Z/OS port of Regina REXX

[David Crayford]

Rick Troff asked me to make my z/OS port of Regina REXX public so I forked the 
repo and pushed it up here https://github.com/daveyc/regina

I had to dust this off a bit and make sure it still worked. It does. Some 
important notes:

*   I have included all the binaries in the repo. You can simply clone the 
repo and set your PATH and LIBPATH environment variables to include the path. 
If you run “make install” it will attempt to install into /usr/local
*   It’s a ASCII port so requires the same environment setup as Rocket and 
IBM ported tools.
*   I switched the build to the new z/OS XL Open C/C++ Clang compiler so if 
you want to build it or extensions make sure you have the compiler installed. 
*   I don’t know why but Regina has two interpreters regina and rexx. Only 
regina supports dynamic calls so to use regutil with all the cool extensions 
always use regina. In fact, always use regina!

Issues are disable and I do not support this port. 
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Rick Troth]

This is excellent news!
Thanks!

-- R; <><


On 5/3/23 03:25, David Crayford wrote:

Rick Troff asked me to make my z/OS port of Regina REXX public so I forked the 
repo and pushed it up here https://github.com/daveyc/regina

I had to dust this off a bit and make sure it still worked. It does. Some 
important notes:

*   I have included all the binaries in the repo. You can simply clone the 
repo and set your PATH and LIBPATH environment variables to include the path. 
If you run “make install” it will attempt to install into /usr/local
*   It’s a ASCII port so requires the same environment setup as Rocket and 
IBM ported tools.
*   I switched the build to the new z/OS XL Open C/C++ Clang compiler so if 
you want to build it or extensions make sure you have the compiler installed.
*   I don’t know why but Regina has two interpreters regina and rexx. Only 
regina supports dynamic calls so to use regutil with all the cool extensions 
always use regina. In fact, always use regina!

Issues are disable and I do not support this port.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[René Jansen]

Great news David, thank you.

best regards,

René.

> On 3 May 2023, at 09:25, David Crayford  wrote:
> 
> Rick Troff asked me to make my z/OS port of Regina REXX public so I forked 
> the repo and pushed it up here https://github.com/daveyc/regina
> 
> I had to dust this off a bit and make sure it still worked. It does. Some 
> important notes:
> 
> * I have included all the binaries in the repo. You can simply clone the 
> repo and set your PATH and LIBPATH environment variables to include the path. 
> If you run “make install” it will attempt to install into /usr/local
> * It’s a ASCII port so requires the same environment setup as Rocket and 
> IBM ported tools.
> * I switched the build to the new z/OS XL Open C/C++ Clang compiler so if 
> you want to build it or extensions make sure you have the compiler installed. 
> * I don’t know why but Regina has two interpreters regina and rexx. Only 
> regina supports dynamic calls so to use regutil with all the cool extensions 
> always use regina. In fact, always use regina!
> 
> Issues are disable and I do not support this port. 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[David Crayford]

I just realised I misspelt your surname. Sorry Rick!

Note that there is a “zos” branch in the Github repository that you will have 
the checkout. 

As I previously mentioned this port could be patched to support EBCDIC but you 
will have a heavy lift patching the Flex lexer and Bison tables. Oh well, 
that’s life dealing with the major suckage that is EBCDIC! Also, the 
interpreter uses 'fopen(filename, “rb”)’ so can in theory can handle source 
files in MVS data sets. Unfortunately, the
parser expects line-ending characters so is not compatible with record oriented 
files.

> On 3 May 2023, at 11:03 pm, Rick Troth  wrote:
> 
> This is excellent news!
> Thanks!
> 
> -- R; <><
> 
> 
> On 5/3/23 03:25, David Crayford wrote:
>> Rick Troff asked me to make my z/OS port of Regina REXX public so I forked 
>> the repo and pushed it up here https://github.com/daveyc/regina
>> 
>> I had to dust this off a bit and make sure it still worked. It does. Some 
>> important notes:
>> 
>> *I have included all the binaries in the repo. You can simply clone the 
>> repo and set your PATH and LIBPATH environment variables to include the 
>> path. If you run “make install” it will attempt to install into /usr/local
>> *It’s a ASCII port so requires the same environment setup as Rocket and 
>> IBM ported tools.
>> *I switched the build to the new z/OS XL Open C/C++ Clang compiler so if 
>> you want to build it or extensions make sure you have the compiler installed.
>> *I don’t know why but Regina has two interpreters regina and rexx. Only 
>> regina supports dynamic calls so to use regutil with all the cool extensions 
>> always use regina. In fact, always use regina!
>> 
>> Issues are disable and I do not support this port.
>> --
>> For IBM-MAIN subscribe / signoff / archive access instructions,
>> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Paul Gilmartin]

On Thu, 4 May 2023 08:39:41 +0800, David Crayford wrote:
>
>Note that there is a “zos” branch in the Github repository that you will have 
>the checkout. 
>
Not in SourceForge, which is mostly mentioned on 
?

What path to Github?  How is that related to:
?

>As I previously mentioned this port could be patched to support EBCDIC but you 
>will have a heavy lift patching the Flex lexer and Bison tables. 
>
Does/would autoconversion help?  (Of course, not for Classic data sets.  But 
DSFS?)

Oh well, that’s life dealing with the major suckage that is EBCDIC! Also, the 
interpreter uses 'fopen(filename, “rb”)’ so can in theory can handle source 
files in MVS data sets. Unfortunately, the
>parser expects line-ending characters so is not compatible with record 
>oriented files.
>
Doesn't XLC fake the "\n" for text files?

How about something like "ADDRESS LINKMVS WITH ?
Challenging because MVS never had an organized concept of stdin, stdout, stderr.

Is there a User's Guide?

--  
Thanks,
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[David Crayford]

> On 5 May 2023, at 1:48 am, Paul Gilmartin 
> <042bfe9c879d-dmarc-requ...@listserv.ua.edu> wrote:
> 
> On Thu, 4 May 2023 08:39:41 +0800, David Crayford wrote:
>> 
>> Note that there is a “zos” branch in the Github repository that you will 
>> have the checkout. 
>> 
> Not in SourceForge, which is mostly mentioned on 
> ?

I don’t do SourceForge and the svn nonsense. Git is the dominant SCM and has 
been ported and is supported on z/OS. 


> 
> What path to Github?  How is that related to:
> ?
> 
>> As I previously mentioned this port could be patched to support EBCDIC but 
>> you will have a heavy lift patching the Flex lexer and Bison tables. 
>> 
> Does/would autoconversion help?  (Of course, not for Classic data sets.  But 
> DSFS?)

It uses autoconversion. For programming languages you have to pick a codesetd. 
I’m sure DSFS would work.

> 
> Oh well, that’s life dealing with the major suckage that is EBCDIC! Also, the 
> interpreter uses 'fopen(filename, “rb”)’ so can in theory can handle source 
> files in MVS data sets. Unfortunately, the
>> parser expects line-ending characters so is not compatible with record 
>> oriented files.
>> 
> Doesn't XLC fake the "\n" for text files?

Yes, with ‘fopen(filename, “r”)’ in text mode. Regina opens the file in binary 
mode so it can parse by character. 

> 
> How about something like "ADDRESS LINKMVS WITH ?
> Challenging because MVS never had an organized concept of stdin, stdout, 
> stderr.
> 
> Is there a User's Guide?
> 

The doc is on 
https://sourceforge.net/projects/regina-rexx/files/regina-documentation/ and is 
good. 

> --  
> Thanks,
> gil
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[René Jansen]

> 
> I don’t do SourceForge and the svn nonsense. Git is the dominant SCM and has 
> been ported and is supported on z/OS. 
> 
I agree 100% here.

René.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Mike Schwab]

On Wed, May 3, 2023 at 7:40 PM David Crayford  wrote:
>

>
> As I previously mentioned this port could be patched to support EBCDIC but 
> you will have a heavy lift patching the Flex lexer and Bison tables. Oh well, 
> that’s life dealing with the major suckage that is EBCDIC! Also, the 
> interpreter uses 'fopen(filename, “rb”)’ so can in theory can handle source 
> files in MVS data sets. Unfortunately, the
> parser expects line-ending characters so is not compatible with record 
> oriented files.
>
You could create a patch for the most common EBCDIC code page
(US.-ENG), then modify and test for other code pages.
--
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Paul Gilmartin]

On Fri, 5 May 2023 12:12:46 -0500, Mike Schwab wrote:

>On Wed, May 3, 2023 at 7:40 PM David Crayford wrote:
>
>> ... Oh well, that’s life dealing with the major suckage that is EBCDIC! 
>>
>You could create a patch for the most common EBCDIC code page
>(US.-ENG), then modify and test for other code pages.
> 
Stay with IBM-1047, preferred by z/OS UNIX and supported by autoconversion.
Don't nurture "major suckage".

But what about the  '¬'?
If I copy-and-;aste it from regina.pdf it doesn't work as a negator, and:
say c2x( '¬' )
C2AC
(It works if I translate it to IBM-1252.)  Text in manuals should Just Work
when copied into programs.

-- 
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Mike Schwab]

Well, some mainframe compilers require a different code page for
special characters they require, so anybody using the compiler might
need a REXXPLI or REXXAPL version.

On Fri, May 5, 2023 at 12:47 PM Paul Gilmartin
<042bfe9c879d-dmarc-requ...@listserv.ua.edu> wrote:
>
> On Fri, 5 May 2023 12:12:46 -0500, Mike Schwab wrote:
>
> >On Wed, May 3, 2023 at 7:40 PM David Crayford wrote:
> >
> >> ... Oh well, that’s life dealing with the major suckage that is EBCDIC!
> >>
> >You could create a patch for the most common EBCDIC code page
> >(US.-ENG), then modify and test for other code pages.
> >
> Stay with IBM-1047, preferred by z/OS UNIX and supported by autoconversion.
> Don't nurture "major suckage".
>
> But what about the  '¬'?
> If I copy-and-;aste it from regina.pdf it doesn't work as a negator, and:
> say c2x( '¬' )
> C2AC
> (It works if I translate it to IBM-1252.)  Text in manuals should Just Work
> when copied into programs.
>
> --
> gil
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN



-- 
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Seymour J Metz]

What code page are you using? What release of Regina? In some code pages Not is 
'AA'X, in others 'AC'X. In Unicode it's U+00AC ('C2AC'X in UTF-8).


From: IBM Mainframe Discussion List  on behalf of 
Paul Gilmartin <042bfe9c879d-dmarc-requ...@listserv.ua.edu>
Sent: Friday, May 5, 2023 1:46 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Z/OS port of Regina REXX

On Fri, 5 May 2023 12:12:46 -0500, Mike Schwab wrote:

>On Wed, May 3, 2023 at 7:40 PM David Crayford wrote:
>
>> ... Oh well, that’s life dealing with the major suckage that is EBCDIC!
>>
>You could create a patch for the most common EBCDIC code page
>(US.-ENG), then modify and test for other code pages.
>
Stay with IBM-1047, preferred by z/OS UNIX and supported by autoconversion.
Don't nurture "major suckage".

But what about the  '¬'?
If I copy-and-;aste it from regina.pdf it doesn't work as a negator, and:
say c2x( '¬' )
C2AC
(It works if I translate it to IBM-1252.)  Text in manuals should Just Work
when copied into programs.

--
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Paul Gilmartin]

On Fri, 5 May 2023 22:25:09 +, Seymour J Metz wrote:

>What code page are you using? What release of Regina? In some code pages Not 
>is 'AA'X, in others 'AC'X. In Unicode it's U+00AC ('C2AC'X in UTF-8).
>
533 $ locale 
LC_COLLATE="en_US.UTF-8"

 trace R; parse version .
 1 *-* parse version .
   >.>   "REXX-Regina_3.9.5(MT) 5.00 25 Jun 2022"

I have no displayable '¬' that works with that Regina, but:
"trace R; parse source .; interpret 'say' 'ac'x 0
 1 *-* parse source .
   >.>   "UNIX SUBROUTINE /tmp/Linux-4.19.0-24-amd64-paulgilm/Rx_4088"
   *-* interpret 'say' 'ac'x 0 
   *-* interpret 'say' 'ac'x 0 
1

On what platform(s)/code page is  it 'AA'X  Does it work there?
Sith what version of Regina?

-- 
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Seymour J Metz]

I don't recall the details, but AA is  '¬' on, e.g., CP852 and AC is  '¬' on, 
e.g., CP850.


From: IBM Mainframe Discussion List  on behalf of 
Paul Gilmartin <042bfe9c879d-dmarc-requ...@listserv.ua.edu>
Sent: Friday, May 5, 2023 7:22 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Z/OS port of Regina REXX

On Fri, 5 May 2023 22:25:09 +, Seymour J Metz wrote:

>What code page are you using? What release of Regina? In some code pages Not 
>is 'AA'X, in others 'AC'X. In Unicode it's U+00AC ('C2AC'X in UTF-8).
>
533 $ locale
LC_COLLATE="en_US.UTF-8"

 trace R; parse version .
 1 *-* parse version .
   >.>   "REXX-Regina_3.9.5(MT) 5.00 25 Jun 2022"

I have no displayable '¬' that works with that Regina, but:
"trace R; parse source .; interpret 'say' 'ac'x 0
 1 *-* parse source .
   >.>   "UNIX SUBROUTINE /tmp/Linux-4.19.0-24-amd64-paulgilm/Rx_4088"
   *-* interpret 'say' 'ac'x 0
   *-* interpret 'say' 'ac'x 0
1

On what platform(s)/code page is  it 'AA'X  Does it work there?
Sith what version of Regina?

--
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Paul Gilmartin]

On Fri, 5 May 2023 23:53:52 +, Seymour J Metz wrote:

>I don't recall the details, but AA is  '¬' on, e.g., CP852 and AC is  '¬' on, 
>e.g., CP850.

I see AA in both.  Look like Windows box drawing code pages.  ("not"
is different from "upper right corner".)

--
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Z/OS port of Regina REXX

[Seymour J Metz]

Sorry, I must have been looking at CP437.


From: IBM Mainframe Discussion List  on behalf of 
Paul Gilmartin <042bfe9c879d-dmarc-requ...@listserv.ua.edu>
Sent: Friday, May 5, 2023 8:10 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Z/OS port of Regina REXX

On Fri, 5 May 2023 23:53:52 +, Seymour J Metz wrote:

>I don't recall the details, but AA is  '¬' on, e.g., CP852 and AC is  '¬' on, 
>e.g., CP850.

I see AA in both.  Look like Windows box drawing code pages.  ("not"
is different from "upper right corner".)

--
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN