World: r3wp

>> i: 0 a: func [z][i: i + 1 a 1] a 1                    == 1704
>> i: 0 a: func [z y][i: i + 1 a 1 2] a 1 2            == 1706
>> i: 0 a: func [z y x][i: i + 1 a 1 2 3] a 1 2 3    == 1705

One stack for calls, one other for parameters perhaps

all is possible

possibly the stack includes a single payload block for calls... this 
would make sense, in preventing stack depth variance.

meaning parameters are not stored in the call stack :)

Carl would not be using GCC for Win builds, because REBOL is older 
than the short time period that GCC has been good on Windows

It would also depend on platform-specific stack handling.

possibly, yes, by indirection.  push stack [func args-block]

and args-block is a mem copy of the args block with local values

this is all speculation though... hehehe

what I am surprised overall is that rebol does not simply increase 
stack space.  isn't that possible in current OSes?

I mean, the interpreter, dynamically.

Function argument stacks are handled in function-specific value frame 
stacks in R2. They are stored in stack frames in R3, which changed 
the stack allocation to be more task-safe and have faster function 
calls.

Brian, i would be very suprised.

Rebol uses his own way to handle stacks, if not it would be unportable.

I think that last post from brian wasn't speculation... I think he 
actualy knows, being in the R3 internal loop  ;-)

i mean it's not platform specific.

This is one aspect of the internals that I actually am aware of, 
since it affects contexts. It's true: REBOL implements its own stacks, 
and doesn't use the C stack to call REBOL functions, just natives.

The platform-specific aspect could be affected by page and/or pointer 
alignment.

well, it an easy deduction, if not Rebol could not be portable

REBOL is not fully portable. The platform-specific parts are wrapped 
in APIs internally.

yes but the VM need to be fully portable

the core engine must be

This is why the R3 host code is separate, and will be open-sourced. 
The host code is non-portable. The VM can call API functions provided 
by the host code. This would be necessary for a variety of reasons, 
not the least of which is that memory management, tasking, networking, 
any number of things are implemented by different platforms differently.

http://www.openldap.org/lists/openldap-devel/200304/msg00123.html


Anyone made a REBOL version of this? It's a UTF-8 <-> ISO-8859-1 
converter in C.

hasn't peterwood just uploaded a pretty nice string en/decoder on 
rebol.org?

will check

Peter's code detects the encoding, and can do several comversion 
between encoding types:
http://www.rebol.org/view-script.r?script=str-enc-utils.r

wow, very nice. thanks.

used on rebol.org I suppose ?  :-)

peterwood can I make a wish?


I really need a function which can detect if a file is binary or 
text... do you think its possible to have such capabilities in your 
module?

I have a printable? function that checks if a string has only ASCII 
printable characters. Would that meed your need, Maxim?

its already very usefull, its for detection of how to process files 
in distro-bot.  right now it just treats any non rebol script as 
a binary, but I'd like to add some processing to other text files.


maybe peter can extend the concept and figure out if there are only 
printable characters in other encodings  :-)

just "discovered" 'alter' fuction...very handy

Sadly, 'alter has not made it into R3:
http://www.rebol.net/r3blogs/0124.html

Doesn't seem a very suitable name for what it does ...

How about 'swap instead ?

>> flags: []
== []
>> swap flags 'resize
== true
>> flags
== [resize]
>> swap flags 'maximize
== true
>> flags
== [resize maximize]
>> swap flags 'resize
== false
>> flags
== [maximize]

Does it read better than 'alter ?

'alter was always a poor choice of name -- short for 'alternate apparently, 
but too short to be obvious.
Anton has suggested 'toggle as a better name for it.

toggle implies something is always there ... but in a different state. 
 Swap is to imply that something goes in and out :)

Rotate is another possibility

Never had the need to use this word though :)

I've used ALTER once for user selection in a list, and it was even 
a special case. I think having the ability to just conditionally 
append to a block if the value does not exist is more useful. I.e. 
an ALTER, split in two functions.

Trying to think of where I could use this function ...

>> a: [ b b c d ]
== [b b c d]
>> alter a 'b
== false
>> a
== [b c d]

'alter is a hard one to name. The operation is:

 *   if the target value exists in the series: remove the first instance 
 of it
 *   if it does not exist in the series: append it to the end.

The "remove first instance"  makes it more generic than rotate/swap/toggle 
when there is more than one instance of the target in the series.

flag/deflag ?

'cull almost works :-)

An english defintion being: "To take someone or something (from somewhere)."

'cull would be a great name for 'remove-each : )

cull deer herd [deer = not smart]

ALTER is still in R3, though its usefulness is so limited it may 
be moved to R3/Plus. The enhanced logic flags that were supposed 
to replace it were removed in the latest release - they were never 
refined or documented to the point of being useful.

In theory, ALTER is a set function, so the series in question is 
not supposed to have duplicates.