World: r3wp

Conservation of initial energy.

true ...

Is anyone able to point me to where to look for an error in some 
simple custom R3-VID-style? The following error message clearly doesn't 
cite my own code, some window handler code seems to try call FACE/FEEL/ON-MOVE 
on a FACE = NONE! oder FACE/FEEL = NONE! The error happens as soon 
as I hover the face in question:


** Script error: path face/feel/on-move is not valid for none! type

** Where: either handle handler while applier wake-up loop applier 
wait do-events catch if either if do begin do
** Near: either drag-gob [
    either drag-min [
        drag-gob/off...

that's inside VID3 alright, but what exactly happens, I don't know.

I can't remember if it is because some function inside your style 
isn't returning a face value or what it is.

Jerry; "why just us"  imho one of the bigger reasons is that we, 
the faithful, communicate in the dark.   There is a buzz, but no 
one hears it.  There is no comp.lang.rebol for people to "bump into" 
for instance.   reboltalk.com gets a few (or less) posts a day, but 
imagine how it would look if the Altme trafic counted.


Sunanda has helped with aga (and RT before) with the web exports, 
but the rebol.org data is "old news" and doesn't allow a surfer a 
chance to take part.


We get excited (a good thing) when we see a REBOL mention on a site. 
 I'd like it if say schemers got excited to hear rebols discuss Arc 
(that would be a better thing).


So the power the Altme (a wonderful comm system) has us in a position 
of "Hoist by his own Petard"

the power the Altme = the power of Altme

Thanks Henrik! Yes, I was assuming some NONE somewhere in my code 
instead of a face, too. Still not solved, though :-(

I've seen similar errors, but it's been a couple of months since 
I last touched VID3. It's a good idea to take a look at an existing 
style to see what's missing. I'm fairly sure that some code is missing 
in your style.

(hopefully this problem completely disappears when the new style 
dialect is done)

Christian: you haven't set gob/data for one of your gobs. don't worry, 
this will mostly be automatic in the next version... :) anyway, gob/data 
has to be set to the face the gob belongs to for each gob you have 
on display.

Thanks Gabriele! I've learned it the hard way, so the good thing 
is, that's something I will never forget again. :-)

drop-file.r show how get draggin from windows enviroment. Is possible 
 to initiate a Drag 'n Drop operation from REBOL3? (ie for to send 
an image to another application).

Not at this time. It requires a COM interface IIRC. I can't remember 
if/how it was done in Windows before that.

In webpage http://www.rebol.net/wiki/VID:_Commands there is an example 
of TEXT-LIST in R3 as follows:

view [
    text-list [
        ["Header col1" "Header col2"]
        ["row1 col1"   "row1 col2"   "value-of row1"]
        ["row2 col1"   "row2 col2"   "value-of row2"]
        ["row3 col1"   "row3 col2"   "value-of row3"]
    ]
    selected [1 3]
]

I tried a lot of things to get it to accept a block! for the rows, 
but couldn't find a way. Any hints?

I'm not sure what yo mean Arie. That example works for me here. Can 
you show us some code?

In the next snippet out is a block! and I want it's contents to be 
displayed as the rows of the text-list:

view/options [
  text-list [
  	["header"]
  	[out]
  ] options [size: 300x200]
] [title: "my example"]

I tried it with and without brackets around out

Also I couldn't find the specific options for text-list

Maybe I have to specify it in the options?

Don't forget it's still Alpha version... it's very far to be complete!

I know ;-) But I'm just trying it; it is not for production anyway 
...

Just experimenting with R3

Arie;  Did you get "out" showing up? 
out: ["1" "2" "3"] 

view [text-list ["col1" "col2" "col3"] [out]]]    will show   out 
none none   in the list, you'll need to compose the out

I wouldn't consider this a real solution, but...

view/options compose/deep [
  text-list [
  	["header"]
  	(out)
  ] options [size: 300x200]
] [title: "my example"]

Ah, there's where the other half of my brain went. It's in Brian's 
head. :-)

:)  Race ya!

Nope, this displays two lines:

header
 and "out" as in all my former tries ;-)

Literally that is I mean.

Yeah, REBOL treats block data as - don't reduce this - by default. 
 Try Gregg's sample.

Same for btiffin and Gregg; it outputs out as a literal string in 
the output row

In R2 you could specify data: out; maybe there is an option doing 
the same trick?

Try this;    a: 1 b: 2 c: 3    d: [a b c]

Did you use the exact code I posted? It works here.

You can copy it from here and use "do read clipboard://" to run it.

Sorry Gregg, I missed the point of the compose! Stupid ... Retrying 
now

Yessss Gregg, it works. But it isn't the most elegant solution. Neat 
trick though :-)

That's why I said I don't consider it a real solution. I don't think 
we have face accessors in place yet.

Maybe Gabriele can shed some light on the inner workings of this.

OK, I'll be patient then. Thanks anyway!

Arie;  This is a REBOL paradigm.  Block data is literal.  Your [out] 
 block is just that, a block with the literal word 'out  in it.  
It needs to be evaluated (reduced) to get at the value of out when 
inside block markers.

btiffin: yes, but I don't know a way yet to achieve this within the 
r3 text-list specs as known today, hence the trich Gregg showed which 
is a kind of meta-reduce :-)

trich is trick

Ok.

note that text-list will be replaced in upcoming VID prototypes. 
it's a stopgap solution.

Henrik: ah, thanks; that is useful info!