World: r3wp

So - everybody going the Cairo way nowadays?

80-100% improvement sounds too good to be true :-) Would we get any 
other speedup, e.g. changing to compound rasterizer?

For the purpose of possible View enhancements, we started following 
document - so please update it, so that your ideas are not lost - 
http://www.rebol.net/wiki/View_Project

re Cairo: Cairo seems to be popular mainstream. But AGG is used in 
a lot commercial projects silently so it is hard to tell. The facts 
are:  
1. Cairo is slower than AGG.(when comparing SW rendering)

2. Cairo is a graphics library, AGG is a 'kit' to build graphics 
libraries.

compound rasterizer: well, this is another type of rasterizer(which 
has been already implemented into AGG2.4 codebase). I haven't made 
any perofrmance tests comparing the scanline AA rasterizer we are 
using but it could be a bit faster because this method shouldn't 
redraw objects one over another.(ie it renders the image 'flattened') 
I can't tell you how much faster it could be but the more layers 
you are using the bigger difference could be seen.

I already told this here but I know AGG was used in this higher profile 
"indie" game: http://www.wikgame.com/

anyone know how to break a call to wait []  so wait returns?

ex: a button with something like [kill-wait] in its action.

the awake handler needs to return TRUE

yeah, I had to patch wake event to add the functionality... would 
be nice to be able to interrupt a wait outside of its event handling. 
 is this doable in R3?

did anyone do a VID compatible tree view control.  the only one I 
can find is the one within rebol.org from the request-dir.r script.


not that its bad, just probing to see if there are others which would 
work more quickly for this project where I need it.

if you are inside wait, you are inside the event handling. so "interrupt 
a wait outside of its event handling" makes no sense to me.

the event handling dispatches code related to the event that triggered 
it.  there is no way for that code to interrupt the even handler, 
especially when the event handler isn't programmed by yourself.


when you add wait []  in a function, there are many reasons why to 
want that wait to return.  in this case, modal operation of some 
code.


wait is a nested system where a second call to wait doesn't break 
the first call to wait.  


To allow this I had to patch view's wake-event directly, but it works 
now.. 


 there is no way for a normal action to interrupt event handling a 
 part from closing all windows, which made this modal operation of 
 my code especially complex to build without an interruption mechanism.


I've had other situations like having a systray and view not collaborating, 
where closing a window ends the app even when the systray is running, 
or where the window wait keeps running even if all windows are closed 
and the systray was then hidden as well.   a timer event would have 
easily solved that issue by interrupting appropriate event handlers 
when it detects some situations.

there is no way for that code to interrupt the event handler

 - right, but you're not "outside" the event handler. you're still 
 inside, and it would be easy to look at the return value of that 
 code. that of course makes VID programming more difficult because 
 you always have to remember to return FALSE.

the alternative is to use global values / flags. the modal (popup) 
system in VID does exactly that.

yep... but an  'interrupt mechanism would be nice to have.  it makes 
a lot of stuff much easier to architecture especially when you have 
multiple ports to handle and state is complex.  you can just add 
a loop which includes all event handlers and interrupt all of them... 
check some states and then decide if you re-enable some or all of 
them, until you decide to stop the loop.


this way, a view event, can also interrupt network xfer for example.

Is there a way to receive wm_gestures ?  Will this be added to an 
enhanced R2 ?

I think you should be able to receive this event thru system port 
(not tested).

The main problem is that there is a bug in system port so the events 
(in general) aren't detected unless rebol console is opened.

It would be great improvement if this is fixed in new R2  release 
but you need to push Carl ;)

interesting ...

My new laptop with multitouch screen has just arrived .. the surface 
demos are interesting ... using two - four fingers to manipulate 
images, or a virtual earth.  Now .. it would be nice if R2 or 3 can 
support this!

What laptop is that Graham?

I'm contemplating getting one and would like to know how well Rebol 
View apps work with it.  I'm assuming they'd work just fine, but 
don't have the kind of money that I can take a chance  ;-)

It's the Dell XT2 http://www.dell.com/tablet
View apps work fine.

Graham, so your arms get tired yet? :-)

There's some issue with Windows 7 RC1 and the multitouch driver ... 
so I have put it down while awaiting the official Wndows 7 dvd from 
Dell.

Graham, did you happen to look at the HP Touchsmart (12.1")?  I see 
it offere at the local big-box stores for about $900 Canadian.  Gave 
it a try and it seemed like a nice machine.

Brock .. no.  I bought a touchsmart IQnnnn ( 19 inch wide I think 
) for someone else but I don't think it was multitouch.

I wanted accurate pen use which needs an active screen, and also 
touch.

Brock the Touchsmart TX2 is also dual mode ... see the reviews.  
There's some mention of lag when using the touch but you might be 
able to live with that.

Yes, the TX2 is what I was actually looking at.  I'll have to see 
if I can find a comparison of the two units.

Google query of:     HP Touchsmart tx2 Dell XT2 
finds lots of info

having trouble with checkbox updating...

this works > chbx1/data: true show chbx1

but if I take the state from a block >
blk: [ "String" true]
chbx1/data: blk/2
IT DOES NOT WORK.

I've also insured its a logic value with> chbx1/data: to-logic blk/2

Any ideas?

reduce blk first

to-logic on a word is always true:

>> to-logic 'true
== true
>> to-logic 'false
== true

Got it! Thanks a lot, Henrik.!

I did not know that about to-logic...
not mentioned in "Word browser"

It's documented in Core manual: http://rebol.com/docs/core23/rebolcore-16.html#section-3.3.3

Also, don't forget SET-FACE.

I'm trying to print a vid face face via conversion to image...
what's the best way to get good text quality...
ie...is the a best font to use...
my out put is a little jaggy

either you use AGG for aliased fonts, or you increase image size 
so its equal to the dpi you want to print.

vid layout

(and scale all the graphics involved in your face)

use 'draw' to lay it out wll produce better image?

the fonts will be anti-aliased (which is what I meant to say above)... 
but in any case, the resolution you use on your printer is always 
much higher than you screen resolution.

How far did Henrik get with his vid to pdf dialect?

My layout is all text except for one small b/w image...

Maxim, so to display a proper page I would need to increase the sacel 
of my vid image quite a bit...larger than screen?

sacel = scale

so if you are creating images for printout, you have to think larger 
faces with larger font sizes... screen dpi is typically between 72-100 
dpi, printers now usually printout at around 600 dpi.... so if you 
want to have clean images, they should be around 5 to 10 times larger.

yep.

when I was doing an art project, I had to create a printout for a 
36 inch wide image... the file was an 508 mb, 300 dpi image.

normal printing, uses vectorial information, so the convertion from 
shapes to points (rasterization) is done directly at print time.