|
View previous topic :: View next topic
|
| Author |
Message |
nareshkp
New User
.jpg)
Joined: 26 Sep 2007
Posts: 28
Location: Bangalore
|
|
|
|
Hi,
I have a physical sequential file (FB, 80B length) which contains a set of mnemonics. Each mnemonic can be of 3 to 8 characters length. In other words, each record in the PS can be of 3 to 8 character length.
Now, I want to search a library (PDS) for each mnemonic of the input file to get the list of PDS members that uses it. The output should be the list of PDS members that gets a hit against each mnemonic separately.
Or else is there a way to get all the lines in PDS library (do not necessarily need the members) that contains any of the mnemonics in Input file?
Thanks,
Naresh. |
|
| Back to top |
|
 |
superk
Global Moderator
Joined: 26 Apr 2004
Posts: 4652
Location: Raleigh, NC, USA
|
|
|
|
Well, I know enough REXX to be dangerous, but I have no idea what you mean by the term "mnemonic", and you haven't provided any details as to how or why using REXX comes into play here, rather than the more obvious option of using SuperC. It'd really be nice if you'd provide some details on the steps and/or processess that you want to do, and then give some details on what part or parts you need help on.
There are 100's of topics here that deal with PDS's (listing members, searching members, changing members, using EDIT, using IPSF EDIT macros, using ISPF Library Management (LM) services, using SuperC, using DFSORT/Syncsort for substring searches and compares, etc.) so you should have a plethora of previous topics or similar circumstances to look at. |
|
| Back to top |
|
|
Pedro
Global Moderator
Joined: 01 Sep 2006
Posts: 2547
Location: Silicon Valley
|
|
|
|
ISPF option 3.15, is search- extended, will produce a job with multiple search arguments, like this:
| Code: |
//SEARCH EXEC PGM=ISRSUPC,
// PARM=(SRCHCMP,
// 'FINDALL')
//NEWDD DD DSN=PEDRO.MISC.SOURCE,
// DISP=SHR
//OUTDD DD SYSOUT=(H)
//SYSIN DD *
SRCHFOR 'TEST123'
SRCHFOR 'TEST456' |
Just do the same thing from rexx by creating the SYSIN file and calling program ISRSUPC. |
|
| Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006
Posts: 6966
Location: porcelain throne
|
|
|
|
| Quote: |
| I have a physical sequential file (FB, 80B length) which contains a set of mnemonics. Each mnemonic can be of 3 to 8 characters length. In other words, each record in the PS can be of 3 to 8 character length. |
you either have a fixed block, fixed length record or you have a variable length record.
what is 80B is that hex for 2059 or are you using your own terms, jargon and notation to discuss a rather well documented subject: IBM Mainframe
the thread is based on very poor definition, total misuse of terms and activities,
and as kevin pointed out,
total lack of interest on the part of the TS to do a little investigation on what is a rather common activity on a computer. |
|
| Back to top |
|
|
nareshkp
New User
Joined: 26 Sep 2007
Posts: 28
Location: Bangalore
|
|
|
|
Thanks all for your responses.
| Quote: |
| but I have no idea what you mean by the term "mnemonic", and you haven't provided any details as to how or why using REXX |
Apologies for providing the in-sufficient details. Mnemonics are the alphanumeric assembler instructions. The input PS contains thousands of assembler statements. The requirement was that I had to find out the number of USERMODs (members of PDS) that are using any of these assembler statements.
| Quote: |
| what is 80B is that hex for 2059 or are you using your own terms |
I meant LRECL=80. Sorry for the confusion.
I know that ISPF option 3.15 can be used for multiple search arguments. But I was thinking, for thousands (around 15500) of search arguments it will not be a feasible solution. So I have made my post here.
Now, my problem has been resolved. One of my colleague helped me in writing a REXX to get the requirement done.
Thanks for all for your advises.
Naresh. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
| Back to top |
|
|
Pedro
Global Moderator
Joined: 01 Sep 2006
Posts: 2547
Location: Silicon Valley
|
|
|
|
| Quote: |
| I was thinking, for thousands (around 15500) of search arguments it will not be a feasible solution |
Agree.
But I think you are getting into a gray area. Things that work in a small scale may not work well at a larger scale. I am not sure rexx is the right choice to search for so many things. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hi Pedro,
It is if you get paid for cpu usage - as this will exponentially increase the cpu needed 
d |
|
| Back to top |
|
|
nareshkp
New User
Joined: 26 Sep 2007
Posts: 28
Location: Bangalore
|
|
|
|
Hi Pedro & Dick,
I did not use the rexx for searching 15500 arguments. I just tested the rexx with 3 arguments and finds it working and kept aside for the implementation. Later I had to filter the arguments as per the new specifications which reduces the list to 185 arguments. And this time I used ISRSUPC for searching the library because it gives the output with HITs being highlited. I did not think about the CPU usage, but it seems unknowingly I saved some CPU usage by not using the rexx  .
| Quote: |
| have the experts in Your organization ever heard of ? |
May be they might have run these reports... but I do not know. My job is is to make sure that the 185 instructions are not being used as user defined variables. The instructions are newly introduced but the USERMODs are old.
Thanks. |
|
| Back to top |
|
|
|
|
