AltME groups: search

Accepting Gregg's response as a second, for a call to end voting; 
 I'd like to motion that we tally up the votes as per the user.r 
Chat example.  Thanks to all participants, and best of luck to our 
candidates.

We'll all pitch in and have someone geographically close to each 
key player get them a nice beverage of their choosing.

I'd like to nominate Nenad Rakocevic for 2010's rebol Of The Year.


As a reminder, for Bob's Rules, this nomination will require a second.

I'd like to propose a round table vote of acceptance of the voting 
code and procedures posted to user.r Chat, by Sunanda. A negative 
outcome would hopefully be followed by new proposals and discussion.

Point of Information:  Recapping current nominations.  Robert Muench, 
Nenad Rakocevic, Nick Antonaccio.  Participants are encouraged to 
get nomincations in for seconding before December 15th.  Starting 
a vote pass by Dec 15th should give everyone a chance to get in a 
vote before the end of Dec 31st.

How would I implement a ssl:// port scheme in the hostkit, that behaves 
just as tcp:// does, but allowing me to hook the native actors (open, 
mainly)

Start is a cmd.exe builtin command on Win2k+ (not Win9x). Try call 
"cmd.exe start textpad" or calling Textpad directly.

Jeez i wasted my time to develop a script to launch remote commands 
from an AS400 on my PC using FTP.

And I fail on the finish line because CALL don't do the last easy 
part of the job.
T_T

Or something like that - I use a different text editor.

you can call a batch script ..

so you're using a AS400 to remote control your PC??

a little strange to use ftp to do that, but it's the only protocol 
allowed here :-)

the reason Notepad works with a direct call is that it is in the 
system path in windows.  Textpad would not, but the path could be 
added, then you wouldn't need the direct path to it.

but when i start a console, "start textpad" works.

So, the default console has the path of the installed apps in his 
env.

As long as you refer to the manual as a "manual" even if it is technically 
a wiki too, that will work.

try a rejoin using two binary values.

Looks like a problem with append:

	append "" ["a"]
	append #{00} [#{01}]

Note that join works (as it's not appending a block).

Already posted a ticket about that today.

FUNC in R3 does a COPY/deep of its spec and code blocks, so it's 
safe to use here.

the new R3 web site is SOOOO much better to navigate on..... 


yeah, it doesn't look like a million, but at least it doesn't get 
in your way, like the old site did. 

just tought I'd share my POV.

Max - it is just a structure, which was reshapen, and the page was 
also de-stylled. The layout, styling and gfx of the new site is not 
done yet, as well as front page segmentation for particular target 
groups interests ...

I probably need to put a no-cache meta directive on it.

Correct.  But, I want to findout what people want the most now.  
I think last time "conference" was about posting the Console code 
from R2 as a starting point for that project.

Otherwise a lot of people remain blocked.

Working on compressed modules this week, inline export already done, 
protected header fields waiting on PROTECT fixes. It would be really 
helpful for there to be a discussion about the design requirements 
for embedded deferred init modules.


Really looking forward to host kit changes. Waiting on READ/as and 
WRITE/as before I start improving the clipboard:// scheme.

here's a personal wishlist:


1. a license for the hostkit, which would allow us to publish and 
distribute patches and custom builds. this would be a necessary precondition 
to bootstrap a healthy hostkit ecosystem (if such a thing is desired).


2. documentation of more hostkit intricacies, such as "documenting 
the boot script build method" mentioned on the priorities page. another 
pet peeve of mine would be documentation on how to add new native 
port types (such as ssl). and fixes (if any) to enable creation of 
such native ports by only touching mezzanines and the host kit.

3. a libr3.dylib built for osx.

right now flushing out (finishing) the extensions is actually the 
most enabling.  so: 


1.  Releasing the view host-kit with improved extensions would be 
the first step into finishing WHAT the first R3 beta will be, in 
terms of architecture. Right now, there are a lot of datatypes missing, 
and a proper REBOL callback mechanism *** with return value***  is 
probably the most frustrating issue with R3 extensibility.  image 
and vector types are critical to extensions... please they have to 
be part of next extensions.


2. Schemes!!!   with a lot of work ALREADY done,  plus willing and 
able helpers, Carl, you should REALLY look into this and orient the 
spontaneous team that is dying to get your attention so they can 
proceed FOR YOU.


3. Documentation ... I'd say.. .TWO complete and final pages a day... 
no need to do more.   At that rate, the documentation will be much 
more helpfull within weeks.  But sections of it will start to be 
worthy of being called documentation... right now... I'd call it 
the "sprawled notes" part of the site.

So, is there a documented dependency tree somewhere ....?

requires a value now?

No, it's function argument handling that has changed. RETURN required 
a value in R2 as well, you could just cheat.

yes, use EXIT if you don't want to return a value

Most important from my perspective:

- finish Extension enhancements
- finish Host enhancements
- move View to the enhanced command! interface

both to the state, that ppl can fully exploit those features.

- multitasking


... for Doc starting to port Cheyenne as mostly the only one REBOL 
killer product

- support porting efforts, get us to the mobile devices (ARM)


... as for Console and some other items - yes it sucks to not have 
them available, but I think that having strong infrastructure in-place 
is really important. But overall we need all the priority list implemented 
for R3 being worth calling a beta ...

.... some ppl also requested enhanced /library interface being done 
as an extension. I started by donating 100USD, and now we are at 
someting like 400 USD ppl would pay for such a feature. So, what 
do we sponsor next? :-)

A GUI table would be nice ...

nearly a month after Carl said he was going to work on a new alpha

host kit will give a bit of parallel work done (Robert, Cyphre, BrianH 
and Maxim are all waiting), but I guess it's up to what motivates 
Carl currently.

host kit will fix remaining graphics issues, which Cyphre is already 
preparing. It will also help prepare for a better console for all 
platforms. Robert and Maxim has done various implementations (OpenGL, 
SQLite) and are doing general tests. BrianH, I can't remember what 
he's doing...

Fixed tickets now at 14 and counting :-) A good sign R3 is back in 
development, although over time we do have few tens of tickets to 
implement :-)

btw, I will start working on a public R3 /library extension as of 
next week, its now on my agenda.

a good thing its only february then  ;-D

The idea here is not build a prototype, but a "end-user" product.

I was going to have a look at the R3 gui sources but they're all 
flattened..  Is there a pretty print for r3?

A script can do the job, I forgot where I'd seen it

That produces an error for higher values of 2.

This should produce an int when it can, and a dec when it can't, 
though only for positive values of 2:

   pow: func [a b /local res][attempt [res: to-integer res: power a 
   b] res]

>> length? read http://idisk.me.com/dobeash/Public/rebdb.r
== 12771
>> size? http://idisk.me.com/dobeash/Public/rebdb.r
== none
>> help size?
USAGE:
	SIZE? target

DESCRIPTION:
	Returns the size of a file.
	SIZE? is a function value.

ARGUMENTS:
	target (file! url!)

I've written a few schemes :)

Oh, that's bad: EQUAL? is doing case-insensitive comparison for binaries. 
That's a bug - I'll report it.

Agreed, that's a priority for me too (though not as high as TO or 
THRU charset not working).

I think using append/insert on binary should be a high priority.

Well, I think that problems with workarounds get a lower priority 
than problems without workarounds, but I agree.

I got a big project I want to turn off of R3DBMS and need many of 
these core fixes.

We neef of a trace scheme, that can be stuck with any other scheme.

But At first,  we need to think about a framework to be able to compose 
several schemes together, like the pipe idea someone else already 
proposed.

The previous build worked fine but this one gives me an error and 
I can't execute my functions.  Currently my script has a halt at 
the end and I just run it and execute the functions from the console.

that a change in the latest release?

How do we now define a word that matches a system defined word?

Looks like we have to use system/contexts/exports to call a system 
word that we have redefined in our script.

File extensions: Can we all agree to use .r3 for R3 related scripts? 
IMO a lot of people will use R2 and R3 in parallel for some time. 
Hence, it's necessary to seperate the two to select the correct interpreter 
etc.

I am not sure. There was a proposal to use REBOL3 [] in a header 
of a script ....

That won't help a lot. As Windooze uses the extension to select the 
correct program.

You can use .r3 if you like. Aside from file extensions that are 
used by codecs or R3 extensions, the file extension is considered 
irrelevant if you specify it explicitly. One gotcha: If you import 
by module name and the module isn't loaded yet, the process of searching 
the library paths currently only looks for .r files. That can be 
worked around by including a module or script in your project that 
imports those modules explicity by filename - after that the import 
by name method will use the loaded module.

FUNC does a COPY/deep of its spec and code block, for safety and 
recursion-safe use within functions (see the source of COLLECT for 
an example of this). However, the internal code uses a non-copying 
version to lower overhead. That non-copying version is assigned to 
FUNCO as the last thing, and the copying version assigned to FUNC. 
FUNCO is left defined for power users who can asses the safety of 
not copying their function specs and bodies.

Can you give me a quick example of how to understand those ramifications 
if we don't copy?

make function! [spec body] doesn't copy the spec or body, it just 
uses them (or does a BIND/copy, I'm not really sure). If that spec 
or body contains literal values that get modified later, they get 
modified in the function as well.

None. User-defined datatypes don't work yet. The datatype is a placeholder 
for now.

And before you ask, I don't know what EVAL is used for either, just 
what it does: It does a DO on block! arguments, but returns the argument 
unevaluated otherwise. I've asked Carl what it's used for and haven't 
gotten an answer yet.

TO-TYPESET creates typeset! values. We don't have R2's pseudo-datatypes 
like series! anymore, we now have typeset!, sort-of an immediate 
bitset for datatypes. It speeds up function calling and allows all 
sorts of other things. I backported a fake version of typesets to 
R2 as well in the 2.7.7 release.

TYPES-OF function! returns a block of argument words and typesets.

Docs need more examples.  If someone just went thought the R3 Functions 
page and separted it into a seciton that says NEW in R3 and explained 
those and gave an example for each we would have awesome docs.

There's already been a blog about that recently: http://www.rebol.com/article/0456.html

TYPES-OF works on other function types as well - we have a lot of 
them.

Since this isn't just a R2 vs. R3 problem, it's also a Mac vs. Linux 
problem, etc.

File associations only affect the main script though, not the modules 
it uses. You can't use an Explorer to load a module.

And since the module lookup process doesn't apply to scripts, this 
is a bit of a non-issue.

>> system/version
== 2.100.96.2.5

>> to decimal! 1%
== 0.01
>> to binary! 0.01
== #{3F847AE147AE147B}
>> to binary! 1%
** Script error: invalid argument: 1%
** Where: to
** Near: to binary! 1%

>> to binary! %a 
** Script error: invalid argument: %a
** Where: to
** Near: to binary! %a

>> to binary! 0x0
** Script error: invalid argument: 0x0
** Where: to
** Near: to binary! 0x0


I think pair! should be binary! encoded/decoded something like this 
(in R2 it was to-binary formed):

to-raw-pair!: make function! [[
	pair [pair!]
	/local x y
][
	x: to binary! pair/x
	y: to binary! pair/y
	while [all [zero? first x zero? first y]][remove x remove y]
	append x y
]]

to-rebol-pair!: make function! [[
	pair [binary!]
	/local length
][

 to pair! reduce [to integer! copy/part pair length: (length? pair) 
 / 2 to integer! skip pair length]
]]

>> to-raw-pair! 0x255 
== #{00FF}
>> to-raw-pair! 0x256 
== #{00000100}

Recently I made some tests comparing the speed of  R3 to R2.

After getting results of R3  up to 100 times slowlier than R2 and 
more, I found that parse is now almost unusable for simple parsing 
under some circumstances.

I will later add a Curecode ticket.

Ah, special mode parsing ... you try to get block of lines, correct? 
I wonder if using some parse rules instead would show so much of 
a difference between R2 and R3?

sqlab - 'split is a mezzanine, hence slower imo ...

G: Dont parse some chars many times, parse many chars a few times.

the above code does not run in R3. Dunno if "end" is forbidden or 
became a keyword, but I get following error, if I don't change end: 
to en:


>> s: now/time/precise loop 1000000 [parse "abc^/def^/ghi^/" [start: 
any [end: n
ewline start: | skip]]] now/time/precise - s
** Script error: PARSE - command cannot be used as variable: end:
** Where: parse loop
** Near: parse {abc
def
ghi
} [start: any [end: newline start: | skip...

even with my random generated data, the number of parts should be 
always the same.

But you can also parse the same data a few times. Every times it 
will be parsed in a different number of parts

end is a keyword!

Hmm... tried to parse 4M chars a few times and r2 ran out of memory 
... r3 worked fine

try running following code:

c: ""
 for i 0 255 1 [append c to-char i]

I get a beep here ....

hmm, we are appending char into string ... who knows what it does 
internally :-) In charset, there are some escape/control codes, so 
maybe it causes a beep, I don't know :-)

could be a GC problem? Maybe above "beep" means there might be some 
memory leak?

parse dataset "^/" returns a different value

do that a few times

he just generates a large set of random chars

http://www.rebol.net/r3blogs/0307.html

A bit of work to do for those interested.

I used to run a routine to fix an email first before running import-email 
on it .....  ie. add missing fields etc

I think there is a blog entry about this...

and a chat message .. tcp outages on rebo.net

Here's a newbie question:  In general, do scripts written for R2 
run under R3 as is or are there specific changes one must go through 
first.  I read that many of you are running older projects in R3 
and wondering how you are approaching the changes that might exist. 
Thanks.

Ladislav and I have both written about our early experiences:
   http://www.rebol.org/art-display-index.r?a=R3

Note the use of the RCO object (REBOL Compatibility Object)  that 
attempts to make scripts R2 and R3 runnable.

Thanks Sunanda. I've been holding off R3 as it takes "my all" just 
to write stuff in R2 and generally they are tools that "have to work." 
I feel badly though because so many of you are putting so much effort 
into it. I could definitely give a non-guru perspective. Thanks for 
all you guys do.

hi, would like to know why have this error want i try to import a 
file in R3.

>> import %carte-joueur.r
>> import %carte-db-joueur.r
** Script error: joueur word is not bound to a context
** Where: make do applier make catch if import
** Near: make object! [
    table: 't_joueurs
    oo: make joueur []

Another method of importing modules is supported. If you use do on 
a module file, it will also be imported. This is provided as a convenience, 
because do has been used in REBOL for years as a way to load and 
initialize additional functions and values.

Claude, DO and IMPORT aren't the same thing at all. Scripts (loaded 
by DO) have a different context model than modules (loaded by IMPORT, 
DO or better yet the Needs header). Your scripts above don't set 
their type to module, don't export words, and have no name - they 
basically aren't modules. You should use DO and treat them as scripts 
until you better understand the module system. Ask in the "!REBOL3 
Modules" group here for more details and help, if you like.

(UN)PROTECT/lock is needed for module header security. Compressed 
modules are done and submitted (in theory), just not released yet. 
Delayed init is delayed until Carl explains a little what he means 
by that. The rest of the core design works,