World: r3wp

And this one give the result PhilB mention :

lay: layout [btn "Test"]
print lay/size
view/new/options lay [resize]
print lay/size
do-events

Notice that the size is not changed. So View think the window is 
not as large as Windows does. Looks like Windows set the window size 
to it's own minimum size.

Funny! With 1.2.1 the not filled part is white, with 1.2.48 it's 
filled with what is in the background (like if it was transparent) 
:-))
But it's not ! Just move the window :-[

(tested under WnXP)

Yes, I observed this behaviour long ago. It should be added to RAMBO, 
if it has not already.

pliss about sample aplicacion fastcgi:// compile exe

bat found table

lay: layout [ 

origin 0x0
space 0x0

list blue 250x220 [
 across
 adres: field 50  

 zmienna: field 50
 pad 0x4
 wartosc: field 50
 pad 0x4
 tag-in: field 50
 pad 0x4
 tag-out: field 50

return ] data

[["join" 100]
 ["yui" 89]
["ytr" 5 ]
["99" 7]
["7878" 5] ]

Q. On Windows you can define a func as:

maximize-face: func [
	"Maximizes a face on screen."
	obj [object!]
][
	obj/changes: [maximize]
	obj
]

which lets you do:

	view maximize-face layout [text "Hello world!"]


but this is of limited use when you want to know how big the [maximized] 
screen will be *prior* to layout.

	help system/view/screen-face/size

isn't quite what we want, so I cobbled up the following:

max-win-size?: has [scr-face view-face] [
	scr-face: system/view/screen-face
	view-face: make face []
	view-face/changes: [maximize]
	;	view the face
	insert tail scr-face/pane view-face
	show scr-face
	wait .001
	;	unview the face
	remove back tail scr-face/pane
	show scr-face
	view-face/size
]

which has two drawbacks:

1) It requires the face to be shown (screen flickers briefly)
2) It doesn't work on Linux

Anyone else mucked around with this sort of stuff?

3) It dosen't work on IOS (1.0.6)

help !! Scrolling panel

plis source code ?

do http://www.lexicon.net/antonr/rebol/gui/demo-scroll-panel.r

You must use Rebol/View 1.2.1 < version <= 1.2.29    (I am sorry, 
for so small range of versions... can be fixed...)

Source for scroll-panel style:  http://www.lexicon.net/antonr/rebol/gui/scroll-panel.r

plis sample code append panel

Guest: why don't you claim an account for Altme in the "Accounts" 
group ??

how I have this to make ?

What is the /away

refinement in show-popup for?

I can't find it in any doc.

I have the impression that the /away efinement on show-popup function 
 will let the focus to the parent window .. but not sure

>> source show-popup

show-popup: func [face [object!] /window window-face [object!] /away][
    if find pop-list face [exit]
    window: either window [window-face] [
        if none? face/options [face/options: copy []]
        if not find face/options 'parent [
            append face/options compose [parent (none)]]
        system/view/screen-face]
    face/action: away
    insert pop-list pop-face: face
    append window/pane face
    show window]

face/action: away

Yeah I never seen that one either.  Someone should get Carl to throw 
that in the docs

It is used in wake-event.

write clipboard:// mold get in system/view 'wake-event

the away refinement try to create modal face which are closed when 
the user click outside the face (like in menu)

Thanks Romano - I was trying to document it in french, but wasn't 
satisfied with the official description "/away - the away refinement"...

hehe

Hi Folks. Where can I find vid-styles to look like and feel a XP? 
I know that SurfDetect uses something like that.. Any tips ?

Today I was asked by my friend to help him with View. I did not know 
he still uses Rebol at all ;-)

He has following "problem" - imagine two faces. First one, container, 
of size 50x50, featuring scrollers etc. Second one, inside the container, 
of size 100x100. Now he tries to get image of the larger face, so 
he does something like to-image container-face/pane ...

The result is image of 100x100 correct size, but the image is cut-off 
of content hidden by parent face ....

the strange thing is, that when he does 'unview, followed by the 
same command, then the image is full, not cutt-off ... anyone?

btw - he uses imo 1.2.10, as he told me he want to use last "stable" 
version .... once again - we should probably do something with View 
release, or ppl will still use old stuff, even if new one is stable 
enough to use ...

I think if you 'view face and then you use 'to-image rendered bitmaps 
are taken - so it's just 100x100. If the face is not viewed at all 
or unviewed, all faces and subfaces must be completely rendered, 
so you get full bitmap....just my thoughts

hmm, so no chance of getting image of 100x100 face?

I have thought that "unview" "destroys" rendered faces? So I am surprised 
he got correct result?

'unview probably "clears" "bitmapcache" so all faces and subfaces 
must be rendered again if you use 'to-face. and 'view probably composes 
all bitmaps together, so everything overlayed is lost when using 
'to-image...but that's just what I think, I don't know how it really 
works internaly

'to-face --- it should be 'to-image

hmm, and once he is in View layout = rendered state, he simply wants 
to press a button and get resulting image ... Maybe he could render 
it once again in the background somehow? But how?

mh, probably make new face (without 'view-ing it) with the content 
of the subface

has he tried "show container-face/pane" before to-image? maybe that 
helps.

why would there be show needed? IMO it is correct that container-face 
clips the image .... hmm, although maybe not, becase the image is 
still 100X100, just content is 50x50 :-)

the buffer is 100x100, but /view knews only the 50x50-part is needed, 
so draws only that in the buffer.

I will send him email to try ...

you must somehow tell the image its not clipped. when you show the 
container, it thinks it is. when you show itself, maybe it renders 
full.

i guess faces remember if they are already rendered or not. but they 
don't remember that they are only rendered partly. 'unview tells 
them they are unrendered. hmm, maybe a 'hide does the same. would 
only flicker, not unview the whole window.