http://www.ardice.com/Recreation/Outdoors/Fishing/
Hello, I found this group at this website.


Moderator's note: The correct link from this web directory is
http://www.ardice.com/Computers/Programming/Languages/REBOL/

The best source I have found for learning programming and software is at http://
www.vtc.com/usa.php Too bad they don't have a course on Rebol. In fact, someone
could do several (Core, View, IOS, AltMe, Command, etc). I have watched many
hundreds of their Quicktime training movies.

On 11-Mar-04, Pete Wason wrote:
> Carl Read wrote:

>> On 10-Mar-04, Pete Wason wrote:
>>
>>> I'm running Core 2.5 on WinXP Home, using Netscape 7.1, and am
>>> trying to use Corel's Webserver (ships with Paradox 8) as a
>>> server.
>>
>>
>>> The server is running, and has a mime type set for "r
>>> application/x-rebol", and its base document path is
>>> D:\Dev\Rebol\rtog.
>>
>>
>>> My cgi script D:\Dev\Rebol\rtog\cgi-bin\first-cgi.r:
>>
>>
>>> #!d:/dev/rebol/rebol.exe --cgi --script REBOL []
>>> print { Content-Type: text/html^/^/Welcome to my first REBOL CGI
>>> script. }
>>
>>
>>> Netscape *also* has a helper application type of
>>
>>
>>>    application/x-rebol
>>>    Rebol Script
>>>    .r
>>>    Open these files with rebol.exe
>>
>>
>>
>>> When I try to access http://localhost/cgi-bin/first-cgi.r, a
>>> netscape "downloading" dialog pops up very quickly, disappears,
>>> rebol opens, probably prints the message above, and closes.
>>
>>
>>
>>> Something simple, I'm sure. If I test this locally (in rebol) it
>>> works fine.


>> I doubt this is the problem, but given there's no HTML being
>> printed, the first line should perhaps be...

>> print { Content-Type: text/plain^/^/Welcome...

> Eventually, there will be HTML; in fact I added some after I sent
> the previous message, and it works as expected in Rebol, but not in
> the browser.


>> Not sure, but maybe the space before Content-Type has some effect -
>> you never know. :)  Also, do file permissions need to be set for
>> Corel's Webserver?  And if so, have you set the script's executable
>> bits?

> Well, in Windows, there aren't a whole lot of permissions to set,
> unless I am missing something. There certainly aren't any mentioned
> in the COrel webserver stuff. Similarly, I can access shares on this
> same WinXP laptop over my network without even entering a username
> or password. Disturbing.

(:

You could try playing around with different CGI lines in your script.
Such as...

#!d:/dev/rebol/rebol.exe -cs

But otherwise, I'd suggest asking on the main REBOL mailing list.
There's not many on this one I'm afraid...

--
Carl Read

Carl Read wrote:

> On 10-Mar-04, Pete Wason wrote:
>
>>I'm running Core 2.5 on WinXP Home, using Netscape 7.1, and am
>>trying to use Corel's Webserver (ships with Paradox 8) as a server.
>
>
>>The server is running, and has a mime type set for "r
>>application/x-rebol", and its base document path is
>>D:\Dev\Rebol\rtog.
>
>
>>My cgi script D:\Dev\Rebol\rtog\cgi-bin\first-cgi.r:
>
>
>>#!d:/dev/rebol/rebol.exe --cgi --script REBOL []
>>print { Content-Type: text/html^/^/Welcome to my first REBOL CGI
>>script. }
>
>
>>Netscape *also* has a helper application type of
>
>
>>    application/x-rebol
>>    Rebol Script
>>    .r
>>    Open these files with rebol.exe
>
>
>
>>When I try to access http://localhost/cgi-bin/first-cgi.r, a
>>netscape "downloading" dialog pops up very quickly, disappears,
>>rebol opens, probably prints the message above, and closes.
>
>
>
>>Something simple, I'm sure. If I test this locally (in rebol) it
>>works fine.
>
>
> I doubt this is the problem, but given there's no HTML being printed,
> the first line should perhaps be...
>
> print { Content-Type: text/plain^/^/Welcome...

Eventually, there will be HTML; in fact I added some after I sent the previous
message, and it works as expected in Rebol, but not in the browser.

>
> Not sure, but maybe the space before Content-Type has some effect -
> you never know. :)  Also, do file permissions need to be set for
> Corel's Webserver?  And if so, have you set the script's executable
> bits?

Well, in Windows, there aren't a whole lot of permissions to set, unless I am
missing something. There certainly aren't any mentioned in the COrel webserver
stuff. Similarly, I can access shares on this same WinXP laptop over my network
without even entering a username or password. Disturbing.

Pete

On 10-Mar-04, Pete Wason wrote:
> I'm running Core 2.5 on WinXP Home, using Netscape 7.1, and am
> trying to use Corel's Webserver (ships with Paradox 8) as a server.

> The server is running, and has a mime type set for "r
> application/x-rebol", and its base document path is
> D:\Dev\Rebol\rtog.

> My cgi script D:\Dev\Rebol\rtog\cgi-bin\first-cgi.r:

> #!d:/dev/rebol/rebol.exe --cgi --script REBOL []
> print { Content-Type: text/html^/^/Welcome to my first REBOL CGI
> script. }

> Netscape *also* has a helper application type of

>     application/x-rebol
>     Rebol Script
>     .r
>     Open these files with rebol.exe


> When I try to access http://localhost/cgi-bin/first-cgi.r, a
> netscape "downloading" dialog pops up very quickly, disappears,
> rebol opens, probably prints the message above, and closes.


> Something simple, I'm sure. If I test this locally (in rebol) it
> works fine.

I doubt this is the problem, but given there's no HTML being printed,
the first line should perhaps be...

print { Content-Type: text/plain^/^/Welcome...

Not sure, but maybe the space before Content-Type has some effect -
you never know. :)  Also, do file permissions need to be set for
Corel's Webserver?  And if so, have you set the script's executable
bits?

All just guesses - I'm no Net guru.

--
Carl Read

I'm running Core 2.5 on WinXP Home, using Netscape 7.1, and am trying to use
Corel's Webserver (ships with Paradox 8) as a server.

The server is running, and has a mime type set for "r application/x-rebol", and
its base document path is D:\Dev\Rebol\rtog.

My cgi script D:\Dev\Rebol\rtog\cgi-bin\first-cgi.r:

#!d:/dev/rebol/rebol.exe --cgi --script
REBOL []
print { Content-Type: text/html^/^/Welcome to my first REBOL CGI script. }

Netscape *also* has a helper application type of

	 application/x-rebol
	 Rebol Script
	 .r
	 Open these files with rebol.exe


When I try to access http://localhost/cgi-bin/first-cgi.r, a netscape
"downloading" dialog pops up very quickly, disappears, rebol opens, probably
prints the message above, and closes.


Something simple, I'm sure. If I test this locally (in rebol) it works fine.


Pete

On 07-Dec-03, Jack Seay wrote:

> I have recently started another Rebol mailing list. It is at
> http://groups.yahoo.com/group/RebolParse. Description: Rebol parsing
> problems and solutions. If you have a project that involves parsing
> text using the Rebol language, ask for help here. No project too big
> or small. Commercial or non-profit projects welcome. Dumb questions
> also welcomed.

Just joined.  I'd long ago forgot my Yahoo ID, never mind my password.
  My advice is never to forget them - it's a journey into hell and
back to sort them out again... :)

--
Carl Read

I have recently started another Rebol mailing list. It is at
http://groups.yahoo.com/group/RebolParse. Description: Rebol parsing
problems and solutions. If you have a project that involves parsing
text using the Rebol language, ask for help here. No project too big
or small. Commercial or non-profit projects welcome. Dumb questions
also welcomed.

Yahoo! India Mobile: Ringtones, Wallpapers, Picture Messages and more. Download now.

Yahoo! India Mobile: Ringtones, Wallpapers, Picture Messages and more. Download now.

Hi!

In one of my reb index.r files, I'm trying to use a background graphic:

	 backdrop tile %backdrop.png

It doesn't seem to have any effect. Are PNGs not supported for background
graphics?


Pete

Hello,

I use a Debian distro and I´ve put REBOL in my /home dir
as a normal user, since I am the only user of my PC

You may install REBOL in /usr/bin and chmod it to 755
so others can user it from there

Carlos


> Hiall!
>
> *Where* am I supposed to put it? The installation instr
uctions ("none
> needed") are very vague, and any reference to "linux re
bol installation"
>   I found was either in german or french...
>
> Looking at my /var/www dir, I see that everything seems
  to be owned by
> root --
  If I make a rebol dir in there, should it and everything
  in it
> also be owned by root? I haven't changed any permission
s since
> installing RH8.0, so if this seems wierd, blame RH.
>
> confusticated,
>
> Pete
>
>
> ------------------------ Yahoo! Groups Sponsor --------
>
> Community email addresses:
>   Post message: Rebol_New@onelist.com
>   Subscribe:    Rebol_New-subscribe@onelist.com
>   Unsubscribe:  Rebol_New-unsubscribe@onelist.com
>   List owner:   Rebol_New-owner@onelist.com
>
> Shortcut URL to this page:
>   http://www.onelist.com/community/Rebol_New
>
> Your use of Yahoo! Groups is subject to http://docs.yah
oo.com/info/terms/
>
>
>


__________________________________________________________________________
Acabe com aquelas janelinhas que pulam na sua tela.
AntiPop-up UOL - É grátis!
http://antipopup.uol.com.br/

Hiall!

*Where* am I supposed to put it? The installation instructions ("none
needed") are very vague, and any reference to "linux rebol installation"
   I found was either in german or french...

Looking at my /var/www dir, I see that everything seems to be owned by
root -- If I make a rebol dir in there, should it and everything in it
also be owned by root? I haven't changed any permissions since
installing RH8.0, so if this seems wierd, blame RH.

confusticated,

Pete

On 15-Aug-03, johnfdutcher wrote:
> Hi again,

> No, the server I'm playing with is %webserv.r by Cal Dixon.

Ah - this one...

http://www.reboltech.com/library/html/webserv.html

Looks much more promising. :)

Blugh - just tried it and can't get it to even display static
web-pages.  Might have a proper look at it over the weekend...

> I'm
> looking at it a bit and I think that unlike Apache and many other
> servers, it doesn't look at line one of the script to decide where
> to find the interpreter....it assumes it to be where (it is)..in the
> same directory as the server script.

It would seem so, going by the docs.

> I'll try the c:/rebol/view/ and see what happens.

> It's interesting, the other simple server I'm playing with,
> TheServer, from Fastlink, seems more sophisticated, and certainly
> much better supported.... but doesn't want to find the html
> pages in its deirectory. I could always revert back to Apache
> which has worked well....but there's a lot more to it than I need
> since I just want to serve up web pages and run Rebol scripts
> that may also connect to MySQL databases.

One advantage of Apache (which I haven't tried) would be that it'd
most likely match what you're likely to be using on remote servers in
the future.

--
Carl Read

Hi again,

No, the server I'm playing with is %webserv.r by Cal Dixon.
I'm looking at it a bit and I think that unlike Apache and many other
servers, it doesn't look at line one of the script to decide where
to find the interpreter....it assumes it to be where (it is)..in the
same directory as the server script.

I'll try the c:/rebol/view/ and see what happens.

It's interesting, the other simple server I'm playing with,
TheServer, from Fastlink, seems more sophisticated, and certainly
much better supported.... but doesn't want to find the html
pages in its deirectory. I could always revert back to Apache
which has worked well....but there's a lot more to it than I need
since I just want to serve up web pages and run Rebol scripts
that may also connect to MySQL databases.

John D.

--- In Rebol_New@yahoogroups.com, Carl Read <carl@c...> wrote:
> On 14-Aug-03, johnfdutcher wrote:
>
> > Hi and thanks again,
> > I have little doubt that the CGI line is probably the issue....but
> > it seems as though there are only so many variations of that line
I
> > can put together trying to find one that might work. My rebol.exe
is
> > on the hard drive (C) in a folder under Rebol named View....hence
> > the path would seem to have to be: #!c:/rebol/view -cs OR
> >                               #!/rebol/view -cs     OR
> >                               #!c:\rebol\view -cs   OR
> >                               #!\rebol\view -cs
>
> > but none of those seem to cut it.
>
> Have you tried...
>
>     #!/c/rebol/view -cs
>
> ?  As the server's written in REBOL I'd assume it would use REBOL's
> file-path format.  Can you find out where in webserver.r it accesses
> the CGI line and add probes to it to see what it does with it?
>
> Hmmm - is this the webserver.r you're using? ...
>
> http://www.reboltech.com/library/html/webserver.html
>
> As it doesn't look very full-featured and I don't see how a script
> could be run from it.  There'd surely need to be a 'do in there
> somewhere, right?  You may need to enhance it to run scripts if it
is
> that one.
>
> > On the home computer (Win 98), I get the actual source of the
script
> > delivered as noted....at work (Win 95), I get a dialog opening
> > suggesting that I have chosen to download a file from the site
> > and asking if I want to open it in place or download it to another
> > location ?
> > It's interesting stuff  :-) :-)
>
> I'd think the difference would just be due to different browser
> settings (or browsers) and not because of the OS.
>
> --
> Carl Read

On 14-Aug-03, johnfdutcher wrote:

> Hi and thanks again,
> I have little doubt that the CGI line is probably the issue....but
> it seems as though there are only so many variations of that line I
> can put together trying to find one that might work. My rebol.exe is
> on the hard drive (C) in a folder under Rebol named View....hence
> the path would seem to have to be: #!c:/rebol/view -cs OR
>                               #!/rebol/view -cs     OR
>                               #!c:\rebol\view -cs   OR
>                               #!\rebol\view -cs

> but none of those seem to cut it.

Have you tried...

     #!/c/rebol/view -cs

?  As the server's written in REBOL I'd assume it would use REBOL's
file-path format.  Can you find out where in webserver.r it accesses
the CGI line and add probes to it to see what it does with it?

Hmmm - is this the webserver.r you're using? ...

http://www.reboltech.com/library/html/webserver.html

As it doesn't look very full-featured and I don't see how a script
could be run from it.  There'd surely need to be a 'do in there
somewhere, right?  You may need to enhance it to run scripts if it is
that one.

> On the home computer (Win 98), I get the actual source of the script
> delivered as noted....at work (Win 95), I get a dialog opening
> suggesting that I have chosen to download a file from the site
> and asking if I want to open it in place or download it to another
> location ?
> It's interesting stuff  :-) :-)

I'd think the difference would just be due to different browser
settings (or browsers) and not because of the OS.

--
Carl Read

Hi and thanks again,
I have little doubt that the CGI line is probably the issue....but
it seems as though there are only so many variations of that line I
can put together trying to find one that might work. My rebol.exe is
on the hard drive (C) in a folder under Rebol named View....hence the
path would seem to have to be: #!c:/rebol/view -cs   OR
                                #!/rebol/view -cs     OR
                                #!c:\rebol\view -cs   OR
                                #!\rebol\view -cs

but none of those seem to cut it.

On the home computer (Win 98), I get the actual source of the script
delivered as noted....at work (Win 95), I get a dialog opening
suggesting that I have chosen to download a file from the site
and asking if I want to open it in place or download it to another
location ?
It's interesting stuff  :-) :-)
John D.
--- In Rebol_New@yahoogroups.com, Carl Read <carl@c...> wrote:
> Hi John,
>
> On 13-Aug-03, johnfdutcher wrote:
>
> > I have installed the Carl Sassenrath scripts from the Rebol script
> > library called webcomment.r and cgicomment.r. As directed I ran
the
> > webcomment.r script which produced a nice html page with a form in
> > it as expected. I retrieved the form, entered a comment, and
> > 'submitted' the page; what was returned was a 'text' display of
the
> > cgicomment.r script....rather than an updated html page showing
the
> > added comment, (the result of 'executing' that script,
presumably).
>
> 1) Have you set the permissions on the script to executable?  Not
sure
> if it'd be required for webserver.r, but it definately would be on a
> remote server.
>
> 2) Have you got the CGI header path correct for your server?  I note
> cgicomment.r has...
>
>     #!rebol -cs
>
> For webserver.r it may need a different path for your settup.  ie,
> something like...
>
>     #!/path/to/rebol -cs
>
> A typical "hello world" script on a server might look something like
> this...
>
> #!/usr/bin/rebol -cs
> REBOL []
> print "Content-Type: text/plain^/"
> print "Hello World!"
>
> Now if you saved that on the server as a file called (say)
> "hello-world.r", set the permissions correctly and entered its URL
in
> your browser, if the CGI path at the top of the script is correct it
> should display a page saying "Hello World".  If it does't, then it's
> probably the CGI path that's wrong.
>
> > I am using the %webserver.r server from Cal Dixon...it seems to
> > serve up html nicely....just wondered if anyone has tried this cgi
> > experiment with the Sassenrath scripts and had better
> > results...which suggest whether a 'script' error, or that the
> > %webserv.r server isn't quite handling the cgi just right.
>
> I've not tried those two scripts, (though I've just had a quick look
> at them), but as it happens I've been playing with something similar
> for the last couple of days, though using POST, not GET.  This is
the
> script I've come up with.: (Call it submissions.r so the HTML page
> matches it) ...
>
>
> #!/usr/bin/rebol -cs
>
> REBOL []
>
> print "Content-Type: text/plain^/"
>
> form-data: make string! 10002
> either system/options/cgi/request-method = "POST" [
>     read-io system/ports/input form-data 10000
>     parse form-data [some [
>         st: "+" (change st " ") |
>         st: "=" (en: change/part st ": {" next st) :en |
>         st: "&" (en: change/part st "}^/" next st) :en |
>         st: "%0D%0A" (en: change/part st "^/" skip st 6) :en |
>         st: "%" (
>             en: change/part st dehex copy/part st 3 skip st 3
>         ) :en |
>         skip
>     ]]
>     append form-data "}"
> ][
>     insert form-data {result: "No Post"}
> ]
> form-data: make object! load form-data
>
> print mold form-data
> print mold system/options/cgi
>
>
> Sorry there's no comments, but what it does is look to see if POST
was
> used and if so, it parses the POST data so it's suitable for
> converting into an object (which is what 'form-data ends up as) and
> then prints it out, as well as the system/options/cgi object.  (To
> turn the output into HTML you'd need to change the "plain" in the
> content-type line to "html" and of course print strings of HTML tags
> instead of plain text.)
>
> The parse routine above  hasn't had extensive testing yet but I
think
> it should be usable for parsing any POST data that's returned.
>
> And the following is the HTML page I've been using to test the
script.
>  It should be in the same directory as the script...
>
>
> <html>
>
>  <head>
>   <title>Test Page</title>
>  </head>
>
>  <body>
>   <h2>Test</h2>
>   <h4>Submission Page:</h4>
>
>   <form action="submissions.r" method="POST">
>    <input type="TEXT" name="input-text" size=100">
>    <textarea name="text-area" rows="10" cols="40">
>    </textarea>
>    <input type="SUBMIT" value="Submit">
>   </form>
>
>  </body>
>
> </html>
>
>
> Hope there's some help amongst that.  The key with CGI is to get the
> simpliest of scripts working first.  Once you've done that, expand
on
> them.
>
> --
> Carl Read

Hi John,

On 13-Aug-03, johnfdutcher wrote:

> I have installed the Carl Sassenrath scripts from the Rebol script
> library called webcomment.r and cgicomment.r. As directed I ran the
> webcomment.r script which produced a nice html page with a form in
> it as expected. I retrieved the form, entered a comment, and
> 'submitted' the page; what was returned was a 'text' display of the
> cgicomment.r script....rather than an updated html page showing the
> added comment, (the result of 'executing' that script, presumably).

1) Have you set the permissions on the script to executable?  Not sure
if it'd be required for webserver.r, but it definately would be on a
remote server.

2) Have you got the CGI header path correct for your server?  I note
cgicomment.r has...

     #!rebol -cs

For webserver.r it may need a different path for your settup.  ie,
something like...

     #!/path/to/rebol -cs

A typical "hello world" script on a server might look something like
this...

#!/usr/bin/rebol -cs
REBOL []
print "Content-Type: text/plain^/"
print "Hello World!"

Now if you saved that on the server as a file called (say)
"hello-world.r", set the permissions correctly and entered its URL in
your browser, if the CGI path at the top of the script is correct it
should display a page saying "Hello World".  If it does't, then it's
probably the CGI path that's wrong.

> I am using the %webserver.r server from Cal Dixon...it seems to
> serve up html nicely....just wondered if anyone has tried this cgi
> experiment with the Sassenrath scripts and had better
> results...which suggest whether a 'script' error, or that the
> %webserv.r server isn't quite handling the cgi just right.

I've not tried those two scripts, (though I've just had a quick look
at them), but as it happens I've been playing with something similar
for the last couple of days, though using POST, not GET.  This is the
script I've come up with.: (Call it submissions.r so the HTML page
matches it) ...


#!/usr/bin/rebol -cs

REBOL []

print "Content-Type: text/plain^/"

form-data: make string! 10002
either system/options/cgi/request-method = "POST" [
     read-io system/ports/input form-data 10000
     parse form-data [some [
         st: "+" (change st " ") |
         st: "=" (en: change/part st ": {" next st) :en |
         st: "&" (en: change/part st "}^/" next st) :en |
         st: "%0D%0A" (en: change/part st "^/" skip st 6) :en |
         st: "%" (
             en: change/part st dehex copy/part st 3 skip st 3
         ) :en |
         skip
     ]]
     append form-data "}"
][
     insert form-data {result: "No Post"}
]
form-data: make object! load form-data

print mold form-data
print mold system/options/cgi


Sorry there's no comments, but what it does is look to see if POST was
used and if so, it parses the POST data so it's suitable for
converting into an object (which is what 'form-data ends up as) and
then prints it out, as well as the system/options/cgi object.  (To
turn the output into HTML you'd need to change the "plain" in the
content-type line to "html" and of course print strings of HTML tags
instead of plain text.)

The parse routine above  hasn't had extensive testing yet but I think
it should be usable for parsing any POST data that's returned.

And the following is the HTML page I've been using to test the script.
  It should be in the same directory as the script...


<html>

  <head>
   <title>Test Page</title>
  </head>

  <body>
   <h2>Test</h2>
   <h4>Submission Page:</h4>

   <form action="submissions.r" method="POST">
    <input type="TEXT" name="input-text" size=100">
    <textarea name="text-area" rows="10" cols="40">
    </textarea>
    <input type="SUBMIT" value="Submit">
   </form>

  </body>

</html>


Hope there's some help amongst that.  The key with CGI is to get the
simpliest of scripts working first.  Once you've done that, expand on
them.

--
Carl Read

I have installed the Carl Sassenrath scripts from the Rebol script
library called webcomment.r and cgicomment.r. As directed I ran the
webcomment.r script which produced a nice html page with a form in it
as expected. I retrieved the form, entered a comment, and 'submitted'
the page; what was returned was a 'text' display of the cgicomment.r
script....rather than an updated html page showing the added comment,
(the result of 'executing' that script, presumably).

I am using the %webserver.r server from Cal Dixon...it seems to serve
up html nicely....just wondered if anyone has tried this cgi
experiment with the Sassenrath scripts and had better results...which
suggest whether a 'script' error, or that the %webserv.r server isn't
quite handling the cgi just right.

John D.

Thanks....again....Carl. I re-installed the %webserv.r server in the
manner originally suggested. It is working now. Your suggestion about
'localhost' led to a solution (yes, 127.0.0.1 and localhost are
interchangeable addresses)....but I had not been adding the actual
suffix: /index.html to my URL entries. Usually that is not necessary
as the 'index.html' is ASSUMED to be the desired page if none is
entered. For whatever reason, the %webserv.r server seems to need a
specific entry attached (or is it the browser, who knows ?). I can
use either
'http://localhost:80/index.html' or 'http://127.0.0.1/index.html' and
get the page.
Thanks again, John D.

--- In Rebol_New@yahoogroups.com, Carl Read <carl@c...> wrote:
> On 10-Aug-03, johnfdutcher wrote:
>
> > Hello,I installed the server in my Rebol/View/local ... folder.
> > Since I cannot relate to the concept of 'changing directories and
> > typing do %webserv.r' as instructed in the scrpt; in a Rebol
> > context, (how does one change directories once they have started
the
> > Rebol/View app) ?
>
> Hi John,
>
> 'what-dir will return the current directory and 'change-dir can be
> used to change it.
>
> > I added webserv.r to my 'local' folder, made an
> > entry in the 'index' file for it...and it seems to start up very
> > nicely in response to a click on its icon in the 'local' desktop
> > display. I put an index.html page in a new folder named 'www' in
the
> > same 'view/local' directory as webserv.r is in as instructed. When
> > keying the 'http://127.0.0.1:80' into either Netscape or
I.E....the
> > browser replies 'page not found' and the log indicates a '404'
> > message etc. Any suggestions from anyone who might run this
server?
> > All 'default' settings are currently
> > in effect. I run on Win 98.
>
> I've not used webserv.r so really can't help with it.  I wrote a
> simple server myself a while back and it used http://localhost:8001/
> as the address.  Is 127.0.0.1 the same as localhost?  I'm a
> rank-amateur when it comes to networking stuff...
>
> --
> Carl Read

On 10-Aug-03, johnfdutcher wrote:

> Hello,I installed the server in my Rebol/View/local ... folder.
> Since I cannot relate to the concept of 'changing directories and
> typing do %webserv.r' as instructed in the scrpt; in a Rebol
> context, (how does one change directories once they have started the
> Rebol/View app) ?

Hi John,

'what-dir will return the current directory and 'change-dir can be
used to change it.

> I added webserv.r to my 'local' folder, made an
> entry in the 'index' file for it...and it seems to start up very
> nicely in response to a click on its icon in the 'local' desktop
> display. I put an index.html page in a new folder named 'www' in the
> same 'view/local' directory as webserv.r is in as instructed. When
> keying the 'http://127.0.0.1:80' into either Netscape or I.E....the
> browser replies 'page not found' and the log indicates a '404'
> message etc. Any suggestions from anyone who might run this server?
> All 'default' settings are currently
> in effect. I run on Win 98.

I've not used webserv.r so really can't help with it.  I wrote a
simple server myself a while back and it used http://localhost:8001/
as the address.  Is 127.0.0.1 the same as localhost?  I'm a
rank-amateur when it comes to networking stuff...

--
Carl Read

Hello,I installed the server in my Rebol/View/local ...
folder. Since I cannot relate to the concept of
'changing directories and typing do %webserv.r' as instructed in the
scrpt;  in a Rebol context, (how does one change directories once
they have started the Rebol/View app) ? I added
webserv.r to my 'local' folder, made an entry in the
'index' file for it...and it seems to start up very
nicely in response to a click on its icon in the
'local' desktop display. I put an index.html page in a
new folder named 'www' in the same 'view/local' directory as
webserv.r is in as instructed. When keying the 'http://127.0.0.1:80'
into either Netscape or I.E....the browser replies
'page not found' and the log indicates a '404' message etc.
Any suggestions from anyone who might run this server? All 'default'
settings are currently
in effect. I run on Win 98.

On 01-Aug-03, johnfdutcher wrote:

> I have posted about this before....email won't work for me using the
> IOS LINK Client OR Rebol View....always get an unsuccessful error
> msg......suggesting I check my 'set-up' values.

> I stumbled onto something more to try...if I press 'setup' in the
> link client menu bar and then Email, I am presented with my current
> settings and a chance to test them. My current settings are:
>       dutcher@...
>       mail.localnet.com
>       mail.localnet.com

> I press 'test the connection'. It successfully makes a connection to
> the server, also successfully engages in the handshake, states that
> the whole exercise was successful.......I say 'save' and then it
> restarts IOS Link.

> Still email doesn't work....??? It seems as though I have fulfilled
> the early suggestion on this board that I must make a connection
> to my email server before the application can work for me.

> Any thoughts ??

Hi John,

I'm not an IOS user, so can't help you with that one.  Have you tried
joining the main REBOL mailing list?  (Find how to join on
www.rebol.com.)  This list doesn't have many on it I think and you're
more likely to get a reply there.

--
Carl Read

I have posted about this before....email won't work for me using the
IOS LINK Client OR Rebol View....always get an unsuccessful error
msg......suggesting I check my 'set-up' values.

I stumbled onto something more to try...if I press 'setup' in the
link client menu bar and then Email, I am presented with my current
settings and a chance to test them. My current settings are:
        dutcher@...
        mail.localnet.com
        mail.localnet.com

I press 'test the connection'. It successfully makes a connection to
the server, also successfully engages in the handshake, states that
the whole exercise was successful.......I say 'save' and then it
restarts IOS Link.

Still email doesn't work....??? It seems as though I have fulfilled
the early suggestion on this board that I must make a connection
to my email server before the application can work for me.

Any thoughts ??

You're the man,  Ingo  !!
The revised code works just fine !!

John

--- In Rebol_New@yahoogroups.com, Ingo Hohmann <ingo@2...> wrote:
> Hi John,
>
> I should _really_ not tell you, how long it took me to find out
it's one
> of the common errors you have to make a few times, and should never
> again forget thereafter ...
>
> Let's have a look at this:
>
>    length? data = 6
>
> When the Rebol interpreter evaluates this, it starts by lookging at
> 'length? it's a word that takes one argument, so you _think_ it
should
> be data, _but_ data is followed by '= which takes two arguments,
one on
> each side ...
>
> so, before 'length? gets a go at data, '= eats up data and 6, and
> returns either true or false, which really isn't a series.
>
> You have to possibilities
> - use brackets:
>     (length? data) = 6
> - (the preferred way by most fluent Rebolers)
>     6 = length? data
>
> After this you will hit the next problem, though (at least if I
> understood your snippet correctly) - you add the _string_ "1 2 3 4
5 6"
> to data, which will always be only 1 element in the block, I guess
what
> you really want is:
>
>    data: to block! ticket/text
>
> which returns a block, consisting of the string being understood as
> Rebol values.
>
> In the end your function looks like this:
>
>
> save-data: does [
>      series: make block! []
>      data: to block! ticket/text
>      probe data
>      print data
>      either 6 = length? data  [
>        print "Good number of values"
>      ][
>        print "Wrong number of values"
>      ]
> ]
>
>
> I hope that helped,
>
> Ingo
>
>
>
>
> johnfdutcher wrote:
> > The portion of a script below takes the value of a View 'field'
> > called 'ticket' and seems to successfully make a 'block' from it.
> >
> > I believed a block IS a series.....but the line that trys to
determine
> > the length? (for me, that's how many individual values does it
> > contain) NEVER works, resulting always in an error message that
> > says the subject of the function length? is NOT a series type !!
> >
> > When 'printing' the block ('data'), if 'ticket' contains: 1 2 3 4
5 6
> > one gets: 1 2 3 4 5 6
> > When 'probing' the block ('data'), one gets ["1 2 3 4 5 6"]
> > Doesn't that suggest one is dealing with a 'series' ??
> >
> > save-data: does [
> >     series: data: make block! []
> >     insert data ticket/text
> >     either length? data = 6 [print "Good number of values"
> >       ][
> >         print "Wrong number of values"
> >        ]
> >     probe data
> >     print data
> > ]
> >
> > Thanks, John D.
> >
> >
> >
> >
> > Community email addresses:
> >   Post message: Rebol_New@onelist.com
> >   Subscribe:    Rebol_New-subscribe@onelist.com
> >   Unsubscribe:  Rebol_New-unsubscribe@onelist.com
> >   List owner:   Rebol_New-owner@onelist.com
> >
> > Shortcut URL to this page:
> >   http://www.onelist.com/community/Rebol_New
> >
> > Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
> >
> >
>
> --
>
> --
> YES!   That's just me, just being!    http://www.2b1.de/
> We ARE all ONE  ---   ingo@2...   ---  We ARE all FREE

Hi John,

I should _really_ not tell you, how long it took me to find out it's one
of the common errors you have to make a few times, and should never
again forget thereafter ...

Let's have a look at this:

    length? data = 6

When the Rebol interpreter evaluates this, it starts by lookging at
'length? it's a word that takes one argument, so you _think_ it should
be data, _but_ data is followed by '= which takes two arguments, one on
each side ...

so, before 'length? gets a go at data, '= eats up data and 6, and
returns either true or false, which really isn't a series.

You have to possibilities
- use brackets:
     (length? data) = 6
- (the preferred way by most fluent Rebolers)
     6 = length? data

After this you will hit the next problem, though (at least if I
understood your snippet correctly) - you add the _string_ "1 2 3 4 5 6"
to data, which will always be only 1 element in the block, I guess what
you really want is:

    data: to block! ticket/text

which returns a block, consisting of the string being understood as
Rebol values.

In the end your function looks like this:


save-data: does [
      series: make block! []
      data: to block! ticket/text
      probe data
      print data
      either 6 = length? data  [
        print "Good number of values"
      ][
        print "Wrong number of values"
      ]
]


I hope that helped,

Ingo




johnfdutcher wrote:
> The portion of a script below takes the value of a View 'field'
> called 'ticket' and seems to successfully make a 'block' from it.
>
> I believed a block IS a series.....but the line that trys to determine
> the length? (for me, that's how many individual values does it
> contain) NEVER works, resulting always in an error message that
> says the subject of the function length? is NOT a series type !!
>
> When 'printing' the block ('data'), if 'ticket' contains: 1 2 3 4 5 6
> one gets: 1 2 3 4 5 6
> When 'probing' the block ('data'), one gets ["1 2 3 4 5 6"]
> Doesn't that suggest one is dealing with a 'series' ??
>
> save-data: does [
>     series: data: make block! []
>     insert data ticket/text
>     either length? data = 6 [print "Good number of values"
>       ][
>         print "Wrong number of values"
>        ]
>     probe data
>     print data
> ]
>
> Thanks, John D.
>
>
>
>
> Community email addresses:
>   Post message: Rebol_New@onelist.com
>   Subscribe:    Rebol_New-subscribe@onelist.com
>   Unsubscribe:  Rebol_New-unsubscribe@onelist.com
>   List owner:   Rebol_New-owner@onelist.com
>
> Shortcut URL to this page:
>   http://www.onelist.com/community/Rebol_New
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>

--

--
YES!   That's just me, just being!    http://www.2b1.de/
We ARE all ONE  ---   ingo@...   ---  We ARE all FREE

The portion of a script below takes the value of a View 'field'
called 'ticket' and seems to successfully make a 'block' from it.

I believed a block IS a series.....but the line that trys to determine
the length? (for me, that's how many individual values does it
contain) NEVER works, resulting always in an error message that
says the subject of the function length? is NOT a series type !!

When 'printing' the block ('data'), if 'ticket' contains: 1 2 3 4 5 6
one gets: 1 2 3 4 5 6
When 'probing' the block ('data'), one gets ["1 2 3 4 5 6"]
Doesn't that suggest one is dealing with a 'series' ??

save-data: does [
     series: data: make block! []
     insert data ticket/text
     either length? data = 6 [print "Good number of values"
       ][
         print "Wrong number of values"
        ]
     probe data
     print data
]

Thanks, John D.

I have used the 'blocked' data style Rebol format for a simple
database project....and it worked beautifully. Does adds, deletes,
searches quickly and accurately. There was a favorable format to the
data however....all 'records' contained all 'value instances' in each
record.
I am considering using the same approach for another project...but
there are many fields in the 'records' which would occur 'varying'
numbers of times or not at all (that is, arrays exist with varying
numbers of occurrences, or none at all). In a fixed record format
they would just exist unvalued at their position. Rebol blocks don't
really seem as though they lend themselves to varyingly populated or
unpopulated blocks....though I consider 'fixed repetitions' of quoted
space characters to 'cover' for the non-valued fields leaving all
records with an essentially fixed number of 'fields' whether in
arrays or otherwise (so that 'looping' logic would always find the
same data in the same position).
Has anyone any experience in trying to use Rebol style data blocks
where this condition exists.....perhaps it's a joke to even think
about doing something this way ?

On 22-Jul-03, johnfdutcher wrote:

> Has anyone tried the Rebol SDK Encapsulator ?

> My little baby is just about born; and I'd like to think it could be
> wrapped in the Encapsulator as suggested....and give me a nice one-
> file install.....but for $250.00...I'd be curious if anyone has had
> good success with it ?

Hi John,

I haven't tried it, but I've heard it works fine, and of course it's
not all you get with the SDK.

Escribe's not working for me at the moment, but I recalled the
following thread about Encap and found it with Graham's search
engine.  "Guest" for name and password...

http://203.79.110.37/rebolml/rep-getmsg?m_no=25825

Those prices there are out of date now of course.

AltMe at http://www.altme.com/ is probably the most wellknown Encapped
REBOL program.

--
Carl Read

Has anyone tried the Rebol SDK Encapsulator ?

My little baby is just about born; and I'd like to think it could be
wrapped in the Encapsulator as suggested....and give me a nice one-
file install.....but for $250.00...I'd be curious if anyone has had
good success with it ?

John D.

THanks much Carl,

The 'area/para/scroll: 0x0' seems to work perfectly !

John D.



--- In Rebol_New@yahoogroups.com, Carl Read <carl@c...> wrote:
> On 19-Jul-03, johnfdutcher wrote:
>
> > I attached a slider to an area ....it scrolls the text nicely.
>
> > When closing the window with the area in it I set the slider/data
to
> > zero so it will be at the top of the bar when the next display is
> > requested. The 'area', however, does not display the next text
> > output at the top....if the window closed with the text at its
> > bottom...the next re-display displays at the bottom of the text as
> > well. (The slider meantime is at its top by virtue of the
> > slider/data: 0). How do I get the 'area' to also re-display at its
> > top rather than at the bottom where it was left the previous
time ?
>
> > John D.
>
> Hi John,
>
> This, I think, should do what you want...
>
> view layout [
>     a: area mold :view
>     do [a/para/scroll: 0x0]
> ]
>
> If you look at a/para after running that you'll be shown this
> object...
>
> make object! [
>     origin: 2x2
>     margin: 2x2
>     indent: 0x0
>     tabs: 40
>     wrap?: false
>     scroll: 0x0
> ]
>
> and scroll determines the position of the text string that's
displayed
> in 'area.
>
> As I didn't know how to do this before you asked, (or had
forgotten;),
> here's how I figured it out:
>
> I first did a probe of the area object...
>
> >> probe first a
> [self type offset size span pane text color image effect data edge
> font para feel saved-area rate show? options parent-face old-offset
> old-size line-list changes face-flags action state style alt-action
> facets related words colors texts file var keycode reset styles init
> multi blinker pane-size dirty? help user-data flags]
>
> I then probed a/text, a/data and a/texts to see if any of them were
> not showing the string at its head, as I assumed that's how area
> would show part of a string - by altering its index.  But it's not
> how it's done, so that was a wild goose chase.
>
> I then probed a/feel to see if that'd give any hints, but that
wasn't
> any help either.  So, having exausted that approach, I decided to
> search the REBOL list archive online at Escribe here...
> http://www.escribe.com/internet/rebol/ , and searching for "area
> head" eventually got me to here...
>
> http://www.escribe.com/internet/rebol/m11896.html
>
> which describes the (undocumented function) scroll-para - which I
> noticed uses para/scroll - which proved to be what was needed!
>
> --
> Carl Read