World: r3wp

This is http://www.rebol.net/tests/view/draw-test.r.
This happens only with Win98.
I tested it with Win2000 and it worked

But there is the difference, that Win2000 was not connected to the 
Internet.

Worked under Win2000 internet connected.

With 1.2.106.3.1 and 1.2.107.3.1 : bug with pen color setting in 
draw block:
view layout [box white effect [draw [
	pen none box 10x10 70x20 ; invisible box
	pen black                                ; pen to black
	pen none box 10x30 70x50 ; invisible box
]]]
is ok,  but not
view layout [box white effect [draw [
	pen none box 10x10 70x20 ; invisible box
	pen black text 12x32 "hello" ; pen to black, some text

 pen none box 10x30 70x50 ; without AGG (->1.2.48), invisible box, 
 with AGG (alpha & 1.2.106+), visible box
]]]
Drawing something with a visible pen breaks invisible pen usage.

(submitted to RAMBO)

Sqlab: Can you provide more detailed info about the crash you mentioned? 
Can be reproduced by repeating the draw-test.r? In what exact part 
of the test Rebol crashed? Thanks.

Vincent: thanks for the report. We know about this problem and will 
be fixed in 1.3.

Another problem with AGG : draw text does not render foreign chars 
:

view layout [text "Après noël en forêt" box 150x20 effect [draw [text 
0x0 "Après noël en forêt"]]]

I wonder what font AGG use ? None of the ASCII 128+ char are displayed, 
 except 251 that show "OBJ" in a box !!

Here is a more complete test :

base: [text (c) 40x20 effect [draw [text 20x0 (c)]]]

lay: [origin 0 space 1 across]
for i 16 31 1 [
	for j 0 7 1 [
		c: copy to-string to-char 8 * i + j
		append lay compose/deep base
	]
	append lay 'return
]
view layout lay

Gabriele: Linux Slackware kernel 2.4.26

Ill post it Rambo

Draw text : added to RAMBO.

I believe that text rendering will not be included in View1.3 version 
of AGG/Draw. It will come a bit later.

what exactly is meant by that? 1.2.107 renders fine anti aliased 
text.

henrik: antialiased text only works in windows, so it will not be 
in 1.3

about crash on win98: should be fixed in upcoming release :-)

DideC: here is result of the test on my machine: http://www.rebol.cz/~cyphre/agg-ansi.png
It looks normal here imo. But Rebol is not unicode yet so this is 
problematic area. Which version of Windows are you using? What default 
codepage/keyboard/national settings?

Draw text and foreign chars: same problem here on Win2k SP4, with 
West European default codepage, Swiss-French national & keyboard.

Gabriele why can't AA fonts it be supported anyways on window?  its 
easy to state in the docs that windows users have an additional option...

what's this about:

>> to-time 721
== 0:12:01
>> to-time 721.4
== 0:12:01.399999999
>> to-time 721.2
== 0:12:01.2
>> to-time 721.3
== 0:12:01.299999999
>> to-time 721.4
== 0:12:01.399999999
>> to-time 721.5
== 0:12:01.5

Henrik: this shows, that time! and decimal! datatypes are a bit "incompatible". 
The time! values are able to represent 0:12:01.4 exactly, while decimal! 
is actually a binary IEEE754 floating point representation, that 
is unable to exactly represent 721.4

Romano, I am seeing some weird face bugs, too, using View 1.2.107

Maxim: "why can't AA fonts be supported on Linux ? they're supported 
on windows..."  ;-)

[Cyphre | Gabriele]: Do you still need more information about the 
Win98 crash or is it solved?

If needed I can send an echoed trace log file to your email address.

Here is what I get with 1.2.48 (standard draw) and 1.2.107 (AGG draw) 
:
http://membres.lycos.fr/didec/images/draw-text-48.jpg
http://membres.lycos.fr/didec/images/draw-text-107.jpg

I have tested on Win2k and WinXP with French Regional settings : 
"French (France)" with "French (France)" keyboard layout.

I don't know the exact code page, 1252 or 850 ? Where can I get it 
?

sqlab: the Win98 bug has been fixed for next release. Thanks for 
your useful bugreport!

Vincent, DideC: thanks for the report. I'll try to investigate more. 
But remember the current font handling in Rebol/AGG will be removed 
 for the 1.3 version. (AA and vectorial fonts will be added once 
we found good crossplatform solution for it)

Romano, Anton: the new error!s were introduced to rebol instead of 
"face in more than one pane" and simmilar console prints.
regarding this error
** Script Error: Invalid graphics face object
** Where: view
** Near: show scr-face
if new [do-events]


I have got it when one of face/size coordinates are lesser than 1.

this is from changes documentation:
Face related warnings like 

face in more than one pane" and "invalid face" are no longer printed 
to the console. They now throw errors and you can catch them or let 
them go to the console. The warning that a face is shown in non-visible 
pane or closed window type errors have been removed. Programmers 
can determine that for themselves."

#3687 : bitwise ops - it was submitted at the start of the /View 
1.3 project (2003/2004). 

Both MacOS 9 and Amiga /View 12.1 (big-endian MC 680xx / PowerPC) 
have this bug for bitwise operations on series.

I had to do a workaround for %gzip.r (painful slow byte per byte 
operations) and %rebzip.r (calculations with integers.)

maxim: it isn't just AA. font support is not multiplatform right 
now so it can't be left in. i'm sure we'll find a solution to this...

One could build a function that catches "face in more than one pane" 
errors that duplicates the face and ties the data values of the faces 
together so that they act like one face but are in fact two faces 
with identical values and therefor allows view to display the face 
in more than one pane...

I noticed a crash with rebol 1.2.107.3.1 on win2k with the script 
logo.r (tet program for AGG)  when I resize the windows ...

the script is http://www.rebol.net/tests/view/logo-maker.r

the crash message is :

** Script Error: Invalid graphics face object
** Where: evt-func
** Near: show out
>>

btw, i have a question: is there anyone using smtp:// directly?

maybe graham with rebmail ?

but i think is using the core VM not the view one

the new error!s were introduced to rebol instead of 

face in more than one pane" and simmilar console prints".the problem 
is that previous versions did not print any messages in the console 
with the same code.

special category of time (only one known value): negative negative 
time
>> time: to-time 2 ** 31
== --596523:-14:-08
>> time = - time
== true
>> time + 1
== -596523:14:07
>> time - 1
== 596523:14:07

Oops, should have actually read the conversation before commenting, 
huh?

romano: indeed, that was a bug, and should be fixed. (though, zero-size 
faces are debatable. vid uses them for sensors, so for compatibility 
that must work, however it doesn't seem such a great idea to me.)

Some bug reports now contain the umbrella phrase "fixed in View 1.3". 
  I would prefer to know the exact version, so I know which version 
to test the bug against, and whether I have that version yet or not.

Can we find out who is doing that ?

Anton try changes? in the lastest consol ... not very detailled I 
know ...

anton: if a response does not have "-Gabriele" in it you can assume 
it was written by Carl :)

about #3016: does it still apply? (PARSE and recursion limits have 
been changed much lately) could anyone provide me with an email that 
breaks the parser?