MVSFORUMS.com

ravikumar_sri2001

Experts,

I have a list of dataset names in a file. From that list I need to extract dataset names ending with R001,R001,R002 and so on till R999. The length of the dataset names also not fixed.

Thanks in advance for your help!

Thanks,
Ravikumar.

papadi · Supermod Joined: 20 Oct 2009 Posts: 594 Topics: 1

What do you want to do when you have found them?

You can identify them using your sort product, or rexx and probably one of several other utilities (depending on what is available on your system).

Depending on the reason for finding them, one tool or another may be better suited.
_________________
All the best,

di

papadi · Supermod Joined: 20 Oct 2009 Posts: 594 Topics: 1

You can also use the FINDALL option in SUPERC.
_________________
All the best,

di

ravikumar_sri2001 · Posted: Sun Aug 21, 2011 2:21 am Post subject:

Thanks for your reply! This can be done thru REXX. But can this be done thru SORT ?

papadi · Supermod Joined: 20 Oct 2009 Posts: 594 Topics: 1

Did you not read my first reply - yes, it can be done with your sort product (assuming your sort is fairly current).

Suggest you look at superc/findall . . .

And you still did not post how this list is to be used. . .
_________________
All the best,

di

dbzTHEdinosauer · Posted: Sun Aug 21, 2011 7:27 am Post subject:

before a sort solution can be provided,
one needs to know what is the record layout of your file.

are the file names delimited by anything? Space?
what are the column positions containing the filename?
a detailed description of you input and output is necessary to provide a solution.

if all you want is a general solution and you do not seem fit to provide any detailed information,
feel free to either look in the dfsort/syncsort/(whatever your sort product is) manuals.
depending on the attributes of the record definition,
PARSE, OMIT, INCLUDE can/could be used.
Also, feel free to look thru the several thousand posts that exist in this forum for a general solution that fits your general needs.
_________________
Dick Brenholtz
American living in Varel, Germany

ravikumar_sri2001 · Posted: Sun Aug 21, 2011 9:24 am Post subject:

Experts,

The LRECL of the input dataset which has file names is 44 bytes.

Sample input:
XXX.XXX.DSN1
XXX.YYY.SDSFD.DSN2
AXX.BXXX.CXXX.R001
AXX.BXXX.CXXX.R002
AXX.BXXX.CXXX.R003
AXX.BXXX.CXXX.R004
AXX.BXXX.CXXX.R005
ASDF.XXXX.CC
SDSD.YYY.VVV
FFFF.GGG.R001
FFFF.GGG.R002
FFFF.GGG.R003
FFFF.GGG.R004
FFFF.GGG.R005
.
.
.
FFFF.GGG.R999

Sample OUTPUT:

AXX.BXXX.CXXX.R001
AXX.BXXX.CXXX.R002
AXX.BXXX.CXXX.R003
AXX.BXXX.CXXX.R004
AXX.BXXX.CXXX.R005
FFFF.GGG.R001
FFFF.GGG.R002
FFFF.GGG.R003
FFFF.GGG.R004
FFFF.GGG.R005
.
..
FFFF.GGG.R999

Thanks,
Ravikumar.

papadi · Supermod Joined: 20 Oct 2009 Posts: 594 Topics: 1

Read about INCLUDE or OMIT in your sort documentation. Also look at PARSE and SS (substring).
_________________
All the best,

di

dbzTHEdinosauer · Posted: Sun Aug 21, 2011 4:06 pm Post subject:

keep in mind since the ds record length is only 44,
you will need to SS up to but not including the 44th column with
variations of '.Rnnnb' (where n is numeric and b is space.)
the the 40th thru 44th must be '.Rnnn'
_________________
Dick Brenholtz
American living in Varel, Germany

ravikumar_sri2001 · Posted: Sun Aug 21, 2011 9:52 pm Post subject:

Experts,

I can give INCLUDE COND=(1,44,SS,EQ,C'R001,R002,R003.....R999') and it will be a lengthy one. Is there any other way ?

papadi · Supermod Joined: 20 Oct 2009 Posts: 594 Topics: 1

If you want to select a series of values (rather than a value within a range) you will need to code all of the individual values. . .
_________________
All the best,

di

Sqlcode · Intermediate Joined: 15 Dec 2006 Posts: 157 Topics: 38

ravikumar_sri2001,
See if below helps...

dbzTHEdinosauer · Posted: Mon Aug 22, 2011 8:53 am Post subject:

sqlcode,
clever use of the SQZ function. claps

_________________
Dick Brenholtz
American living in Varel, Germany

ravikumar_sri2001 · Posted: Mon Aug 22, 2011 2:47 pm Post subject:

SQLCODE,

Excellent solution!!...Thank you so much!!!

Ravikumar.

Frank Yaeger · Posted: Mon Aug 22, 2011 2:58 pm Post subject:

Yes, good solution. But here's a slightly simpler version: