AltME groups: search

Is this a bug or by design?


A faces pane, if of type block!, may contain not only face objects, 
but allows for words too (and maybe other types, but I didn't check 
this). At least it works and View doesn't complain.  

I sometimes use this for easy access of faces in crowdy pane; e.g. 
for faces A, B, C and F I can set do F/PANE: REDUCE ['a A 'b B 'c 
C].


But now I noticed in some popup-faces' code that SYSTEM/VIEW/WAKE-EVENT 
fails in such cases because of the line IF GET IN ITEM 'ACTION [BREAK/RETURN 
FALSE], which I had to patch to IF ALL [OBJECT? ITEM IN ITEM 'ACTION] 
[BREAK/RETURN FALSE] to make it work through such panes.


Well, I'm really uncertain whether it is a good idea to build panes 
that way and if I have to think of WAKE-EVENT as being buggy, or 
if by design pane blocks really should consist of nothing but faces 
and the bug is me ignoring that.

So, what do you think?

Christian, I don't think there are any official docs on the contents 
of pane, but I would avoid this approach. Most code would be expecting 
only faces, eg. code that shuffles faces in order to do depth arrangement 
is probably not expecting words in there too (and, if it did expect 
them, should it assume the words are associated with the faces ?)

there are two problems: some mezzanine code doesn't work with that 
(but this could be fixed), and view actually tries to lookup the 
words.

that is, if your pane is [b], and b references a face, view will 
look it up and use it as if the face object was there instead of 
the word. because of this, your faces will be in the pane TWICE as 
per the example above.

Gabriele, thanks for the "being in the pane TWICE" explanation: you 
pointed me to a mysterious "face in more than one pane" error message 
I wasn't able to track down to it's origin. Should've read "face 
more than once in pane", though ;-)  Then again, it seems to work 
fine if the word-lookup doesn't evaluate to a face (even non-face 
objects seem to be allowed).


Did you had some special mezzanines in mind which may stumble over 
this? WAKE-EVENT currently is the first and only one I know of for 
now. Of course the number of idioms like FOREACH FACE PANE [FACE/...] 
is legion, so this really may turn out to be problematic.

It's just that the approach seems to fit into the 'code is data and 
data is code' idea quite nice.

(that is, i once had a panel with buttons that had a key defined; 
the keys worked in one case but not in another. basically, in one 
case i was using a style that used words instead of faces in the 
pane - so it was easier to probe it for testing - and the function 
find-key-face didn't recurse into that pane)

The more I mess with ImageMagick and PIL, the more I like Draw :-)

is there a quick way to set a maximum length of text in a field and 
automatically unfocus and move to the nexst field in the tab order? 
E.g. day, month and year fields with limits 2, 2 and 4 and automatically 
jump to the next field

I know of a way, but I wouldn't call it quick - override the engage 
func. for a field and take advantage of the 'parent-face var in the 
object... I'll start a quick example if you want to see it

it's a bit buggy... when I select a field and start typing something 
when it already has a length > 3, the first character I write is 
printed and it jumps to the other field

As I said, I'm still new to styles - hopefully someone with more 
experience will come along and do something better...

Thank you  =^)


And I like the idea, maybe if I have spare time I'll attempt a bit 
better one....

Does somebody face with the plugin problem that I got!? I tried to 
setup an Rebol script as a Plugin in webpage, but got message Cannot 
run ...script name.r. Then to be shure that all is o'k I just copied 
an example (web page and Rebol script) from www.rebol.com and put 
on my localhost web server. All of those examples works fine on my 
IE6 and WinXP.  But when I tried the same example on my localhost 
web server I got the same message. Of course I changed parameter 
LaunchURL to point to the same (copied) script on my localhost. What 
is a diffrerence in environments that I don't know and how to solve 
this problem?

in this VM there is nothing or near nothing done and yet we get a 
lot of memory used for nothing ...

and keep recycling until it is stable.

We should not have to worry about how and when recycle works. It 
should just work.

instead of having frames not related to a rebol word and  still allocated 
in memory until the GC decides to do he's job ;)

But I also admit my opinion is probably useless until I've studied 
and tested various garbage collection schemes.

What is the true basic problem ?? I have in rebol memory chunks unrelated 
with world on the diccionnary of rebol  and this make the memory 
management just impossible because it dépends on too many things 
unrelated with coding ...

C is definitely not Rebol. But consider this:
	a: b: context [data: "love me"]
	a: none
Obviously the object should not be freed yet.

The question is, should a full recycle occur (checking for unattached 
values) ?

I think that would make setting words very slow and degrade performance 
terribly.

Anton it depends if a; and b; store each the cloned content ;)

and

And you should recycle and check stats multiple times before and 
after each test.

and obviously you will clear only single related data

no such file. created an empty user.r and now it won't start at all

you're trying to fight the system without coming up with a better 
one, that's all.

 OH OH Comon we are talking about a serrious issue and that's not 
 because I make  a criticise of the system that I have a better solution 
 ....

ypu can say I hate capitalism because it makes  little people very 
ricth and most of the people just poor  without having a rechange 
solution ;)

shadwolf: try setting all inbuild functions to none and see if the 
system needs less memory. maybe that helps

Mmm.. When you figure it out, write a quick "install script", which 
detects debian (if you can), and alerts the user with the advice. 
Then, when someone asks in future, you can just give your script. 
Also, this may evolve into a fully cross-platform installer script.

/etc/termcap ? i lack some concentration, early morning here. carl 
has changed keyboard handling for redirection and now you need that 
stuff. but cant remember the names. apt-getted something and it worked.

I remember the problem with Rebol/view, Linux and foreign character 
input in field/area (spanish/portugese).
What was the solution, if any ?

could such errors could add a field for the face? so that we can 
look into /stye and /facets? could help

I have a proble to submit to VID GURUS... My problem is the following 
I writing a click and scroll algorithm. (ike In MDP-GUI you have 
the floating toc window panel when you click on on of the title shows 
in the floating toc window the visualisation scroll to the related 
title. But this show the title on the middle of the screen and no 
on the top of the visualisation pane. The forumal I apply is sld/data: 
related-offet/y  / sub-pane-visualisation/size/y  and then I apply 
the scroll function. What must be the for mula to have always the 
related title in the visualisation pane on the top part of the screen

I make an image to show you the dilema I build an image that explain 
what is the actual problem and what I want http://shadwolf.free.fr/clic-scroll-img.jpg

for example title 4  with related-offse/yt: 1586 and sub-pan-visualisation/size/y: 
3142 i get a sld/data: 0.504774029280713 and I have the title around 
the middle of the visualisation box. If I use the scroller to place 
the title 4 on the top of the visualisation pane I get sld/data: 
0.602247191011236

Shadwolf: I guess first title is shown at top and last is shown at 
bottom, yes or no ?

Rebolinth anyone can write an SVG emitter, just like there is a Flash 
emitter. I think this complex and unique enough that it shouldn't 
be included in the base. But 3rd party script will do.

thanks to DideC, Volker and most specificly to Brock witch helps 
me a lot

the problem was complicate to understand and chatting with brock 
help me a lot to think

and find the solution

Allen - you may not be interested if Carl's rotary serves your needs 
but I find the FX5 rotary that Frank Sievertsen created some years 
ago to be a bit more usable because he added the left and right arrow 
navigation to the base rotary.   Clicking on the rotary itself rolls 
it forward but with the FX5 one, you have a back option instead of 
having to cycle through the list.    The rotary is still a bit cumbersome 
without key board support that is specific to the widget when it 
has focus.

...and the inbuilt rotary allows to go back using the alt-click (right 
mouse button).

I just hacked the graph-layout stuff found in Rebol-Framework into 
a standalone using the new AGG stuff. Very cute and fast!! Take a 
look at: http://www.robertmuench.de/download/graph-layout.r

It's not perfect yet, so please feel free to improve (and let me 
know). Next thing is to add boxes for nodes.

Updated to show boxes. Now, is there any way to associated an ID 
with a box and get it back if the users clicks inside the box?

... and its size.  So you make a canvas/feel to trap the mouse clicks, 
then search your nodes to find the one affected.

Your nodes need DEPTH, so they can be depth-arranged. Double-click 
can send a node to the head of nodes and shift-double-click can send 
a node to the tail.

Could you try following on your system? Run new View alpha, go to 
desktop, run Bubbles and then start Particles demo - both demos slow 
down dramatically. Those are separated processes though (that can 
be proven by closing the Desktop - demos will stay running). Cyphre's 
theory is, that it is normal, and that CPU usage is high and so OS 
have to divide its CPU time between the two. But normally my PC can 
run several videos without noticable glitch, I wonder if Rebol will 
give it a hard time ...

Bubbles was made to always take 100% of your CPU time. it easily 
gets to 50+ fps, and this is a great result considering the amount 
of work it is doing.

put in a 'wait for a short amount of time and cpu usage may drop 
dramatically. it works well with responsiveness for scrolling large 
panes even if the frame rate drops a bit

I've been looking at %view-edit.r recently (and Romano's excellent 
http://www.rebol.it/~romano/edit-text-undo.txt), and have the following 
three code change suggestions:


1) Allow Shift-Tab to cycle back through the first / last pane objects 
(as Tab does):

	back-field: func [face /local item][
		all	[
			item: find face/parent-face/pane face

   any [if head? item [item: tail item] true] ; new line added here
			while [face <> first item: back item][
			...

2) Implement a new function to hilight the current word.

	current-word: function [str] [s ns] [
		set [s] word-limits
		s: any [all [s: find/reverse str s next s] head str]
		set [ns] word-limits
		ns: any [find str ns tail str]
		;	hilight word
		hilight-text s ns
		show view*/focal-face
	]


3) Refactor the engage / down action to allow double-click selection 
of a word (something I use all the time in almost every editor I 
use).

Current code:

	down [
		either not-equal? face view*/focal-face [
			focus face
			view*/caret: offset-to-caret face event/offset
		][
			view*/highlight-start:
			view*/highlight-end: none
			view*/caret: offset-to-caret face event/offset
		]
		show face
	]

Proposed change:

	down [
		either event/double-click [
			current-word view*/caret
		][
			either face <> view*/focal-face [focus face] [unlight-text]
			view*/caret: offset-to-caret face event/offset
			show face
		]
	]

Comments?

Yes, but if it were about changing the font, that probably may be 
possible by changing the fields font dynamically at the right time, 
may be by modifing CTX-TEXT.

What I meant is the way highlighting is "drawn". It looks much like 
if the text is rendered inverse with some sort of keying. You'll 
see what I'm trying to express by looking at


view layout [field "MMM" 210x100 effect [gradient 1x0 blue red] bold 
font-size 78]


and selecting the text. Obviously something like FONT [COLORS: REDUCE 
[WHITE BLACK]] here doesn't work, because FIELD/FEEL (i.e. SWIPE) 
doesn't make use of FONT/COLORS.

But more so, it seems like it's currently technially impossible to 
do so. In the end, the answer really seems to be No! for now.

Maybe I'll find "a little spare time" to do a field which draws it's 
text in the effect block and rewrite the stilll somewhat buggy text 
editing functions ;-)

(Not a serious comment though, because that's definitly not the league 
I'm playing in.)

Is there a way to show a pop-requester with a time-out? I want to 
show if for X seconds and if nothing happens return a default action.

And this allows (yes) (no) (cancel) as well? Didn't used it yet.

Has anyone tried 1.2.117 yet? I'm experiencing problems here getting 
it to load my user.r on startup, even though I've already tried it 
with different places and registry settings.

re versions I'm trying to run Didec's delete-emails which requires 
View 1.2.8+. I downloaded the latest version of View and it shows 
up as 1.2.1? When I look at the downloads dir I see 1.2.8 circa june 
2005 but the more recent files are labeled version 1.2.1. Can someone 
explain why newer code has a lower version number?  Thanks

The latest View beta is 1.2.119, which was released yesterday (and 
possibly throws up some new bugs/issues for older programs.)
http://www.rebol.net/builds/031/rebview1211931.exe

Otherwise I can recommend you try View beta 1.2.48, which is not 
so new and therefore more likely to have all its issues understood 
by delete-emails:
http://www.rebol.net/builds/031/rebview1248031.exe

It's not a final version, but is completely usable. What is not finish 
is the editor for the rules.

So, this rules are stored in the rules.txt file (this file appear 
after your run it one time) and should be edit manually.

Gregg, I added com12 to the system/ports/serial and it gets thrown 
out when I OPEN.  Will work if I used com7 on another PC - seems 
to only like a single digit for N in comN??

Thanks, I've tried some com port mapping software (maps com12 to 
com1) and then all works.  I can keep the REBOL com port as com1 
and change it in the mapping software.  A work around for now.  Tnks 
for the help

and please - could someone come with more attractive ViewTop skin? 
Maybe we could do a small contest, no? :-)

Henrik - your zoomer is nice - I vote for it to be included in View 
Desktop test folder - nice example of AGG usage .... although REBOL 
is bitmap, or not? Once enlarged, I can see R and O letter with jaggies 
... :-)

wow...you guys are pritty ahead... I must still let 1.3 do the work 
and you are already talking 1.4 ;-)

I'm not sure what else it should be... AmigaDE is actually the same 
principle. They sell you a program which launches kind of a desktop 
where you can launch AmigaDE games and programs.

I wonder why 'list style is not based upon Carl's IOS Developer's 
version, which was said being fixed and more advanced?

Is that OK though? I have to find some windows app and try it :-)

That's right. Radio buttons make no sense unless they are in a group. 
If you want a single on and off option then you must use a checkbox.

well, thank to others. I just asked few days ago if there is new 
version of toggle, and someone (Ashle?) gave me an advice :-)

Ok, that was weird!  I just started View and it told me there was 
a new version to be downloaded and when I clicked download it popped 
up a file requester asking me where I want to download view.exe to. 
 The weird part is the default path it popped up was C:/Program Files/Common 
Files/System/Mapi/1033/NT  Where the hell did that come from?

Hm...  Somehow VIEW is triggering my Virus Scan Software...  Not 
sure just what is going on but when I run View 1.2.125 it is telling 
me that there is a new version of View available and immediately 
thereafter my Virus Scan is popping up a requester claiming files 
that I KNOW aren't viruses are viruses.

I just ran the View desktop, and was advised of a newer version.. 
which I duly downloaded.

And the familiar rebol icon is back again.

rebol 1.3.0.3.1 final - upgrade command, and the result? "You are 
running an unknown version of REBOL." :-)

Can someone try 'call and 'sound interface? At least 'call seems 
to work, while when you start Desktop, those components are listed 
as "(View/Pro). If those are unlocked, then Desktop text should be 
enabled ...

that is bad, and probably mistake by RT, not intention - they claimed 
sound will be free for 1.3. 'call seems to work here, but I have 
license key, so I wonder if I don't run /Pro.

Is there a "proper" way to set font properties for a single element, 
such as 'text after the layout is created?

I'm trying:
view layout [a: text "12345" b: text "12345"]


and set the font size for a: a/font/size: 14 (b is 12 at this point)

then I unview it and:
view layout [a: text "12345" b: text "12345"]
once more. Now b has font size 14 as well.

They seem to share the same font object....

and set the font size for a: a/font/size: 14 (b is 12 at this point)

 <--- I don't set a to a/font/size there although it looks that way, 
 sorry.

should be "and set the font size for the variable a to: a/font/size: 
14"

A and B share the same, so what you need to do is

Shadwolf, yes, you're right. Today VID is a bit loosing the focus 
in favour of REBOL/View 1.3.0 and probably RebGUI, but I'm sure there 
will be efforts again.

but my ask in the reflexion process is why to consum less code less 
memory less CPU  and be sharpen if we are only at the level of what 
we can see with winAPI ....

We must IMOO  have a widget set that is similar than what are  people 
habbits and be smoother and able to be distinguish directly

we abstract from OS capabilities and set of widget and we introduce 
a commun set

this is in my opinion the same intent as GTK+, WxWindow, TK, QT and 
other in this mood libraries ;)

As I said - we need tree-view, grid (or little better list), tabs, 
group-box, and maybe even menu. Those missing styles will make it 
difficult to newbies to start using View as full-fledged dev-tool 
replacement ....

Chris your code can be bettered more people will work on it more 
sharp it will be and if we have as goal to include it into a state 
of art basement library like rebGUI I think this will be very enthousiastic 
for people who will participate ;)

I make a listview widget multicolumn multi resizing with drag&drop 
Viewed bar effet like in Win32API and multi widget type capable in 
only 3 weeks and using less than 300 That's the illustration that 
REbgui and VID base concept are pretty good  !!

http://hmkdesign.dk/rebol/listbin.r<--- very, very quick and dirty 
demo...

And man, it's fast. I notice no scrolling delay!

hehe... well, there litterally are about 50 things missing yet. I'm 
emphasizing on having speed even on huge lists and minimal editing 
capabilities.

christianE: what it does is I create simply a fixed grid of textboxes 
and replace the text in them on the fly. that's much faster than 
setting offset for a very big pane. this way it doesn't matter how 
big the list is, it'll stay fast

Since the very beginning of VID we have seen lot intent of widgets 
improvements make by pationnated people but we never take the time 
to enclose those work into a trully thinked system that's why we 
don't have a reable and durable widget design effort and RebGUI  
tryes to fit this need

that doesn't mean that we can't anymore work on our side to search 
for better widgets and better interactivity system but havng RebGUI 
 improvement as goal makes our work to have a meanning

once our work is achieve we can confront it to the other ones and 
than enhance it more and that's the assurance that if one day we 
are not anymore interrested on develpping this idea our whole work 
wont fall to the pitfall of forgive

right now there still are a ton of functionality improvements to 
come, and it'll be even faster than it is now. I want to implement 
dynamic column sizing, graphics in lists, drag'n'drop and proper 
handling of list actions

Henrik truelly RebGUI  is simple you have basic object (rebface) 
with only the needed things and starting from this point you can 
make new widgets enheriting of rebface basic things and extended 
them ;)

Same here, I expected to need WinXP simulating menus and thought 
others may be interested, too. No deeper strategy involved here, 
shadwolf  ;-)

you have the 	assurance that your new enherited object will only 
countain the needed things and no more ;)

shadowolf: ok... mine is still a bit backwards. I have a list object 
which contains all list functions and the raw layout for the list. 
when I want to use the list, I create a box in a layout and assign 
it to the pane of that box, voila. :-) the list adapts to the size 
of the box, so I think live resizing would work though columnsizes 
are still fixed