Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, 17 Apr 2012 22:23:09 -0500, John McKown wrote: I do use a UNIX subdirectory on my SYSEXEC concatenation. But, as you said, it cannot be first. So I have an empty PDS with FB/80 as the first DSN in the concatenation. A clumsy work around, but at least it works for me. When you do this, do you get sporadic ABENDs, typically 0C4? I never get more than one per ISPF session -- after one I seem to be immune. And it happens only on ENDing a panel, so in a sense it's harmless; I can endure it. Ops probly doesn't welcome the dumps. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
Hi Tom, In respect of your recent post on the issue of the title, and your last paragraph: It clearly says only directories and paths, but I suppose a confirmation that datasets are not supported would be nice. I will try to get that into the next version of the Programming Guide. If anyone has a better suggestion for where this confirmation would go, let me know! In Chapter 23 Compiler direction statements Page 520 contains information about the COPY Statement which states: When compiling from JCL or TSO, only the first eight characters are used as the identifying name. When compiling with the cob2 command and processing COPY text residing in the Hierarchical File System (HFS), all characters are significant. By implication the restriction information is in the manual so perhaps a cross reference of the two items is all that is required. Kind Regards - Terry Director KMS-IT Limited 228 Abbeydale Road South Dore Sheffield S17 3LA UK Reg : 3767263 Outgoing e-mails have been scanned, but it is the recipients responsibility to ensure their anti-virus software is up to date. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, Apr 17, 2012 at 10:33 PM, John McKown joa...@swbell.net wrote: Guess I am spoiled by the assembler (as command) and C/C++ compiler's support of both UNIX paths and z/OS PDS[E]s. Spoiled by common sense :-) z/OS Unix commands should not be walled out from z/OS datasets. I have a, to me, clumsy way to get around it. I started up the NFS server and exported the high level qualifier that had the PDS[E]s that I needed, then did a MOUNT onto a z/OS UNIX subdirectory. Clumsy, but it does work. Its especially silly considering that BPAM already supports reading concatenations of PDS[E] and Unix directories.Apparently, the cob2 developer, or maybe the underlying COBOL compiler has protected itself from this support. I may write my own version of the cob2 command which does what _I_ want. Should be fairly easy. Just set up the allocations as needed using DYNALLOC, then use BPX1ATM UNIX routine to ATTACH the IGYCRCTL program. Or with a pretty simple REXX shell script. It might be that the COBOL compiler (IGYCRCTL) has open exits that detects SYSLIB mixed PDS/Unix contatenations (but unlikely I would guess). The main problem that I have with my UNIX programs is that I can only use HLASM (no C compiler license). And I have not figured out how to handle UNIX signals. So the cntlc to abort a command doesn't work. see signal() On Tue, 2012-04-17 at 15:04 -0700, Tom Ross wrote: On 4/16/2012 3:26 PM, McKown, John wrote: snip It clearly says only directories and paths, but I suppose a confirmation that datasets are not supported would be nice. I will try to get that into the next version of the Programming Guide. If anyone has a better suggestion for where this confirmation would go, let me know! Cheers, TomR COBOL is the Language of the Future! -- John McKown Maranatha! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, 17 Apr 2012 22:23:09 -0500, John McKown wrote: I do use a UNIX subdirectory on my SYSEXEC concatenation. But, as you said, it cannot be first. So I have an empty PDS with FB/80 as the first DSN in the concatenation. A clumsy work around, but at least it works for me. Me, too. It could even be a temp DSN; NEW,DELETE. One drawback, though, is that DDLIST won't show members in the USS catenand(s). PMR submitted, years ago. WAD. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
I've tried, but I seem to have a mental block on getting the NFS server working on z/OS. I use the client to a Linux server with no problem. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Jerry Whitteridge Sent: Monday, April 16, 2012 6:29 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Anybody use the cob2 command on a UNIX shell to compile COBOL? Has anyone tried (apart from Paul G) exporting the PDSE via NFS and mounting it at a z/Unix mountpoint on the same system ? That should be able to provide your path as well a classic access Jerry Whitteridge Lead Systems Programmer Safeway Inc. 925 951 4184 If you feel in control you just aren't going fast enough. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Steve Comstock Sent: Monday, April 16, 2012 3:21 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Anybody use the cob2 command on a UNIX shell to compile COBOL? On 4/16/2012 3:26 PM, McKown, John wrote: If so, have you figured out how to specify a PDS as an include or copybooksource? The documentation for this command basically stinks. There is an -I switch. But it apparently only accepts UNIX path specifications. The ld command (binder) accepts a PDS name via -L //pds.name. The as command (HLASM) and C compilers accept a -I //pds.name. I may end up writing my own version of the cob2 command, if I really decide that I want to compile COBOL from the UNIX shell. I'll likely model it after the as command. Yes. I discuss this command in our course Developing Applications for z/OS UNIX. But you're right, the doc is very poor. The assumption seems to be that copy books must be in HFS directories and my experiments produce error messages that would support that. In a way, that's too bad. OTOH, I, too, was assuming that copy books would reside in HFS directories so I had never tried to access copy books as members of a PDS before. You would think that if you used the classic clue that a library was a PDS/E, the compiler could figure it out. Maybe: export SYSLIB=//'SCOMSTO.U520.LIBRARY' which I tried, and the message from the compile comes back: LineID Message code Library phase message text 24 IGYLI0049-S The COPY library SCOMSTO.U520.LIBRARY/APODEFS was not found. Skipped to the period terminating the COPY statement. so it was pretty clearly expecting (nay: requiring) a z/OS UNIX file, not a PDS. I'm copying this over to ibm-main, too, as Tom Ross occasionally follows that group so maybe he will chime in with some info. John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For MVS-OE subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO MVS-OE -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
Thanks. It's reassuring that I'm no stupider than usual. I don't have PL/I, but from looking at the manual, it seems to suffer from the same problem. http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/ibm3pg80/1.4.1.4 And FORTRAN doesn't even seem to support any z/OS UNIX usage at all. There is nothing in the books. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Steve Comstock Sent: Monday, April 16, 2012 5:21 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Anybody use the cob2 command on a UNIX shell to compile COBOL? On 4/16/2012 3:26 PM, McKown, John wrote: If so, have you figured out how to specify a PDS as an include or copybooksource? The documentation for this command basically stinks. There is an -I switch. But it apparently only accepts UNIX path specifications. The ld command (binder) accepts a PDS name via -L //pds.name. The as command (HLASM) and C compilers accept a -I //pds.name. I may end up writing my own version of the cob2 command, if I really decide that I want to compile COBOL from the UNIX shell. I'll likely model it after the as command. Yes. I discuss this command in our course Developing Applications for z/OS UNIX. But you're right, the doc is very poor. The assumption seems to be that copy books must be in HFS directories and my experiments produce error messages that would support that. In a way, that's too bad. OTOH, I, too, was assuming that copy books would reside in HFS directories so I had never tried to access copy books as members of a PDS before. You would think that if you used the classic clue that a library was a PDS/E, the compiler could figure it out. Maybe: export SYSLIB=//'SCOMSTO.U520.LIBRARY' which I tried, and the message from the compile comes back: LineID Message code Library phase message text 24 IGYLI0049-S The COPY library SCOMSTO.U520.LIBRARY/APODEFS was not found. Skipped to the period terminating the COPY statement. so it was pretty clearly expecting (nay: requiring) a z/OS UNIX file, not a PDS. I'm copying this over to ibm-main, too, as Tom Ross occasionally follows that group so maybe he will chime in with some info. John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For MVS-OE subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO MVS-OE -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Mon, 16 Apr 2012 18:38:22 -0500, Paul Gilmartin paulgboul...@aim.com wrote:
On Mon, 16 Apr 2012 17:29:23 -0600, Jerry Whitteridge wrote:
Has anyone tried (apart from Paul G) exporting the PDSE via NFS and mounting
it at a z/Unix mountpoint on the same system ? That should be able to provide
your path as well a classic access
We tried and failed. But we didn't try very hard because I
couldn't justify expending much systems programmer resource
on my largely experimental need. We have legacy data sets
exported and mounted on Solaris mountpoints. We have
Solaris filesystems exported and mounted on z/OS. If I need
something in both places, I keep it on a Solaris server.
John M. may have tried; I don't know with what degree of success.
I hadn't done it with a PDSE, but I had with a PDS. I just tried it and it
works
the same. :-)
I ran an exec similar to this from TSO to mount the PDSE on the local system:
/* rexx */
Address TSO
MOUNT FILESYSTEM(NFS_ZELD) TYPE(NFS) ,
MOUNTPOINT('/u/zelden/testnfs') ,
PARM('SYST:ZELDEN.TEST.PDSE,text,xlat(Y)')
You can change ZELDEN.TEST.PDSE to ZELDEN.TEST in the mount parm, then all
my data sets
starting ZELDEN.TEST are mounted under /u/zelden/testnfs/ and the PDSE members
fall under
/u/zelden/testnfs/PDSE/*.
My export looks like this, so I can write to that PDSE as well:
ZELDEN-access=SYST|SYSX,rw=SYST|SYSX
Mark
--
Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS
mailto:m...@mzelden.com
Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html
Systems Programming expert at http://expertanswercenter.techtarget.com/
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, 17 Apr 2012 10:59:59 -0500, Mark Zelden wrote:
I ran an exec similar to this from TSO to mount the PDSE on the local system:
/* rexx */
Address TSO
MOUNT FILESYSTEM(NFS_ZELD) TYPE(NFS) ,
MOUNTPOINT('/u/zelden/testnfs') ,
PARM('SYST:ZELDEN.TEST.PDSE,text,xlat(Y)')
Can you automount?
How do permissions work? (From Solaris we use mvslogin.)
Ideally, on a single system it should be transparent to RACF:
Whatever permissions you have on a data set should be
available when it's mounted as a filesystem.
You can change ZELDEN.TEST.PDSE to ZELDEN.TEST in the mount parm, then
all my data sets
starting ZELDEN.TEST are mounted under /u/zelden/testnfs/ and the PDSE
members fall under
/u/zelden/testnfs/PDSE/*.
Can you strip the parm to null string so all exported
data sets can be mounted? (This is how we mount
data sets on Solaris.)
My export looks like this, so I can write to that PDSE as well:
ZELDEN-access=SYST|SYSX,rw=SYST|SYSX
Thanks,
gil
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
To the original question, you should be able to use a Unix directory and files in place of a PDS or PDSE in any program (with limitations), since BPAM provides simulation for these. Details can be found in z/OS DFSMS Using Datasets under Processing z/OS Unix files - http://publibfp.dhe.ibm.com/cgi-bin/bookmgr/BOOKS/dgt2d440/3.9?DT=20050602124524 So, allocate the equivalent of SYSLIB as a contenation with one DD pointing to a Unix directory and invoke the COBOL compiler. Even in the absence of a shell command front-end, you could write your own REXX shell script to do allocations via BPXWDYN and then invoke the COBOL compiler with LINKMVS. Similarly, it should not be difficult to write a REXX shell script for FORTRAN. Kirk Wolf Dovetailed Technologies http://dovetail.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, 17 Apr 2012 12:35:19 -0500, Paul Gilmartin paulgboul...@aim.com wrote:
On Tue, 17 Apr 2012 10:59:59 -0500, Mark Zelden wrote:
I ran an exec similar to this from TSO to mount the PDSE on the local system:
/* rexx */
Address TSO
MOUNT FILESYSTEM(NFS_ZELD) TYPE(NFS) ,
MOUNTPOINT('/u/zelden/testnfs') ,
PARM('SYST:ZELDEN.TEST.PDSE,text,xlat(Y)')
Can you automount?
Don't know. Never tried nor looked into it. I put required NFS mounts for
production
(needed for distributed unix) either at the end of /etc/rc or into a script
executed at
the end of /etc/rc and use the mountx rexx shell script provided in the
/samples directory.
How do permissions work? (From Solaris we use mvslogin.)
Ideally, on a single system it should be transparent to RACF:
Whatever permissions you have on a data set should be
available when it's mounted as a filesystem.
Different options. RTFM.
Mark
--
Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS
mailto:m...@mzelden.com
Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html
Systems Programming expert at http://expertanswercenter.techtarget.com/
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
As the OP, I thought I'm mention that I finally got the z/OS NFS server to allow me to mount a high-level onto a z/OS UNIX subdirectory on the same system. I could then successfully use the cob2 command to compile a COBOL program which did a COPY CEEIGZCT from the mounted subdirectory. I mounted them at /$SYSNAME/nfs-legacy/hlq. I.e. I put -I /\$SYSNAME/nfs-legacy/cee/SCEESAMP on my cob2 command. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On 4/17/2012 12:53 PM, McKown, John wrote: As the OP, I thought I'm mention that I finally got the z/OS NFS server to allow me to mount a high-level onto a z/OS UNIX subdirectory on the same system. I could then successfully use the cob2 command to compile a COBOL program which did a COPY CEEIGZCT from the mounted subdirectory. I mounted them at /$SYSNAME/nfs-legacy/hlq. I.e. I put -I /\$SYSNAME/nfs-legacy/cee/SCEESAMP on my cob2 command. Clever and resourceful. But certainly just a workaround for the situation: you would like the compilers to be able to use MVS data sets directly for SYSLIB (and other DD names). -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
Definitely I would prefer that the command directly supported PDS PDSE datasets. NFS is a horrible overhead to me, being a CPU miser. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Steve Comstock Sent: Tuesday, April 17, 2012 2:07 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Anybody use the cob2 command on a UNIX shell to compile COBOL? On 4/17/2012 12:53 PM, McKown, John wrote: As the OP, I thought I'm mention that I finally got the z/OS NFS server to allow me to mount a high-level onto a z/OS UNIX subdirectory on the same system. I could then successfully use the cob2 command to compile a COBOL program which did a COPY CEEIGZCT from the mounted subdirectory. I mounted them at /$SYSNAME/nfs-legacy/hlq. I.e. I put -I /\$SYSNAME/nfs-legacy/cee/SCEESAMP on my cob2 command. Clever and resourceful. But certainly just a workaround for the situation: you would like the compilers to be able to use MVS data sets directly for SYSLIB (and other DD names). -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
Awesome - glad we could get it worked out. Jerry Whitteridge Lead Systems Programmer Safeway Inc. 925 951 4184 If you feel in control you just aren't going fast enough. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of McKown, John Sent: Tuesday, April 17, 2012 11:54 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Anybody use the cob2 command on a UNIX shell to compile COBOL? As the OP, I thought I'm mention that I finally got the z/OS NFS server to allow me to mount a high-level onto a z/OS UNIX subdirectory on the same system. I could then successfully use the cob2 command to compile a COBOL program which did a COPY CEEIGZCT from the mounted subdirectory. I mounted them at /$SYSNAME/nfs-legacy/hlq. I.e. I put -I /\$SYSNAME/nfs-legacy/cee/SCEESAMP on my cob2 command. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN Email Firewall made the following annotations. -- Warning: All e-mail sent to this address will be received by the corporate e-mail system, and is subject to archival and review by someone other than the recipient. This e-mail may contain proprietary information and is intended only for the use of the intended recipient(s). If the reader of this message is not the intended recipient(s), you are notified that you have received this message in error and that any review, dissemination, distribution or copying of this message is strictly prohibited. If you have received this message in error, please notify the sender immediately. == -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On 4/16/2012 3:26 PM, McKown, John wrote: If so, have you figured out how to specify a PDS as an include or copybooksource? The documentation for this command basically stinks. There is an -I switch. But it apparently only accepts UNIX path specifications. The ld command (binder) accepts a PDS name via -L //pds.name. The as command (HLASM) and C compilers accept a -I //pds.name. I may end up writing my own version of the cob2 command, if I really decide that I want to compile COBOL from the UNIX shell. I'll likely model it after the as command. Yes. I discuss this command in our course Developing Applications for z/OS UNIX. But you're right, the doc is very poor. The assumption seems to be that copy books must be in HFS directories and my experiments produce error messages that would support that. In a way, that's too bad. OTOH, I, too, was assuming that copy books would reside in HFS directories so I had never tried to access copy books as members of a PDS before. You would think that if you used the classic clue that a library was a PDS/E, the compiler could figure it out. Maybe: export SYSLIB=//'SCOMSTO.U520.LIBRARY' which I tried, and the message from the compile comes back: LineID Message code Library phase message text 24 IGYLI0049-S The COPY library SCOMSTO.U520.LIBRARY/APODEFS was not found. Skipped to the period terminating the COPY statement. so it was pretty clearly expecting (nay: requiring) a z/OS UNIX file, not a PDS. I'm copying this over to ibm-main, too, as Tom Ross occasionally follows that group so maybe he will chime in with some info. Well, I actually ALWAYS follow this group, but via digest, so by the time I see the COBOL questions someone on IBMMAIN has already given the answer that I would have given! A perfect example is the GO TO discussion where the poster did not use the NOT AT END phrase of READ. Thank you listers for being so COBOL savvy! On the topic of where your source can be for the cob2 command, it is documented in the Programming Guide. It did take me a minute to find it, but the location seems to make sense to me. In COBOL V4R2 version, see Chapter 15. Compiling under z/OS UNIX, under the heading Setting environment variables under z/OS UNIX: SYSLIB Specify paths to directories to be used in searching for COBOL copybooks if you do not specify an explicit library-name in the COPY statement. Separate multiple paths with a colon. Paths are evaluated in order from the first path to the last in the export command. If you set the variable with multiple files of the same name, the first located copy of the file is used. For COPY statements in which you have not coded an explicit library-name, the compiler searches for copybooks in this order: 1. In the current directory 2. In the paths you specify with the -I cob2 option 3. In the paths you specify in the SYSLIB environment variable It clearly says only directories and paths, but I suppose a confirmation that datasets are not supported would be nice. I will try to get that into the next version of the Programming Guide. If anyone has a better suggestion for where this confirmation would go, let me know! Cheers, TomR COBOL is the Language of the Future! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, 17 Apr 2012 13:07:18 -0600, Steve Comstock wrote: On 4/17/2012 12:53 PM, McKown, John wrote: As the OP, I thought I'm mention that I finally got the z/OS NFS server to allow me to mount a high-level onto a z/OS UNIX subdirectory on the same system. Clever and resourceful. But certainly just a workaround for the situation: you would like the compilers to be able to use MVS data sets directly for SYSLIB (and other DD names). More than just a workaround. It has the considerable collateral benefit of making legacy data sets available for processing by arbitrary USS commands. The next thing I'd like to see (but I don't do COBOL) is removal of contrived restrictions on the use of USS files and directories by applications using QSAM/BSAM/BPAM. For example I'd like to be able to use a USS directory as SYSEXEC. The only reason I can't is that Rexx makes a test (only on the first catenand) on DSORG that excludes USS directories. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On 4/17/2012 4:04 PM, Tom Ross wrote: On 4/16/2012 3:26 PM, McKown, John wrote: If so, have you figured out how to specify a PDS as an include or copybooksource? The documentation for this command basically stinks. There is an -I switch. But it apparently only accepts UNIX path specifications. The ld command (binder) accepts a PDS name via -L //pds.name. The as command (HLASM) and C compilers accept a -I //pds.name. I may end up writing my own version of the cob2 command, if I really decide that I want to compile COBOL from the UNIX shell. I'll likely model it after the as command. Yes. I discuss this command in our course Developing Applications for z/OS UNIX. But you're right, the doc is very poor. The assumption seems to be that copy books must be in HFS directories and my experiments produce error messages that would support that. In a way, that's too bad. OTOH, I, too, was assuming that copy books would reside in HFS directories so I had never tried to access copy books as members of a PDS before. You would think that if you used the classic clue that a library was a PDS/E, the compiler could figure it out. Maybe: export SYSLIB=//'SCOMSTO.U520.LIBRARY' which I tried, and the message from the compile comes back: LineID Message code Library phase message text 24 IGYLI0049-S The COPY library SCOMSTO.U520.LIBRARY/APODEFS was not found. Skipped to the period terminating the COPY statement. so it was pretty clearly expecting (nay: requiring) a z/OS UNIX file, not a PDS. I'm copying this over to ibm-main, too, as Tom Ross occasionally follows that group so maybe he will chime in with some info. Well, I actually ALWAYS follow this group, but via digest, so by the time I see the COBOL questions someone on IBMMAIN has already given the answer that I would have given! A perfect example is the GO TO discussion where the poster did not use the NOT AT END phrase of READ. Thank you listers for being so COBOL savvy! On the topic of where your source can be for the cob2 command, it is documented in the Programming Guide. It did take me a minute to find it, but the location seems to make sense to me. In COBOL V4R2 version, see Chapter 15. Compiling under z/OS UNIX, under the heading Setting environment variables under z/OS UNIX: SYSLIB Specify paths to directories to be used in searching for COBOL copybooks if you do not specify an explicit library-name in the COPY statement. Separate multiple paths with a colon. Paths are evaluated in order from the first path to the last in the export command. If you set the variable with multiple files of the same name, the first located copy of the file is used. For COPY statements in which you have not coded an explicit library-name, the compiler searches for copybooks in this order: 1. In the current directory 2. In the paths you specify with the -I cob2 option 3. In the paths you specify in the SYSLIB environment variable It clearly says only directories and paths, but I suppose a confirmation that datasets are not supported would be nice. I will try to get that into the next version of the Programming Guide. If anyone has a better suggestion for where this confirmation would go, let me know! Cheers, TomR COBOL is the Language of the Future! So, in other words, you'll make the exclusion explicit instead of removing the restriction. :-) But, I gather from some of the discussions, and your comments, on this list, you are rather tied up with other compiler priorities these days. Hope all is well. -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Tue, 17 Apr 2012 16:29:13 -0600, Steve Comstock wrote: Cheers, TomR COBOL is the Language of the Future! So, in other words, you'll make the exclusion explicit instead of removing the restriction. :-) But, I gather from some of the discussions, and your comments, on this list, you are rather tied up with other compiler priorities these days. Hope all is well. Errr... Does that compiler not already contain code to read input from legacy data sets in the case when it's invoked in batch, or is it a completely different compiler? What happens in the batch case if the copybook concatenation contains a USS directory? -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
I do use a UNIX subdirectory on my SYSEXEC concatenation. But, as you said, it cannot be first. So I have an empty PDS with FB/80 as the first DSN in the concatenation. A clumsy work around, but at least it works for me. On Tue, 2012-04-17 at 17:27 -0500, Paul Gilmartin wrote: On Tue, 17 Apr 2012 13:07:18 -0600, Steve Comstock wrote: On 4/17/2012 12:53 PM, McKown, John wrote: As the OP, I thought I'm mention that I finally got the z/OS NFS server to allow me to mount a high-level onto a z/OS UNIX subdirectory on the same system. Clever and resourceful. But certainly just a workaround for the situation: you would like the compilers to be able to use MVS data sets directly for SYSLIB (and other DD names). More than just a workaround. It has the considerable collateral benefit of making legacy data sets available for processing by arbitrary USS commands. The next thing I'd like to see (but I don't do COBOL) is removal of contrived restrictions on the use of USS files and directories by applications using QSAM/BSAM/BPAM. For example I'd like to be able to use a USS directory as SYSEXEC. The only reason I can't is that Rexx makes a test (only on the first catenand) on DSORG that excludes USS directories. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN -- John McKown Maranatha! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
Guess I am spoiled by the assembler (as command) and C/C++ compiler's support of both UNIX paths and z/OS PDS[E]s. I have a, to me, clumsy way to get around it. I started up the NFS server and exported the high level qualifier that had the PDS[E]s that I needed, then did a MOUNT onto a z/OS UNIX subdirectory. Clumsy, but it does work. I may write my own version of the cob2 command which does what _I_ want. Should be fairly easy. Just set up the allocations as needed using DYNALLOC, then use BPX1ATM UNIX routine to ATTACH the IGYCRCTL program. The main problem that I have with my UNIX programs is that I can only use HLASM (no C compiler license). And I have not figured out how to handle UNIX signals. So the cntlc to abort a command doesn't work. On Tue, 2012-04-17 at 15:04 -0700, Tom Ross wrote: On 4/16/2012 3:26 PM, McKown, John wrote: snip It clearly says only directories and paths, but I suppose a confirmation that datasets are not supported would be nice. I will try to get that into the next version of the Programming Guide. If anyone has a better suggestion for where this confirmation would go, let me know! Cheers, TomR COBOL is the Language of the Future! -- John McKown Maranatha! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On 4/16/2012 3:26 PM, McKown, John wrote: If so, have you figured out how to specify a PDS as an include or copybooksource? The documentation for this command basically stinks. There is an -I switch. But it apparently only accepts UNIX path specifications. The ld command (binder) accepts a PDS name via -L //pds.name. The as command (HLASM) and C compilers accept a -I //pds.name. I may end up writing my own version of the cob2 command, if I really decide that I want to compile COBOL from the UNIX shell. I'll likely model it after the as command. Yes. I discuss this command in our course Developing Applications for z/OS UNIX. But you're right, the doc is very poor. The assumption seems to be that copy books must be in HFS directories and my experiments produce error messages that would support that. In a way, that's too bad. OTOH, I, too, was assuming that copy books would reside in HFS directories so I had never tried to access copy books as members of a PDS before. You would think that if you used the classic clue that a library was a PDS/E, the compiler could figure it out. Maybe: export SYSLIB=//'SCOMSTO.U520.LIBRARY' which I tried, and the message from the compile comes back: LineID Message code Library phase message text 24 IGYLI0049-S The COPY library SCOMSTO.U520.LIBRARY/APODEFS was not found. Skipped to the period terminating the COPY statement. so it was pretty clearly expecting (nay: requiring) a z/OS UNIX file, not a PDS. I'm copying this over to ibm-main, too, as Tom Ross occasionally follows that group so maybe he will chime in with some info. John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For MVS-OE subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO MVS-OE -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
Has anyone tried (apart from Paul G) exporting the PDSE via NFS and mounting it at a z/Unix mountpoint on the same system ? That should be able to provide your path as well a classic access Jerry Whitteridge Lead Systems Programmer Safeway Inc. 925 951 4184 If you feel in control you just aren't going fast enough. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Steve Comstock Sent: Monday, April 16, 2012 3:21 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Anybody use the cob2 command on a UNIX shell to compile COBOL? On 4/16/2012 3:26 PM, McKown, John wrote: If so, have you figured out how to specify a PDS as an include or copybooksource? The documentation for this command basically stinks. There is an -I switch. But it apparently only accepts UNIX path specifications. The ld command (binder) accepts a PDS name via -L //pds.name. The as command (HLASM) and C compilers accept a -I //pds.name. I may end up writing my own version of the cob2 command, if I really decide that I want to compile COBOL from the UNIX shell. I'll likely model it after the as command. Yes. I discuss this command in our course Developing Applications for z/OS UNIX. But you're right, the doc is very poor. The assumption seems to be that copy books must be in HFS directories and my experiments produce error messages that would support that. In a way, that's too bad. OTOH, I, too, was assuming that copy books would reside in HFS directories so I had never tried to access copy books as members of a PDS before. You would think that if you used the classic clue that a library was a PDS/E, the compiler could figure it out. Maybe: export SYSLIB=//'SCOMSTO.U520.LIBRARY' which I tried, and the message from the compile comes back: LineID Message code Library phase message text 24 IGYLI0049-S The COPY library SCOMSTO.U520.LIBRARY/APODEFS was not found. Skipped to the period terminating the COPY statement. so it was pretty clearly expecting (nay: requiring) a z/OS UNIX file, not a PDS. I'm copying this over to ibm-main, too, as Tom Ross occasionally follows that group so maybe he will chime in with some info. John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -- For MVS-OE subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO MVS-OE -- Kind regards, -Steve Comstock The Trainer's Friend, Inc. 303-355-2752 http://www.trainersfriend.com * To get a good Return on your Investment, first make an investment! + Training your people is an excellent investment * Try our tool for calculating your Return On Investment for training dollars at http://www.trainersfriend.com/ROI/roi.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN Email Firewall made the following annotations. -- Warning: All e-mail sent to this address will be received by the corporate e-mail system, and is subject to archival and review by someone other than the recipient. This e-mail may contain proprietary information and is intended only for the use of the intended recipient(s). If the reader of this message is not the intended recipient(s), you are notified that you have received this message in error and that any review, dissemination, distribution or copying of this message is strictly prohibited. If you have received this message in error, please notify the sender immediately. == -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Anybody use the cob2 command on a UNIX shell to compile COBOL?
On Mon, 16 Apr 2012 17:29:23 -0600, Jerry Whitteridge wrote: Has anyone tried (apart from Paul G) exporting the PDSE via NFS and mounting it at a z/Unix mountpoint on the same system ? That should be able to provide your path as well a classic access We tried and failed. But we didn't try very hard because I couldn't justify expending much systems programmer resource on my largely experimental need. We have legacy data sets exported and mounted on Solaris mountpoints. We have Solaris filesystems exported and mounted on z/OS. If I need something in both places, I keep it on a Solaris server. John M. may have tried; I don't know with what degree of success. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
