Article View: comp.lang.cobol
Article #50Re: STRING, UNstring & INSPECT are not recommended
From: Kerry Ellis
Date: Fri, 30 Jun 2000 00:00
Date: Fri, 30 Jun 2000 00:00
219 lines
9359 bytes
9359 bytes
--------------72A782B134CB0BD7E7AF84A7 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sorry, I wouldn't have the foggiest idea. I think that the last time I worked on an IBM DOS system was in 1971. Anyway, I doubt that the "IBM Guy" is anywhere near being correct. See: 4.8.1.6 "COBOL/VSE Programming Guide" via IBM BookManager BookServer which appears to have become applicable as long ago as 1983. Even DOS/VSE has emerged into the 20th century (even if everybody else is in the 21st). Kerry Patrick Lee wrote: > Kerry, > Thanks for your detail explanation. But IBM guy told me the issue > applies to COBOL/VSE even through it state in the COBOL manual that it > only applies to DOS/COBOL. Is that true?? > > PL. > In article <3959284F.3C8D8027@msdw.com>, > Kerry Ellis <Kerry.Ellis@MSDW.com> wrote: > > OS/VS Cobol was oblivious to the CICS environment. The Inspect, > > String and Unstring statements invoked routines that issued OS (not > > CICS) Getmains. Because OS/VS Cobol programs had to be term- > > inated with an EXEC CICS RETURN, the Cobol runtime routines > > were never given a chance to clean up; eventually, the CICS region > > would run out of what used to be called "OSCOR" and incur an OS > > abend such as S80A. Nowadays, with Cobol II and the "new" LE > > conforming compilers, all run time storage acquisition requests are > > handled by environment specific library routines, so Inspect, String > > and Unstring are given CICS DSA storage that is not different than > > any other piece of transaction storage. In addition, such programs > > can be ended with a STOP RUN or an EXEC CICS RETURN; > > both will go through normal program/run unit cleanup; however, > > any task duration storage that is missed will be freed by task term- > > ination. (BTW, the restrictions against String, Unstring and Inspect > > were dropped IIRC with the arrival of the Cobol II compiler which > > was =E6ons ago.) > > > > Kerry > > > > "William M. Klein" wrote: > > > > > I believe that you are correct. I tried to find this documented in > the= > > COBOL > > > manuals, but couldn't (easily). HOWEVER, I can find the list of > what "= > > is" > > > restricted under the current COBOL with CICS - and it doesn't > include t= > > hose > > > verbs. Check out: > > > > > > http://www.s390.ibm.com/bookmgr-cgi/bookmgr.cmd/BOOKS/IGYPG204/4% > 2e10%2= > > e1%2e6 > > > > > > -- > > > Bill Klein > > > wmklein <at> ix dot netcom dot com > > > "Clark F. Morris, Jr." <cfmtech@istar.ca> wrote in message > > > news:39587336.17BE2312@istar.ca... > > > > I am forwarding this to comp.lang.cobol to verify my impression > that = > > VS > > > > COBOL II, COBOL for MVS and VM and COBOL for MVS and VM have no > > > > restrictions on the use of STRING, UNSTRING and INSPECT. All of > thes= > > e > > > > are 31 bit compilers. > > > > > > > > Clark Morris, CFM Technical Programming Services Inc., > cfmtech@istar.= > > ca > > > > > > > > -------- Original Message -------- > > > > Subject: STRING, UNstring & INSPECT are not recommended > > > > Date: Tue, 27 Jun 2000 09:14:00 GMT > > > > From: Patrick Lee <patricklee@whbhk.com> > > > > Organization: Deja.com - Before you buy. > > > > Newsgroups: bit.listserv.cics-l > > > > > > > > Dear all, > > > > According to IBM APAR II06138, we should not use STRING, UNSTRING, > > > > INSPECT because they will eat up our 24bit Getvis area. Any > > > > recommendation of alternatives for these 3 words???? > > > > > > > > Thanks. > > > > > > > > Patrick. > > > > > > > > > > > > Sent via Deja.com http://www.deja.com/ > > > > Before you buy. > > > > > > Sent via Deja.com http://www.deja.com/ > Before you buy. --------------72A782B134CB0BD7E7AF84A7 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html> Sorry, I wouldn't have the foggiest idea. I think that the <br>last time I worked on an IBM DOS system was in 1971. <br>Anyway, I doubt that the "IBM Guy" is anywhere near <br>being correct. See: <p> <a href="http://www.s390.ibm.com:80/bookmgr-cgi/bookmgr.cmd/BOOKS/IGYVA001/4%2e8%2e1%2e6?ACTION=MATCHES&REQUEST=cics&TYPE=FUZZY&SHELF=IGYVS001&searchTopic=TOPIC&searchText=TEXT&searchIndex=INDEX&rank=RANK&ScrollTOP=FIRSTHIT#FIRSTHIT">4.8.1.6 "COBOL/VSE Programming Guide" via IBM BookManager BookServer</a> <p>which appears to have become applicable as long ago <br>as 1983. Even DOS/VSE has emerged into the 20th <br>century (even if everybody else is in the 21st). <p>Kerry <p>Patrick Lee wrote: <blockquote TYPE=CITE>Kerry, <br>Thanks for your detail explanation. But IBM guy told me the issue <br>applies to COBOL/VSE even through it state in the COBOL manual that it <br>only applies to DOS/COBOL. Is that true?? <p>PL. <br>In article <3959284F.3C8D8027@msdw.com>, <br> Kerry Ellis <Kerry.Ellis@MSDW.com> wrote: <br>> OS/VS Cobol was oblivious to the CICS environment. The Inspect, <br>> String and Unstring statements invoked routines that issued OS (not <br>> CICS) Getmains. Because OS/VS Cobol programs had to be term- <br>> inated with an EXEC CICS RETURN, the Cobol runtime routines <br>> were never given a chance to clean up; eventually, the CICS region <br>> would run out of what used to be called "OSCOR" and incur an OS <br>> abend such as S80A. Nowadays, with Cobol II and the "new" LE <br>> conforming compilers, all run time storage acquisition requests are <br>> handled by environment specific library routines, so Inspect, String <br>> and Unstring are given CICS DSA storage that is not different than <br>> any other piece of transaction storage. In addition, such programs <br>> can be ended with a STOP RUN or an EXEC CICS RETURN; <br>> both will go through normal program/run unit cleanup; however, <br>> any task duration storage that is missed will be freed by task term- <br>> ination. (BTW, the restrictions against String, Unstring and Inspect <br>> were dropped IIRC with the arrival of the Cobol II compiler which <br>> was =E6ons ago.) <br>> <br>> Kerry <br>> <br>> "William M. Klein" wrote: <br>> <br>> > I believe that you are correct. I tried to find this documented in <br>the= <br>> COBOL <br>> > manuals, but couldn't (easily). HOWEVER, I can find the list of <br>what "= <br>> is" <br>> > restricted under the current COBOL with CICS - and it doesn't <br>include t= <br>> hose <br>> > verbs. Check out: <br>> > <br>> > <a href="http://www.s390.ibm.com/bookmgr-cgi/bookmgr.cmd/BOOKS/IGYPG204/4%">http://www.s390.ibm.com/bookmgr-cgi/bookmgr.cmd/BOOKS/IGYPG204/4%</a> <br>2e10%2= <br>> e1%2e6 <br>> > <br>> > -- <br>> > Bill Klein <br>> > wmklein <at> ix dot netcom dot com <br>> > "Clark F. Morris, Jr." <cfmtech@istar.ca> wrote in message <br>> > <a href="news:39587336.17BE2312@istar.ca">news:39587336.17BE2312@istar.ca</a>... <br>> > > I am forwarding this to comp.lang.cobol to verify my impression <br>that = <br>> VS <br>> > > COBOL II, COBOL for MVS and VM and COBOL for MVS and VM have no <br>> > > restrictions on the use of STRING, UNSTRING and INSPECT. All of <br>thes= <br>> e <br>> > > are 31 bit compilers. <br>> > > <br>> > > Clark Morris, CFM Technical Programming Services Inc., <br>cfmtech@istar.= <br>> ca <br>> > > <br>> > > -------- Original Message -------- <br>> > > Subject: STRING, UNstring & INSPECT are not recommended <br>> > > Date: Tue, 27 Jun 2000 09:14:00 GMT <br>> > > From: Patrick Lee <patricklee@whbhk.com> <br>> > > Organization: Deja.com - Before you buy. <br>> > > Newsgroups: bit.listserv.cics-l <br>> > > <br>> > > Dear all, <br>> > > According to IBM APAR II06138, we should not use STRING, UNSTRING, <br>> > > INSPECT because they will eat up our 24bit Getvis area. Any <br>> > > recommendation of alternatives for these 3 words???? <br>> > > <br>> > > Thanks. <br>> > > <br>> > > Patrick. <br>> > > <br>> > > <br>> > > Sent via Deja.com <a href="http://www.deja.com/">http://www.deja.com/</a> <br>> > > Before you buy. <br>> <br>> <p>Sent via Deja.com <a href="http://www.deja.com/">http://www.deja.com/</a> <br>Before you buy.</blockquote> </html> --------------72A782B134CB0BD7E7AF84A7--
Technical Details
Message-ID:
<395D0385.76ADF25D@msdw.com>Path:
arc1-ca.pugleaf.net!unknown!not-for-mailReferences:
<39587336.17BE2312@istar.ca> <8jalpm$d5t$1@slb7.atl.mindspring.net> <3959284F.3C8D8027@msdw.com> <8jbq6r$eif$1@nnrp1.deja.com>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 10f713,f86dda0c5a889e71 X-Google-Attributes: gid10f713,public X-Google-Thread: 10d15b,f86dda0c5a889e71 X-Google-Attributes: gid10d15b,public From: Kerry Ellis <Kerry.Ellis@msdw.com> Subject: Re: STRING, UNstring & INSPECT are not recommended Date: 2000/06/30 Message-ID: <395D0385.76ADF25D@msdw.com> X-Deja-AN: 640989709 References: <39587336.17BE2312@istar.ca> <8jalpm$d5t$1@slb7.atl.mindspring.net> <3959284F.3C8D8027@msdw.com> <8jbq6r$eif$1@nnrp1.deja.com> To: Patrick Lee <patricklee@whbhk.com> X-Accept-Language: en,ja Content-Type: multipart/alternative; boundary="------------72A782B134CB0BD7E7AF84A7" Organization: Morgan Stanley Dean Witter & Co. Mime-Version: 1.0 Reply-To: Kerry Ellis <Kerry.Ellis@MSDW.com> Newsgroups: comp.lang.cobol,bit.listserv.cics-l