World: r3wp

For that matter Volker, you could set view-root to the standard sandbox 
directory like this:
view-root: join to-rebol-file get-env "APPDATA" "REBOL"

Yes, but then i get security-questions.

And i dont like to rely on user.r . personal preference.

Yeah, those security questions.

As for relying on user.r, I prefer rebol.r for things like this.

On the plugin-side it can never be secure enough ;)

Hmm... should check that out.

and having a %public/alongside my /core-script work quite well for 
me.

Sure, why not (don't answer that).

I try to stay out of user.r because rebol installer has its way with 
it. I suspect rebol.r is also in the same category. You could lose 
your changes. Therefore I prefer to redirect to another script where 
all this action occurs.

i like to poitn peoples to install rebol and run a little launch-script 
i send them.

No, rebol.r used to contain the feedback function and so was included 
in the Core package. It is not written automatically. You can keep 
control of it if you want.

and have that run without problems.

It was supposed to be that global settings were contained in rebol.r 
and user-specific or local settings in user.r, but it never worked 
that way with Core because REBOL only looked for the location of 
those files once for both, rather than once for each, so you couldn't 
put user.r in a user-specific place and keep rebol.r is a global 
place. VIew does it right with version 1.3 though.

set-net [ "[carl-:-rebol-:-com]" smtp.rebol.com ]

should return an error ... but it doesn't.

as this happens

** User Error: ESMTP: Invalid command
** Near: insert smtp-port reduce [from reduce [addr] message]

Anton, why do the actual numbers for FOR matter? The thing is, those 
numbers are of type decimal! as my app has to use decimal! values 
for calculation.

Has someone a nice function to extract tuples by level? For example:
	1.0.0
	1.1.0
	1.2.0
	2.0.0
	2.1.0


I just need all 1st level tuples, than only the second level tuples 
etc.

tup: 1.2.3
i: 2
probe tup/:i
; ?

I need the function to return only 1.0.0, 2.0.0 or 1.1.0, 1.2.0, 
2.1.0

All tuples that have a specific level set.

I'm just playing around with a multiplier pattern like zero? (1.0.0 
* 0.1.1). This works expect for the 3rd level.

Robert, I was just wondering what actually triggered your initial 
question.

I thought perhaps we could refactor your original code to use WHILE 
or REPEAT etc..

I'm not arguing with your main point, which seems to be that the 
current behaviour of FOR is uncomfortable in this respect. (Or is 
it deficient ?)

Robert, how about this ?

tuples-of-level: func [level tuples /local compare][
	result: copy tuples
	remove-each tuple result [
		repeat n length? tuple [
			compare: get pick [<> =] n > level
			if compare 0 tuple/:n [break/return yes]
		]
	]
]

; test
blk: [1.0.0 1.1.0 1.2.0 2.0.0 2.1.0]

tuples-of-level 1 blk  ;== [1.0.0 2.0.0]
tuples-of-level 2 blk  ;== [1.1.0 1.2.0 2.1.0]
tuples-of-level 3 blk  ;== []
tuples-of-level 0 blk  ;== []

This version should scale better.

tuples-of-level: func [level tuples /local compare][
	result: copy []
	foreach tuple tuples [
		if repeat n length? tuple [
			compare: get pick [<> =] n > level
			if compare 0 tuple/:n [break/return no]
			yes
		][append result tuple]
	]
	result
]

Cool... I had this idea with the comparators too but not to use pick 
and the level as the selector. Cool stuff!

:)

Has anybody noticed problems with files that have "%" in name?

Yes, can't open them in WinXP.

No, yes you can.

read %"Hi%there.txt"     <-- this filename works

yes, but load mold %"Hi%there.txt" does not

>> load mold "Hi%there.txt"
== "Hi%there.txt"

what version of rebol, what os ?

1.3.2 on WXP

no, sorry, you're right. I missed the initial %

I think % should be automatically translated to %25 (like space is 
translated to  ) to prevent these problems

It does not mold into a loadable string. --> RAMBO.

OK, I'll post it

Note also this bug!  http://www.rebol.net/cgi-bin/rambo.r?id=3189&

well, that's really strange. see this >>
>> load mold to file! "[s-:-a]"
** Syntax Error: Invalid email -- %[s-:-a]
** Near: (line 1) %[s-:-a]
>> load mold to file! "[aas-:-df]"
== ª[s-:-df]

everything in range from "aa@.." to "ff@.." produces the second result

Add this information with the link to #3189 to your report. They 
should insert it into the original bug report.

these ways can be used to circumvent the filename character problems 
if needed:
load {%"aaa%aa"}
load "%aaa%25aa"

how to catch following error? error? try [to-date 29-Feb_2006]