AltME groups: search

and in so much as rebol runs as a command line client, then it can 
be readily done.

when you use file! call does a to-local-file automatically, and wraps 
it in quotes.

Looks like Sterling's detach.r script in the library to detach attachments 
from emails has a bug.

the first line is redundant, and the second line can be changed to 
a parse ...


   if parse headers/content-type  [ thru "boundary=" opt {"} [ copy 
   boundary to {"} to end | copy boundary to end ]][
        ;remove back tail boundary
        print ["Boundary string:" boundary]
   ]

which I think fixes it.

Hmm. Boundary won't have a value if the parse fails ... so

either parse ...  [

][
	 boundary: none
]

Hello! I'm receiving the following error when processing a cgi form 
to send an e-mail of the response.  I correctly can send from my 
laptop, Rebol/Core -- command line, with my set-net settings in user.r 
but when I do the same thing on my hosting account I get the following 
error:

Hello!  I'm having trouble with my hosting account to send via e-mail 
a cgi form response.  I can get rebol to post the data to the page, 
I just can't get it to send the data to me via e-mail.   I'm able 
to send e-mails from my laptop but when I use the same set-net settings 
on my host account I get the following error: ** User Error: Server 
error: tcp connection failed
** Near: smtp-port: open [scheme: 'esmtp] 
either only

Hello!  I'm having trouble with my hosting account to send via e-mail 
a cgi form response.  I can get rebol to post the data to the page, 
I just can't get it to send the data to me via e-mail.   I'm able 
to send e-mails from my laptop but when I use the same set-net settings 
on my host account I get the following error: ** User Error: Server 
error: tcp connection failed
** Near: smtp-port: open [scheme: 'esmtp] 
either only

most smtp servers do ip filtering on the input.  this allows them 
to know who is sending the email to them and will only allow ips 
they serve to connect.


since your hosting account is probably on a remote server, it won't 
be allowed to send data via your home smtp server account.

when you say hosting account, you mean a server on the net with its 
own dns/ip which is different than the ip being served to you by 
your access provider... yes?

Yes, it's a shared hosting account; my provider is GoDaddy.

SMTP Relay lets you send email messages through your email account 
using your existing email service. For example, you can continue 
to use Microsoft Outlook to compose, receive, and send email messages, 
but the actual email messages are processed through our SMTP relaying 
services. This lets you bypass ISP restrictions on your outbound 
email messages and allows you to use your professional looking "[sales-:-coolexample-:-com]" 
email address rather than a general “[sales-:-ispname]” address.

so, von, if you have a godaddy mail account setup, you can connect 
to its smtp server directly. 


is your hosted machine's set-net function setup to connect to something 
like [smtp-:-godaddy] ?

well, doesn't it ask you for a password??

esmtp requires a password

so you need to encode the password somewhere if you're going to use 
a script

well, then you need to a trace/net to see where the issue is

do you have a shell account?

make it a cgi script and see what is written to %log.txt

would there be instances where write/lines/append would write a quarter 
or half a line? I'm logging tests of several script instances into 
the same file and write/lines/append sometimes produces only half 
a line in the log.

could you create something like a trie in rebol or would you have 
to go lower level for it to be normally eficient?

(let's say I want to use key-value (string-int) pairs for 5M words 
.. hash tables are probably more memory consuming for such a big 
set of data?)

hash tables for such a big set are the only way to go... they will 
be magnitudes faster on access.

but when things are in the millions, sometimes using a disk-based 
on-demand caching algorithm is fastest... it really depends on the 
application.


cause think of it this way.  every byte used by each element becomes 
a MB  so adds up quickly.


5 million pairs of (10 byte) strings and pairs... is just about 350MB 
!

 >> b: make block! 5000010
>> m: stats
== 84172417

>> loop 5000000 [append b copy random "1234567890" append b random 
10000000]

== ["5862713409" 4765171 "2546013987" 2726704 "9528013746" 3565380 
"4591302786" ...
>>  stats - m
== 348435008

Maxim .. thanks a lot for your answers.. very interesting .. I know 
from distance how hashtables work internally but I don't know details.. 
should a block take roughly the same space as hashtable of the same 
block (in rebol) or factor(s) different?

hum... lets see:  ;-)

a: stats
b: make block! 5000010
print stats - a
== 80001039

a: stats
b: make hash! 5000010
print stats - a
== 80005071

>> a: stats b: make block! 1000 repeat i 1000 [ append b random "abcdef" 
random 100000 ] print stats - a
48671

>> a: stats b: make hash! 1000 repeat i 1000 [ append b random "abcdef" 
random 100000 ] print stats - a
81454

but... filled up....

b: make hash! 5000010
m: stats

loop 5000000 [append b copy random "1234567890" append b random 10000000]
print stats - m
== 188430448

here its half the space.


a ha!  depending on the string input... hash tables can actually 
be smaller...   :-)

stats is a cool command , with many refinements also .. I didn't 
know about it

in REBOL, we're a newbie a few minutes... every day.... even after 
a decade of using it   ;-)

I am nevbie a little longer each day :)

hm.. I have a very newbie question .. do you most effectively add 
new pairs to hashtable by appending to it as a block ? can't figure 
out how to change a value .. set doesn't work that way

a: make hash! [ "33" 33 "44" 44 "55" 55]
select a "33"
change find a "44" ["88" 88]
== make hash! ["33" 33 "88" 88 "55" 55]

Transfering memory based hash! (map! in R3) datatype into disk based 
shema automatically keeping the hash table computation and lookup 
hidden from user gives you a RIF. Holly grail of all rebollers :) 
long long time promissed, still waiting to be done. Anyway hash tables 
are always usually unsorted, when necessary to search in usually 
some type of additional index is used (B-tree for example), for simple 
information if the key is in the set, bitmap vectors are used with 
advantage, when the set is really big (and bitmap vector doesn fit 
into memory) comressed bitmap may be used and usually bitwise operations 
on those vectors are much quicker than on uncompressed. 

Thisi is why it should be used for bitset! datatype anyway. The number 
of byte aligned (BBC,Packbit,RLE)od word aligned (WAH) schemes exists. 
 It is used in very large datasets when index also resides in disk 
file. Once again bitwise operation may be much quickier even in memory 
on those schemes.

For those interrested a Fastbit webpage is good source of docs.

I will be rebuilding the callback example with a much better/simpler 
design. but they work very well, basically I have mapped the Reb_Do_String() 
and Reb_Print() functions so that they can be called from within 
any extension.

I am also building little helper funcs like a REBOL datatype centric 
version of sprintf  which acts a bit like a C-side rejoin for REBOL.

this way we can create rebol code directly from strings and native 
data very easily.


there is currently a size limit on executed strings, its a simple 
question of optimisation.  this means we can't use the wiredf function 
for creating large datasets via strings (for now).

but I'm already doing stuff like:


wiredf("rogl-event-handler make wr-event [new-size: %p]", win-w, 
win-h);


calls rebol's do with %p replaced by a pair, using 2 ints.  this 
is a varargs function.

i was just thinking again about the idea of IF (etc.) keeping a reference 
to the condition argument for you, that is, so that instead of writing:

    if x: select block value [do-something-with x]

you can write:

    if select block value [do-something-with it]


The reason people say it's not worth it is usually that of having 
to bind/copy the block - you don't want that in every IF call and 
probably not even in the ones where it would be useful (and, there's 
really no other name you could use for the function).

IT could be a function that returns the thread-local top of the stack 
of implied subject values. IF would then push a value on that stack, 
and pop the value off when it returns. Might be tricky to make error-throw-safe, 
but easy to make thread-safe :)

A *lot* of code uses the trick of having IF or UNLESS return none 
when the condition is not met, so your other suggestion is unlikely.

A *lot* ?
somewhat exaggerated :-)

For complex control flow rules, i rather prefer CASE.

Most of the time, combitations of ALL ANY, can be replaced by a CASE 
structure (which is faster and more readable)

Gabriele, it occurs to me that if IT was native it could look up 
the stack to get its value. I'll try writing a (security hole) REBOL 
version of the function later today - it would require debug privileges 
to run so that it can call the STACK function.

a sort of native POP function

I don't know how the values evaluated are stacked by the VM. But 
i see the advantage of having a POP function.
We could easly create postfix functions.
e.g:
CONCAT: func [v][join pop v]

>> "a" concat "b"
=="ab"

All sort of new operators would be easy to construct

Steeve, R3 evaluation doesn't work that way - it's not a stack machine.

I have an old IF-IT function, which just does a bind/copy. I used 
it a bit when I first wrote it, but it hasn't become a part of my 
daily life.

Brian: a lot of code uses IF returning none, agreed, on UNLESS i'm 
not really sure, it's quite new. besides, it's not like R2 scripts 
run unchanged on R3; but anyway i was just thinking out loud, not 
really proposing anything.

This function returns a copy of everything until it finds the value 
specified and then it breaks.

a: [b c d]
copy/part a find a 'd
== [b c]

Brian, even if the Rebol's VM  is not a true stack machine. It has 
a data stack, so that, the POP function could be emulated in some 
way.

A forth kernel, is nothing else than that. To simulate a stack machine 
on a processor that is not designed that way initially.
However, your response suggests that the cost would be high.

The reason Steeve's proposal doesn't work is because the result of 
the prior expression is thrown away, not pushed on a stack.

Sorry Bolek, I should have been more specific. I meant Steeve's POP 
proposal and the CONCAT example wouldn't work (for reason's stated 
above). Gabriele's IT proposal and Steeve's sample implementation 
of it would work a little, but would need modification.

There are 45 scripts on FREBOL.org that appear to use UNLESS (there 
may be a few false positives in this -- if, say, the word UNLESS 
has been reused):
   http://www.rebol.org/search.r?find=unless+[b]

I was reading http://www.chalicegames.com/swym/SwymWebIntro.html
and some concepts were interesting to me(especially ETC), so I made 
REBOL equivalents:

http://box.lebeda.ws/~rebolek/rebol/swyv.r


There's a documentation in the script, so just few examples of what 
it can do:

SERIE:


>> serie [etc 1 2 4 .. 20 cycle [1 2 3] length 5 iterate [x: x + 
10] from 10 5]

== [1 2 4 8 16 1 2 3 1 2 10 20 30 40 50]

COMPARE:

a: [1 2 3 4 5 6 7 8 9]
b: [2 4 6]
>> compare a b [some a > every b]

== true


>> compare a b [one a > every b] 

== false

FILTER:

>> filter serie [iterate [x: x + 1] 10 ] [[x > 2] [x < 5]]

== [3 4]

>> filter etc [3 6 9] 100 [x > 250]

== [252 255 258 261 264 267 270 273 276 279 282 285 288 291 294 297 
300]

>> filter serie [1 .. 10] [[x > 5][zero? x // 2]]
== [6 8 10]


It's written in R3 but should also work in R2 (not tested). It's 
not optimized so if you're interested in it, feel free to do whatever 
you want to improve it (more patterns that ETC can recognize...).

Rebolek: thanks for the link, lots of good food for thought here. 
That would be great to support it at native level (using a R3 extension). 
I also wonder how much of it could be implemented efficiently using 
'map-each and 'apply. Anyway, this could be a really great addition 
to R3 (or even R2). Keep up the good work.

Well it's interesting as a study dialect. But to be honest guys, 
i don't see the interest  to have them in Rebol.

Because we can do much of the use cases rebolek showed us  with one 
or two lines of rebol code.

And i don't need to say that it will got lightning speed by comparison

But anyway, It's lot of fun to do such things with Rebol.

Bolek, that's very interesting for me because I was searching for 
just such a declarative dialect for sound generation and music composition.

Pekr, vectors are really great. But they need few improvements and 
 bugfixes here and there. I wrote a document what doesn't work and 
should some time ago (has been two years already? I think so). I 
haven't looked at them recently, so maybe they're improved already. 
I should check my R3 AIFF/WAV loaders/savers wheter they work as 
they have been the best test for vectors I had.

Smart guys. This may seem elementary but I need to check if certain 
ports are open on a windows machine. For example, port 8881. I use 
something like error? try [close open to-url "tcp://:8881"]  (building 
these strings with various port numbers). My problem is I don't know 
how to check if it is working. I turn on the firewall and it doesn't 
seem to make a difference. Perhaps my thinking is all wrong and all 
I am doing is checking within the firewall. Any thoughts?

all that does is trying to open a server port

if you want to check if there is a server port listening then you 
can do

open tcp://localhost:8881


and if you want to see if that port is open to the outside, then 
you need to use another PC to probe that port address

Thanks Graham and Sqlab. One step further if you please. What would 
you suggest the steps would be to test the code. Right now when I 
run the test I get no opened ports. I've turned the firewall off 
and on but the results are the same. 

I've been studying the nettools.r code and thinking that maybe I'm 
not thinking this right. What I want to know is if a certain port 
will allow it to be open so that this particular application has 
can use.it. What is happening is customers are installing the app 
and having trouble because these certain ports are unavailable. What 
I wanted to create was a quick and easy pre-install test to verify 
these ports were open. This partly due to the fact that they way 
the software was written it doesn't tell you that a closed port is 
the problem. It simply stops worting. 
I appreciate the feedback you have given.

what the code checks for is if anyone is listening at a certain port 
...not whether the firewall is open or closed.

If the firewall is off, but no one is listening ... then you'll get 
a closed port
same as if the firewall is on, and someone is listening...

what about R3 status ? make uptodate  R2 is very fine but i would 
prefer a R3 version with GUI and ODBC or MYSQL ..................;

some miracle

 in this case meaning a community member with the time volunteering 
 to do the work.

do you have a plan for the beta realase ?  one week, month,

It's a little flexible - outside circumstances have affected the 
timing already.

i am just like an end user on R3.  and for me i am not a guru like 
you and others !!!!!

There will be no miracle that will bring the GUI or database to the 
first R3 release - they just aren't done yet, and can't be done with 
the level of community involvement that the alpha releases have engendered. 
Too many critical people are waiting for a beta or full release before 
they will even start to get involved. Fortunately we are on the rapid 
release model, so there is no such thing as a "final" or "full" feature 
set, just the feature set of a particular release.

If non-gurus can't use the GUI, it's a design flaw that needs to 
be fixed.

Ok, let's create a new r3 group ...

anyone got a code color rinser for R source?

>> a: [1 2]                 
== [1 2]


>> reduce [a swap a next a]

== [[2 1] [2 1]]

Why it doesn't return [[1 2] [2 1]] ?

This works as I expected:
>> a: [1 2] 
== [1 2]
>> compose/deep [[(a)] [(swap a next a)]]

== [[1 2] [2 1]]

Hm, I understand that now:

>> a: [1]

== [1]


>> reduce [a append a [1]]
== [[1 1] [1 1]]

Interesting. I never knew that.

Regarding capturing of globally defined words in a local context, 
as Gabriele likes to say : "for every code you would write that would 
work for you, I can write an example code that will break it" ;-)

setting no global words: use a context

I imagine when you now set some word in some function it looks at 
/local words and it it's there it creates a local word, if not it 
just sets a word (globaly).. let's say that there is funcstrinct 
that in second case just throws an error (probably something like 
that could be made in rebol itself)

Henrik: but I don't want to make each function a context (probably:) 
) .. I have to admit I don't know much about contexts .. only that 
it's like object

I am changing to using a context now for all words that are defined 
outside functions on rps pages to make them local to that pageload

Just remembered about querying system/words, that would give you 
a mean to detect new global words.

>> query/clear system/words

== [end! unset! error! datatype! context! native! action! routine! 
op! function! object! struct
! library! port! any-type! any-word!...
>> context [set 'a 5]
>> query system/words
== [a]

Doc .. thanks a lot for that query word ... it' awesome to see on 
a page what all got set in the process.. this will help me the to 
make code more strong A LOT!


Doc.. how does the runtime binding to function local words work? 
func is probably not a  mezzaine or something where we could peek 
into what it does with /local words ?


Henrik: I need to learn more about them and how to use them .. are 
there any good docs to read about what contexts are used for maybe?

Janko, a function is a context! value like objects. You can use the 
following mental analogy to see how it is related :

foo: func ["for demo" a [integer!] /local b][...]


would be *roughly* equivalent to constructing an object like that 
:

foo-def: make object! [
	hidden-ctx: make object! [a: none local: none b: none]
	body: [...]
	spec: ["for demo" a [integer!] /local b]
]


The body is bound to the 'hidden-ctx context at function creation. 
When calling 'foo, the interpreter will set the 'hidden-ctx object 
words values according to passed arguments and refinements and then 
DO 'body.


There's no differences on how REBOL treats "arguments" and "local 
words", it's part of the illusion. The /local refinement is used 
by *convention* only, to set "local words", you could just decide 
to use any other refinement for the same job. Here's an example :

>> a: func [/local b][print b]
>> a/local 5
5


Additionnaly, when you apply the ordinal natives on a function! value, 
you get :

>> first :foo
== [a /local b]		;=> the hidden context words

>> second :foo
== [...]			;=> the function body block

>> third :foo
== ["for demo" a [integer!] /local b]	;=> the original spec block

trying:

source context

might be a revelation too.

ok .. query/clear will find me seek out leaking globals so this problem 
has got a solution in a way

Object! and function! have different creation and usage semantics, 
AFAIU, they share a common internal parent datatype, context!. So, 
context! (which is not directly accessible) should be "cheaper". 
Objects and functions have different purposes, so this might be like 
comparing apples and oranges...in a closed blackbox.

BIND on a block! does a full recursive traversal of the block! and 
for each word! found, does a fast lookup in the target context (probably 
hashed). So the cost is directly proportional to the size and depth 
of the argument block!.

well, how else would you do it? I guess you need kind of an escape 
sequence to evaluate a word or something.

Terry, show use case you think is cumbersome and we'll show you a 
simpler way (i hope)

<<< is a here-doc in php

so.. the result is a button that alerts.. 

\'hello\'s {Worlds \';

...and prints...

 'and {so "on';

after the button

There has been talk in the past of including a substituion function, 
REWORD being the R3 func for it. In R2 we have build-markup, which 
shouldn't be hard to hack, but I don't know of a version that anyone 
has done for a given substitution syntax.

Now, that's norribly naive, and doesn't work because of that. e.g. 
it needs a space before the $ marker, so a var at the beginning of 
the text gets missed.

yes seems a little messy Greg ;-)

Needs a different name too, as Ladislav has a nice BUILD func that 
works on blocks.

I need a spec first. ;-)