AltME groups: search

If you backup so many files, you should open just one connection 
and comunicate with the server. Not to open and close connections 
for every file.

Read and write is working... at least now:-)

I'm doing an open and copy on the source machine and I'm writing 
to the FTP server using 'write/binary/append

no... and I see, the zip is not complete, I uploaded complete script 
to:  ftp://rebol3:[rebol3-:-69-:-12-:-157-:-73]/test/ftp-sync_latest.r

And it certainly require some fixes

as it douesn't recognize, that the file is already uploaded and with 
same or newer date

there is just a little mess with the source and target path, as I 
was using it in the scenarie, where the target path was accessed 
thru http, it should not be difficult to fix it, but I don't have 
more time now

Smooth-Copy has an adjustable wait mechanism built in, but it may 
make the overall length of the transfer too long.  I'll have to put 
some debugging statements in prot-ftp.r to see if I can isolate the 
problem and figure out how to go around it.

If I find a solution, I'll post the fix and let you know.

there should  be a timeout, yes. Would  trigger an error too, how 
 is that done in the right way with ports? Because i do not know 
what this code  does without a port/subport.

what confuses  me, this code runs also in passive  mode?


an idea of mine is to listen on both ports, control and this one. 
if the server sends the 405, its never processed because we dont 
listen there. Should  listen on both ports and still handle commands. 
But do not know what the  right vars are.  Waiting for the skilled 
 ftp-patchers  before examining^^

In limited testing, my patch above (slightly modified to remove typos) 
works.  If the 425 occurs, after a timeout period the patched protocol 
picks up right where it left off.  But yes, I would really like the 
skilled ftp-patchers to take a look at the problem and see if they 
can come up with a better fix.

I presume I need one server script and one client script

IIRC Robert and the author where chatting  once, maybe  he remembers?

just run ipconfig /all via shell and parse it out ;-)

and probably nobody was expecting, someone will need so much memory

BACK versus past-tail indices. Every index (except for 1) may become 
past-tail during the execution of a script (when its block/string 
"shrinks") and "legal" again (when its block/string grows sufficiently). 
Let me mention the SAME? function as an example of a consistent behaviour 
- no matter whether the index is past-tail or not, the SAME? function 
takes it into account when comparing series. The BACK function, however, 
handles past-tail indices differently than the "normal" ones. any 
wishes/notes/proposals?

Graham: re ports, there's not much art to it.  You try and break 
up access to any particular service according to supported accessors. 
 Almost all built in protocols are wrappers to a TCP port, containing 
all the commands for that particular protocol.  Note that 'read on 
a port uses 4 accessors: init, open, copy, close; 'write also uses 
4: init, open, insert, close; 'make will call 'init, whereas 'open 
will first 'init (unless you provide a port! as your argument -- 
eg. open make port! foo::bar), then 'open.  You can use all the other 
accessors on an open port: insert, change, remove, etc.  And of course 
'query which you can assign to provide information about the port 
without opening it.  I'd be happy for anyone to elaborate or correct 
this description...

The sandbox port I wrote may serve as an example, except that it 
wraps a file port and not TCP.  I hope to put together an article 
on this (before ports are changed completely in R3 :)

HTTP is broken up thusly -- INIT breaks down the url/spec, OPEN actually 
does all the heavy lifting, sending the request and parsing the response, 
COPY takes the final download from the buffer, and CLOSE closes the 
TCP port.

the one thing I find strange comming from carl, is the fact that 
within Amiga, everything was a hook.  so you could very easily re-implement 
everything... and most of the deep internals of  REBOL are pretty 
boxed in... the stuff is still in box... but opening most of these 
secrets is like opening a can of worms...

The second and third arguments look like they are already referentially 
transparent, but I suppose you are ensuring this in case FOREACH 
ever changes its spec.

I cloned the esmtp protocol to use on port 465, and added a /secure 
to 'send so that it uses my new ssmtp protocol.

set-net [ [compkarori-:-gmail-:-com] smtp.gmail.com ]
send/secure [compkarori-:-gmail-:-com] "testing .. "


and a trace/net shows that it sends the message but then hangs waiting 
for a 250 response from the 'check-write

mucking around, i changed the system/words/insert port/sub-port "^/." 
to system/words/insert port/sub-port "^M^J.^M^J" and that gives an 
error.

as I need so often to write something and have the name same as the 
path :(

a 10MB xml file loaded from the net and loaded with xml2rebxml  took 
about 100mb of ram.  the same file loaded in firefox  took up 600mb 
of ram.  I was pretty shocked !

For example I call:
	a-function 1


and if this function has an error, it calls an error-function that 
ends with: make error! ...

This works for me:
a-function: func [] [make error! "problem"]
error? a: try [a-function]
***

If that's not quite what you have in mind, take a look at 'throw 
and 'catch -- they may be more suited.

I want to create a little spreadsheet application where users have 
access to the mathematical functions in rebol and nothing else.

Is there some way to constrain 'do to work within a specific context 
of some mathematical functions, and nothing else?

turn that into Rebol first ... and then only check the first value?

when using a parser, you can even use operator precedence rules and 
such

oh?  and third is the locals ?

I was thinking it should be after third and not before :(

third was added later than first and second IIRC

Graham - it was a joke. And I thought as maybe ppl would not understand, 
if I would post it in Announce group, I posted it rather in Humour 
group :-)

yes, the transformation to string and to block back and forth looks 
a bit messy

and  then fill your functions in.
use first system/words reduce[f1: (:f1) ... 'some-word]

another option:


math-words: [+ - * / ** = == =? < <= > >= and or xor square-root 
negate]
math-words: use math-words reduce [math-words]
foreach word math-words [set word get in system word]
etc.

yes. but keep in mind bind ignores functions and objects, so you 
must  recurse into them too.

can anyone confirm that these two statements should return the same 
value for the same blk and val?

select blk val
second find blk val


as long as val does exist in blk.

my god... REBOL is going nuts!  I have a situation where the above 
is not true...  and the select pairs are [aword an-object bword another-object]

and only the find works... if I try to select using 'aword   it always 
returns none

Ok, then fork the entire code and cut away chunks until it no longer 
displays the problem.

ehh impossible... we are talking bout an entire app here... 200-300kb 
of code. probably... and the problem occurs within an event loop.

but look at the simplicity of the two lines above... can you image 
a way in  which  the select would return none and the find would 
return the object?

A confusion between lit-words and words maybe ?

unfortunatetely, I can't print the block, cause the objects have 
nested objects, and that causes an out of memory err, If I try to 
mold it  :-(

write clipboard:// mold/all blk
and paste it (and mold/all VAL) into a new editor window.

I've discoverd another way to discover where a set occurs... within 
objects, you look at the first... and the natural order of words, 
is a telling clue...

well found it... and its a biggie  :-(   deep within glayout... well... 
its actually pretty easy for fix there, though.

It would be nice in R3 if we could remove things from an object when 
deriving with make... (we can , sort of, using the 'THIRD accessor 
on an object and editing it....)

binding would then have to replace any object-local bindings of select 
to the global context and raise an error if it doesn't exist, as 
it would if you had tried to define a new object with select undefined 
in any place.

One think that I like in JavaScript is the flexibility to freely 
add and remove members from object instances and object prototypes 
.

If I'm calling a function with many different refinements and the 
function has to work recursively and all refinements must be maintained 
throughout recursion, is there an easy way to get the function call 
with all refinements?

I'm using:


>> f: func [/a /b] [to-path remove-each p reduce ['f all [a 'a] all 
[b 'b]] [none? p]]

>> f
== f
>> f/a
== f/a
>> f/b/a
== f/a/b ; lose refinement order

I'm not sure if this helps but I sometimes make a wrapper function 
which calls the actual recursive function. The wrapper could take 
the refinements and set some flags in a context shared by all calls 
to the inner recursion function.

there is something about that you don't know the actual function 
name inside the function? if only the refinements and their order 
could be managed.

well, that only brings up a block and the refinement order is still 
off, which would be a problem if your refinements have arguments

Since I've passed in the code to call the function, it knows how 
it was called, and the order of refinements. The refinements are 
listed in spec order in the function context though.

it's a file packing function. refinements are used to determine which 
files should be skipped and which files should be processed with 
other external functions

And why does it need to be recursive ? Recursing directories ?

if you have control over the function, write a (recursive) function 
without refinements and a wrapper with refinements that calls it. 
(this is the best solution)

in immortal words of Kurt Cobain "who knows? not me."
and this probably won't help very much:
>> ?? a
a: make object! [
    b: 1
]
>> path a 'b
>> ?? a
a: make object! [
    b: end
]

just try to add url! into "from" argument and try it

Huh.  Looks like it's marking the end of the block without actually 
removing elements, so that they're still accessible if really needed.

>> A: [a b c d e]
== [a b c d e]
>> path A 3      
>> ? A
A is a block of value: [a b]
>> A/2
== b
>> A/3
>> A/4
== d
>> mold A        
== "[a b]"
>> back tail A
== [e]
>> back back tail A
== [d e]
>> back back back tail A
== []
>> back back back back tail A
== [b]


I can see myself using this...  And I think I'd actually had a problem 
a while back where it would have been helpful.

that works, but the find function doesnt seem to be working, the 
two lines it returns for matching //Database are var $new_schema 
= true; and case 'ACL': return true; ---

and remove (<> none), it's useless

and why is that evaluated first, I thought things were left to right 
?  YOu know where in the manual i can re-read this ?

it's there so we've got something to fiddle with and don't ask about 
new releases too much ;)

I wonder what is 2.7 good for, as R3 official release is at DevCon, 
and it is less than 3 months? ;-)

Pekr it's official preview and not release

I think R3 is not really delayed... I think the plan has simply changed, 
and this time Carl is finally working at delivering what he promised 
at devcon 2004.  He might have realised that its a little bit more 
work to make something so open... you can't hide the ugly things, 
so I'm pretty sure he is doing a lot of cleaning up...  and we have 
been hearing about the fact that this time, view isnt' being left 
behind...

I had function somewhere, parsing E-+value, and generating string.

and henrik and anton would be asking to not convert it  ;-) 


 I always thought to-string should cleanup the decimal! datatype...


what do you think ladislav?  its not load or mold, its specifically 
asking for a string version of a decimal... and in any case the normal 
notation of  0.000004  is still a valid decimal, so I see little 
problems...

Seems a good suggestion - and it appears this formatting issue is 
not in the RAMBO database yet.

if you clear it first and there are other references to it, you will 
have better cleanup, but then, you could as well be trashing your 
tool!  so... clear carefully.

I have just block where I append string values. I use it in recursion, 
so when go down in recursion, I copy the block into stack clear it 
and reuse the variable. When I'm going back I restore previous block 
values if any or set it to none, when it's the last one (end of recursion). 
Don't know if my explanation is clear, but will still use the 'clear' 
on my block:)

And there should not be any references into string values in my block 
(I hope:)

No. If it would be local, I would not need to clear it and store 
myself, wouldn't I?

And I must say it again... I use ATTEMPT only in cases where I do 
not expect error. Many people use it just a shortcut for error? try 
which I don't lik. As I understand attempt as higher function for 
reporting not wanted errors (for example on server). If you use it 
in cases, where like attempt [to-integer "a"] it's not good as it 
would report many errors (in some bigger context)

And Henrik, I don't think Carl will add it into Rebol as it's too 
easy to implement yourself.

And I would not use it, as I usually use something like this:
if error? try [a: to integer! a][a: 0]

And date! does not behave like integer!:

>> a: 1
== 1
>> b: "2"
== "2"
>> c: "u"
== "u"
>> to-date reduce [a b c]
** Script Error: Invalid argument: 2
** Where: to-date
** Near: to date! :value

and if I would use it, I would probably use it like:
to-date?: func[a][ not error? try [a: to date! a]]

And, perhaps worse:
    type? (1)
    == integer!

which is not what you might expect..... You have to do this to get 
type? to work:
     type? first [(1)]
    == paren!

Or do I have to read it and try to DECLOAK it?

As far as I know, an encloaked string is just a jumbled up, binary 
version of the original string: it carries no prefix signature so 
you can't tell at a glance it is an encloaked string rather than 
another bit of binary.

So, yes, as far as I knoww, you'll have to read and attempt decloaking. 
 Or, if you have control over the writing, ensure some sort of identifiable 
prefix is added)

>> do to-path reduce ['now 'none]   
** Script Error: now/none has no refinement called none
** Near: do to-path reduce ['now 'none]
>> do to-path reduce ['now none] 
== 25-Feb-2007/20:40:18+1:00
>> do probe to-path reduce ['now none]
now/none
== 25-Feb-2007/20:40:28+1:00
>> now/none
** Script Error: now has no refinement called none
** Near: now/none


I guess the difference is between 'none and none!, but I thought 
it was curious that the none! is accepted as a refinement.

I only have the problem when not using read/binary but just have 
used WRITE and READ... strange.

Yes, READ and WRITE are in text mode and translate line terminators, 
so LF can become CRLF and vice versa.

and what do you want... the string or the value?

Do you mean using it to generate file names that will be acceptable 
on any platform that REBOL runs on?
If so, I don't know, but I wouldn't take the risk.....

....I use checksum/secure and remove #{} part -- gets me a file name 
that is just letters (a-f) and digits.

henrik: the detective only updates the interpreter if it is needed; 
when it is needed, it downloades the new version as update.exe, and 
runs it, then quits. update.exe overwrites the main executable with 
itself then launchs it and quits. there is no other way on windows 
that i know of.

My apologies to Jaime and everyone, I deserve what happened. Posting 
a correction:

Quick dumb question: I have to post some data to a web site that 
requires authentication. How do I specify the username and password? 
I forget...

That worked, but now I'm getting a "Method not allowed" error from 
the server. I'm trying to enable junk email folders on an Exchange 
server and apparently the only way to do so is with a 100+ line VBScript 
that translates to a one-line REBOL script, but neither seem to work 
with this server.

I tried host and path and that builds the url properly, but now I 
get a different server error that REBOL doesn't print all of so I 
can't see it. I can't seem to trap the error so I can print the whole 
thing.

And what would that way be? :)

Yeah me too,  This is the kinda code that my brain treats as fire-and-forget. 
 Need to do it again...look it up again.  :)

According to MS's blogs, the only way to enable Exchange's server-side 
junk mail folders is to do so through Outlook Web Access on a per-user 
basis. The only way they suggest to do so for all users is to post:

cmd=options
junkemailstate=1
cmd=savejunkemailrule


to every user, logging in as a domain administrator. They provide 
a 100+ line VBScript to do this for a list of names in a file. When 
that failed, I figured that I could do the same in 1 or 2 lines of 
REBOL, and I was correct: I get the same exact failure the VBScript 
gets in 1 line of REBOL :(