World: r3wp

He's in deep think mode

or, he's drinking too much wine ... not sure

hum I vote for the 3rd forget the damn carret and find something 
better :P

text handling is so basic and old

well ... most of my best achievement where done in drunk stage .... 
So keep  going on Carl  ^^

maybe that's why your demo seems fast enough??

after the drunken master the drunken coder  like in kung-fu in coding 
the more you drink the best you are  ^^

the drunken master isn't drunk!  He's just pretending...

hum GRaham hum that's cause i didn't drink anything .... my neural 
activity is too slow i need to test every single line to be sure 
i'm not missing something

I test every word ... makes me slower

for vid what I like the most in most ofen case is not to have to 
write 1000  lines to display a button and what I hate is to have 
to write 10000  when i want to do exotic faces   ^^

Yes, that is the problem

It's an unfinished opus

yes  .... but that's a pitty because it's really open and powerfull

but text selection in my RTE-line project will makes me face the 
same dilema

but I think I will set an internal var where you can find selected 
text and 2 other internals var with offset start and end of the selection

like: rte-face/selected-text rte-face/sel-start rte-face/sel-end

i need to implement so more things to get a complete multi line rte 
that's giving me a head hach

and more i need to find a good equation to remplace the damn switch 
for cursor motion

remove don't work the way it should be too ... that force me to do 
a foreach loop and that's slowing the delet process

example: >> a: [[a][b][c]] remove a [b]
== [b]
>> probe a
[[b] [c]]
== [[b] [c]]

but way [a] is gone and why [b] is still here ....

remove that means the passed argument is delete from the list and 
nothing else

remove a [b] == remove process done [b] have been removed from the 
list a. Have a nice day shadwolf  you know you really rock  !!!  
probe a  == the list a contains [[a][c]] have a good day shadwolf 
you still rock you know i love you .... That's would be  the best 
^^

REbol3 the language that really talks to you and really love you 
  ^^

if after that rebol is not the main big computing thing I really 
change my professsion to anything else ...

ok i made a quick patch for the bug graham noticed to me ( the right 
arrow not going  on the right side of the last char) you can find 
the new version here  http://shadwolf.free.fr/RTE-line-Shad-03.r

the formula I use in the patch is (font-size of the last entry in 
the char table  / 2)  -1

cursor motion with mouse actionh ttp://shadwolf.free.fr/RTE-line-Shad-04.r

http://shadwolf.free.fr/RTE-line-Shad-04.r

This demo is cool. I think you just need to replace probe with an 
empty function or comment it out so it runs as fast as it can.

hum but it's fast on my computer and that's a pre-alpha double dash 
++ X 2008 version  (that's way there is the probes) If I go enough 
far in the process obviously I will do a clean share. That's just 
to show  wich direction it takes and puts some animation here while 
we are pending for news from R3 ^^. I plan even to declinate it to 
wrok with rebGUI and do a synntaxe colored widget wich could be used 
for an IDE for example

I will think in a better data scheme too because the way it's designed 
it's not to handle big text amount anyway ...

on my previous intent i was handeling the text as a subdivision of 
same way formated strings of char  and that was a pain then to get 
a precise text cursor motion and to insert new chars into an already 
existing and formated string

so with the actual way to handle the data each char haves it's own 
position and format stored that enhance alot the précision and the 
posibilities but the side effect is to have to handle much more data

a way to reduce the data scheme will be to hum encode them into a 
fixed size string for example a12black for arial black 12 pt char 
 instead of storing for each char the corresponding values as splited 
 thing

i use as "index" for my seachies into the list the offset position 
of each char displayed

so my idea to speed the search process instead of having to use foreach 
loops would be to convert my list into an hash table like data structure 
using the postions as index

and i need to find a good way to handle multiline fonctionalities

in multiline statement the most problematical issue is to deal with 
the multilne text selection and to deal with  the text wraping

some more evolutions http://shadwolf.free.fr/RTE-line-Shad05.rthis 
version gets some bugs corrections, no active probe and the handling 
of the del. key

http://shadwolf.free.fr/RTE-line-Shad-05.r

No Rebol3 Frontline blog entry for 57 days
No Rebol3 updates to Docbase for 70  days
Grape picking season approaching

perhaps something has been brewing in the vats?

yep very quiet

It would helpful if there were some community driven projects for 
designing R3 protocols. This is something that Carl doesn't need 
to be involved in until final inspection to see if it can be added 
as standard to R3. Although it's been said by Carl that he would 
change the TCP ports a bit as they were found to be too simple, I 
don't think it would be much of a hindrance to work on simpler protocols.


I wouldn't mind seeing out of the box support for LDAP, if it can 
be done in 4-5 kb of code. I know DICT is being worked on.

(I know next to nothing about LDAP, so 4-5 kb of code may be ludicrous)

The full feature set of LDAP is quite complicated, but I think a 
basic interface would already help a lot

Even if it were just read-only at first

I just realized, yes it's quite complicated. :-)