well. For now this one "wait" 
seems to be doing its job and I just hope it is not system-dependent 
in terms of the time I have set the wait for.

Can someone confirm this issue: secure allow permits all access, 
but library access still pops up security window.

>> secure allow

== [net allow library allow shell allow registry allow envr allow 
file allow]

>> dll: load/library %REb-Excel.dll
** Access Error: REBOL - Security Violation
** Near: dll: load/library %REb-Excel.dll

The problem still there even I start rebol.exe with -s option.

But if I start rebol.exe with --secure option then it doesn't pop 
up security check window. I tested this on Windows XP/2008.

So it looks -s and --secure don't behave same. Which can be a bug 
in current R2/View.

I've never had the problem. (even on w7)

I always use -qs.

secure allow should behave like that, right?

AFAIK yes.

if we don't use --secure allow or -s.

so maybe there is a bug within the secure function.  did you check 
on RAMBO?

yep, I checked rambo, no ticket for this.

I've never seen this either. But all my scripts that use load/library 
are started with -s... (or encapped)