ISPF macro - Extracting symbolic variable and its value

chidams78 · New User Joined: 29 May 2006 Posts: 59 Location: India

Hi all,

I want to extract the symbolic variable and its value in 2 separate REXX variables.

In the below mentioned PROC, I want to extract the symbolic parm CONDB in one variable and the '0,NE in the other variable.

//TST0121 PROC CONDB='0,NE',CONDQ='0,NE',CONDR='0,NE',CONDT='0,NE',
// TAPERET=',TRTCH=COMP',

Note: I had handled the PROC statement and stored the line after PROC in the THPARMS variable
PARSE VAR ALINE 'PROC' THEPARMS

Below code gets the symbolic parm and its value in 2 diff variables - PARM1 and PARM2

enrico-sorichetti · Posted: Wed Sep 10, 2008 2:13 am

here is a pointer to a working jcl analyzer/reformatter I wrote

the I/O model is unix/linux oriented ( linein/lineout) but it works
it should take very little to port it to the EXECIO/IRSREDIT model

http://ibmmainframes.com/viewtopic.php?t=32494&highlight=linein

chidams78 · New User Joined: 29 May 2006 Posts: 59 Location: India

Thanks Enrico...

But If i can get the help in solving it by means of ISREDIT macro instead of exporting the JCL parser, it will be great.

Thanks
Chidam

enrico-sorichetti · Posted: Wed Sep 10, 2008 3:41 am

chidams78 · New User Joined: 29 May 2006 Posts: 59 Location: India

Thatz right...

But If anybody can help me in modifying this code to handle the above concern, that will be helpful

Pedro · Posted: Wed Sep 10, 2008 9:49 pm

When I find constants that look like command syntax (or delimiters in this case), I temporarily change them to something else.

Your example:

chidams78 · New User Joined: 29 May 2006 Posts: 59 Location: India

Pedro,

I agree with the logic that you had mentioned but the scenario here is:

I need to extract each parm and its value from PROC in two diferent REXX variable by reading each PROC line and then search the entire PROCLIB till null character to find out whether that particular symbolic parm is used before the NULL character or not. If it is used after NULL charcter, blank it out using ISPF edit macro. I am able to do the search process but finds a bug in the initial PARM extract process.

For the below PROC, we need to extract each symbolic parm (Eg: CONDB) in one REXX variable PARM1 and the corresponding value (0,NE) in second REXX variable PARM2.

//TST0542 PROC CONDB='0,NE',CONDQ='0,NE',CONDR='0,NE',CONDT='0,NE',
// CONDV='0,NE',CONDZ='0,NE',COND4='0,NE',COND5='0,NE',
// TAPERET=',TRTCH=COMP',
// GDG='GDG,',

The code is
PARSE VAR THEPARMS PARM1 "=" PARM2 " ', " THEPARMS

The problem with this code is that it won't extract correctly when it comes to the variable TAPERET since the above code looks for = first and store it in PARM1 and thatz fine. But after that the code looks for ', and store only spaces in PARM2 which is wrong (Even if convert that to ~, the same thing happens). The whole value ',TRTCH=COMP' should get stored in PARM2.

Note: We need this code too as it extracts the parms correctly except for TAPERET.

I am not finding any clue how to handle this. Can you plz help me in that.

dick scherrer · Posted: Thu Sep 11, 2008 1:49 am

Hello,

Pedro · Posted: Thu Sep 11, 2008 2:07 am

Start with:

Pedro · Posted: Thu Sep 11, 2008 2:12 am

Sorry, I did not see earlier:

chidams78 · New User Joined: 29 May 2006 Posts: 59 Location: India

Hi Dsch

Sorry If i am saying something wrong. I had tried the code which was put forth. After that only I replied becaz it didn't satisfy my reqt.

For the below mentioned PROC

//TST0542 PROC CONDB='0,NE',CONDQ='0,NE',CONDR='0,NE',CONDT='0,NE',
// CONDV='0,NE',CONDZ='0,NE',COND4='0,NE',COND5='0,NE',
// TAPERET=',TRTCH=COMP',
// GDG='GDG,'

the code offered is

"ISREDIT C '7E7D6B'X '7E7DA1'X "

So if I change my parm extraction code to
PARSE VAR THEPARMS PARM1 " ='~ " PARM2 " ', " THEPARMS

This code will handle my issue with TAPERET=',TRTCH=COMP',
but again I will land up in trouble when handlign some other PROC line like
// CONDV='0,NE',CONDZ='0,NE',COND4='0,NE',COND5='0,NE',

Here itz not the special char which puts up the issue but the way in which the parm was extracted.

PARSE VAR THEPARMS PARM1 "=" PARM2 " ', " THEPARMS
==> handles PROC lines 1 ,2 and 4

"ISREDIT C '7E7D6B'X '7E7DA1'X ALL"
PARSE VAR THEPARMS PARM1 " ='~ " PARM2 " '~ " THEPARMS
"ISREDIT C '7E7DA1'X '7E7D6B'X ALL"
==> Handles PROC line 3

I was just looking for a logic which handles both the scenario

Pedro · Posted: Thu Sep 11, 2008 2:21 am

We appended at almost the same time... please read my earlier append

===
Sorry if I was not clear. Sort of a repeat:

When I find data that looks like command syntax (or delimiters in this case), I temporarily change the data to something else.

Your example:

dick scherrer · Posted: Thu Sep 11, 2008 2:48 am

Hello,

chidams78 · New User Joined: 29 May 2006 Posts: 59 Location: India

Pedro,

Earlier, I was not able to understand the logic that you had given.
I was of the impression of changing the parm extraction step instead of handling the special character.

Now I had implemented it and it works fine...:-)

Sorry for the confusion.

Thanks all.
Chidam