AltME groups: search

actually, this is the form I used 

>> res: read dig://203.96.152.4/MX/compkarori.com
connecting to: 203.96.152.4
>> probe res/to-str
{;; REQUEST: MX(15) for compkarori.com

;; id: 53188 AA: 0.0 RD: 1.0 RA: 1.0 ANSWER(s): 2 AUTHORITY(s): 3 
ADDITIONAL(s): 3

;; ANSWER(s):
compkarori.com.^-247^-MX^-0   pop.compkarori.com.
compkarori.com.^-247^-MX^-10   pop.compkarori.com.

;; AUTHORITY(s)
compkarori.com.^-596^-NS^-ns3.webxess.net.
compkarori.com.^-596^-NS^-ns1.webxess.net.
compkarori.com.^-596^-NS^-ns2.webxess.net.

;; ADDITIONAL(s)
pop.compkarori.com.^-596^-A^-203.79.110.37
ns1.webxess.net.^-69731^-A^-216.166.83.11
ns2.webxess.net.^-69731^-A^-216.166.83.12
}

what does this mean?


   * Generate 64 bits of randomness from a good, well-seeded random 
   number generator;

ie. how large a seed do I need to get 64 bits ?

http://www.jwz.org/doc/mid.html
	

 In summary, one possible approach to generating a Message-ID would 
 be:

    * Append "<".


    * Get the current (wall-clock) time in the highest resolution to 
    which you have access 

    (most systems can give it to you in milliseconds, but seconds will 
    do);


    * Generate 64 bits of randomness from a good, well-seeded random 
    number generator;


    * Convert these two numbers to base 36 (0-9 and A-Z) and append the 
    first number, 

    a ".", the second number, and an "@". This makes the left hand side 
    of the message ID be only about 21 characters long.


    * Append the FQDN of the local host, or the host name in the user's 
    return address.

    * Append ">".

enbase doesn't accept a value of 36

a 64 bit number has 64 digits in base 2, and 12-13 digits in base 
36

anyway, since you just need some randomness, a random 12 character 
string will do

not sure if any char in base 64 could be a problem as a message id 
char, but i think MTAs do use base64 for message ids so probably 
it's fine.

a: make object! [
    b: 4
    c: 6
]

You can create a new object based on your object that adds or omits 
values, the latter being trickier.

construct head remove remove find third a to-set-word 'b

a: context [c: to-set-word 'b b: 123]

strange result, I would expect obtaining error here?
>> unset in a 'b
>> probe a

make object! [
    b: unset
    c: 6
]
>> a/b
>>

how to add object e: 7 to a ?

a: make a [ e: 7 ]

Or did you mean something more like...

foreach [word integer] [a 1 b 2 c 3] [
    print ["Word: " to string! word newline "Integer: " integer]
]

I'm getting a confusing error about using paths on a logic! object 
when trying to use the methods of a created object. I figure I'm 
missing something obvious, but I can't figure out what it is:

test-result: make object! [
    run-count: 0
    error-count: 0

    test-started: does [
        run-count: run-count + 1
    ]

    test-failed: does [
        error-count: error-count + 1
    ]

    summary: does [
        return join  run-count [ " run, " error-count " failed" ]
    ]
]


; ...


            ed: make test-result [ ]
            ed/test-started
            ed/test-failed
            assert [ ed/summary == "1 run, 1 failed" ]

; output of code:

[[wisti-:-us1-dhcp-227-65] xUnit]$ rebol xunit.r
** Script Error: Cannot use path on logic! value
** Where: test-failed-result-formatting
** Near: ed/test-started
ed/test-failed
assert [ed/summary == "1 run, 1 failed"]

Let me make the whole script available. I was hoping it was a very 
simple logic error in those lines.

A favorite subject of mine. ;-)

in 'test-case-test, you are redefining 'test-result as a function, 
not globaly so you aren't actually changing the definition but adding 
a new one to the current context.

so when you 'make 'test-result you are making a function not the 
object that you thought that you were making...

Is there a flag I can set to warn when something is being redefined?

You can keep your current set up with the redefined 'test-result 
but you will need to run 'compose on the spec block being passed 
to 'test-case-test and enclose 'test-result in a paren

You're creating a NEW definition in a NEW context.

'compose evaluates any parens in the block passed to it.  If you 
are passing 'compose a block containing blocks that contain values 
you would like composed then you need to use the /deep refinement 
of compose

obj-spec: func [
    "Returns the object spec as a single line (flat) string."
    obj [object!]
    /only "no surrounding brackets"
    /mold "Multi-line MOLDed format"
    /local res
][
    res: copy find/tail system/words/mold obj "make object! "
    if not mold [trim/lines res]
    if only [res: trim/auto next head remove back tail next res]
    res
]

remove-words: func [

    "Returns a copy of the object with the specified words removed."
    object [object!]
    words  [word! block!] "The word, or words, to remove"
    /local spec
][
    spec: load obj-spec object
    foreach word compose [(words)] [
        remove/part find spec to set-word! word 2
    ]
    make object! spec
]

block: [ "a:" "string1" "b:" "string2"    ]

Fremowe: func [x y][return remove remove find x  y ]

Fremowe block "a:"

not remove a ?

I want to get in a Rebol script what I print on a standard windows 
printer.

I use Redmon (part of ghostview) to redirect what the printer get 
to my rebol script
RedMon : http://www.cs.wisc.edu/~ghost/redmon/index.htm
I'm under Windows.

How can I get data from the standard input ?

I have tried "copy system/standard/input" and also "input", but get 
nothing !!
plis help ;-)

or use a temporary file?

Ah, this looks like a cgi-stype call. i would try --cgi then. it 
should do this "call and get data from stdin".

conn: first port
wait conn ; such things in /awake may be a problem.

(Script is in %inp.r)
I have tried this command line (I'm on Win2k, Core 2.5.6) :

C:\> echo this is a test | rebol.exe inp.r
C:\> rebol.exe inp.r < test.txt
C:\> type test.txt | rebol.exe inp.r

i understand from the webpage that redmon launches the program and 
send it input, just as a webserver does with cgi. then try it from 
the real program, not from the console. i don't know about xp, but 
in win9x console-pipes are pretty broken.

echo this is a test | rebol.exe -cw test.r

you get "this is a test" in test.txt

if you PROBE it, to see it on the command window you need a | more 
or something like that.

I used to know this, but for which datatypes is a hash! indexed? 
Just strings, or words too?

It used to be just strings that were hashed. Other data types were 
just sitting there, and had to be found with a linear search like 
with other blocks.

Is there a  Perl compatible regular expression parser written in 
REBOL

note to volker:   what did I say  !!!!  ;-)


I was just telling volker privately, a few hours ago, how having 
regexp within REBOL would help those who migrate to it  ;-)

Joe, not that I know of.   Try to learn the 'Parse word.  There are 
a lot of parse experts hanging on this site... (not that I am one 
of them..)

is there a way to set the datestamp when writing a file?

Larry's list of RegEx shortcomings http://www.perl.com/pub/a/2002/06/04/apo5.html?page=2

it's a start ...

Does one say what is the syntax to specify parity on a serial port 
?

I saw "parity: none" on docs, but what does it mean, even or odd 
? And so what is the other syntax (odd or even) "parity: ????"

WRT RegEx's -- I did a simple wildcard matcher (emulates VB's Like 
operator), and looked at hooking up PCRE, but it had a funky interface 
to it and I didn't get it working in the limited time I spent on 
it.

Is it possible to set the modification date on a directory?  I keep 
getting errors whereas it works with files in win32.

>> set-modes %xml-parse.r [ modification-date: 1-Jan-2005 ]
>> set-modes %www/ [ modification-date: 1-Jan-2005 ]
** Access Error: Cannot open /D/rebol/rebXR/www/
** Near: set-modes %www/ [modification-date: 1-Jan-2005]
>> set-modes %www [ modification-date: 1-Jan-2005 ]
** Access Error: Cannot open /D/rebol/rebXR/www
** Near: set-modes %www [modification-date: 1-Jan-2005]
>>

Graham I did something similiar which allows me to write the date 
to another target just as it appears in the original - kinda like 
a backup.  However, I need to modify it so that the target dates 
are stamped as they are originally.  Currently it changes the stamp.

Hi, is it possible to restore the date of a file which was read via 
ftp?

>> foo: [a [does this]]
== [a [does this]]
>> bar: [b [does that]]
== [b [does that]]
>> rejoin [foo bar]
== [a [does this] [b [does that]]]

Is there an equivelant function that would give me:
== [a [does this] b [does that]]

>> A: []
== []
>> insert A bar
== []
>> insert A foo
== [b [does that]]
>> ? A
A is a block of value: [a [does this] b [does that]]

Yep, I can fudge it that way, but I was looking for a single function, 
like 'rejoin or 'reduce

Or, if you don't mind changing 'bar:
>> insert bar foo
== [b [does that]]
>> ? bar
BAR is a block of value: [a [does this] b [does that]]

How about...

>> foo: [a [does this]]
== [a [does this]]
>> bar: [b [does that]]
== [b [does that]]
>> append copy foo bar
== [a [does this] b [does that]]
>> compose [(foo) (bar)]
== [a [does this] b [does that]]

I have a feeling that if 'reduce doesn't do it, then nothing will...

>> one-word: func [block /local val][val: copy [] foreach word block 
[append val get word]val]
>> one-word [foo bar]
== [a [does this] b [does that]]

Although, I don't understand why.  It is a one-line solution so I 
don't see why it should make a difference if it is built in or not...

There are quite a few words in REBOL that, IMHO, should have automatic 
block handling.

Hmm, and I was thinking of a refinement to 'reduce.  'Get makes more 
sense...

... Is there a quick trick to compare 2 pairs??  Seems 'lesser? or 
greater? dont work on pairs...

Chris:

>> foo: [a [does this]]
== [a [does this]]
>>  bar: [b [does that]]
== [b [does that]]
>> foobar: compose [(foo) (bar)]
== [a [does this] b [does that]]

Yes it is - but he wants [foo bar] to become  [a [does this] b [does 
that]] without the () or anything else inside..

Yep, I'm resigned to that.  (and I'll word my queries a little better 
next time :^)

I've just built a direct printing library for R2, Windows only. It's 
a wrapper on Win32 Print API, so it supports all printers. It support 
a subset of Draw dialect as input. I was needing it to print reports 
for the project I'm currently working on. It still needs some additionnal 
work to be released publicly (like adding a port scheme layer for 
more intuitive usage).

My library is just a thin layer that maps Draw dialect to Win32 Printing 
API.

I also need to add extend Draw dialect with a new command: text-box. 
It's an improved version of 'text that allow you to define a bouding 
box, align the text horizontally and vertically and auto-wrap text.

For R3 you might look into the rich text support. I am less familiar 
with R2's Draw (and that's saying a lot).

I had a quick look at XPS API, but it looked more complicated and 
required more work than GDI API. There was also the compatibility 
issue, I needed a solution that would work with any printer. I'll 
gave a deeper look at XPS latter.

It might make more sense for R3, mostly as a thought experiment to 
help us decide on the semantics of the REBOL printing model.

From what I can tell, they did a lot of interesting research when 
they came up with XPS - food for thought.

it would have been a lot more fun if they just used postscript :-)

Here's a nice introduction to XPS : http://msdn.microsoft.com/en-us/library/ms742418.aspx

No, it would have been horrible. There is a reason that even Adobe 
has moved away from Postscript - its model has major problems.

Read up on the research on PDF sometime before you start promoting 
Postscript. It is even a good idea to use PDF instead if you are 
outputting through Ghostscript - it can handle it.

That's why Apple based its Quartz model on PDF, when they already 
had a Postscript model from NeXT.

XPS is like a cleaned-up, extended PDF, with an XML representation 
if you're into that. The models are similar.

It would probably be easier to get AGG to output stuff in a form 
GDI would like though, with more overhead from pushing around all 
of that bitmap data of course.

It might have had problems, but it would have been a much better 
starting point, had Microsoft embraced postscript from the start. 
There would have been a common starting point and a much larger incentive 
for building hardware postscript printers at the time. If that had 
been done, printer drivers would not be necessary under any platform 
today, or they would be limited to being postscript rasterizers.

At the start, postscript printers cost thousands of dollars but dot 
matrix printers cost a couple hundred. If MS had gone with Postscript, 
printing would have been stillborn outside of large companies.

That was considered a hard problem on 8086 computers. Remember how 
far back the "beginning" was...

Remember that the procedural model of Postscript meant that a Postscript 
printer was a computer, and definitely a more powerful and more expensive 
computer than most people could afford. Even faking Postscript support 
required a computer of at least the same scale.

Well, I still think postscript should have become more widespread 
than it ended up being. And you can't change my opinion on that. 
:-) I crave standardization. OK, so if postscript was too hardware 
hungry, then a lighter version could have helped, which is why I 
wonder why PDF came so late.

Well, if you are using the OS's facilities for printing you are using 
the API version of the semantics, not necessarily the source. What 
really matters is the semantics - the source is just a generated 
representation.

It wouldn't be the wrapping of the Windows API that would help Linux 
users, it would be his initial work on making a Draw-like printing 
dialect. Defining the dialect is a large part of the process of supporting 
printing in REBOL. There will be non-Windows-specific parts of Doc's 
implementation that can be adapted to a general printing model for 
REBOL, one that can have multiple implementations with different 
backends. For that matter, there would need to be at least 3 backends: 
GDI (for Windows), Postscript (for Ghostscript) and PDF (for Mac 
Quartz), with a possible XPS backend as a minor variation on the 
PDF one.

No docs for now, look at the sample %test-page.r script and at the 
scheme implementation. Input dialect is a subset of Draw dialect.

Draw dialect maps very well with Windows drawing API (GDI). It's, 
in most cases, a one to one mapping.

But Draw dialect is really too level for a daily use. A higher level 
dialect with relative positionning and higher level constructs (e.g. 
tables support), like VID or HTML is needed.

Thanks but this isn't really such a great piece of code (Windows 
API is doing the real job), even if it fills a gap in REBOL (at least 
for Windows). Btw, in my company, we're using Gab's pdf-maker for 
years now to generate and print all our documents. I made this library 
only because I needed a direct printing solution for a customer and 
I must admit it was a fun work to do.

mine focuses more on the UI side, offering various methods of printing 
postscript. there is also a printer queue system as well as a printer 
server.

who wants to clone Crystal Reports ?
  

We would need a /bloat refinement to do that. :-)

For information, I've successfully tested direct printing in Linux 
and OS X using PostScript format documents and CUPS as backend. I'm 
currently trying to implement a Draw dialect compiler targeting PS. 
Unix and OS X support wasn't needed for my project, but I couldn't 
resist to give it a try ;-).

I know that Geomol has built a PS lib but, unfortunately, it doesn't 
take Draw dialect as input.

Sure, if he has access to that part of the source code. Henrik, did 
you made a RAMBO ticket for that issue ?

the idea was to use size-text to produce the needed position, but 
the result is not usable, because I don't think DRAW uses the concept 
of a bounding box for text.

Update on the work-in-progress : http://softinnov.org/tmp/test-page.zip


Both files are printed from the same Draw dialect source, using my 
printer:// scheme. The PDF file is printed through Bullzip PDF Virtual 
printer. The PS file is directly generated by the printer scheme 
(for UNIX/Cups direct printing).


Most of the PostScript support is done (see %test-page.ps), but there's 
still a lot of details to enhance/fix/add: 

o Add center/right alignement support

o Add underline style for fonts

o Fine-tune positionning and bold level.

o Fix minor differences with the GDI version.

don't bother with underline style!  It's a relic of type setting.

There are a few good justifications schemes available for postscript

I found a justification routine (doing also alignement). I need to 
study it to see if it fit my needs : align and line-wrap at the same 
time.

there's also a very nice bar code generator for postscript.

Btw, I need to make a bar code generator for my current project, 
so that's something I'll work on soon.