|
View previous topic :: View next topic
|
| Author |
Message |
ratnakumar
New User
Joined: 07 May 2009
Posts: 30
Location: Banglore
|
|
|
|
I tested with SYNCSORT, And it failed with A data exception, S0C7.. Pasted below is the error message in ABENDAID
| Code: |
* Error Location *
*******************************************
The next sequential instruction to be executed in load module SS13X02C
was at displacement 0000218A from the load point of the module.
A B E N D - A I D PAGE 2
The module was dynamically loaded.
SS13X02C was loaded from the extended fixed link pack area (EFLPA).
The module is 00002568 bytes long.
*******************************************
* Call Trace Summary *
*******************************************
*******Calling******* ****Return**** *******Called********
Load-Mod Program Type Value Program Load-Mod
*SYSTEM Links to QKQMAIN SYNCSORT
The program causing the error is not on the calling
chain.
One or more modules loaded from the LINKLIST.
Current LINKLIST library set is LNKLSTXX
|
Please let me know if you need any other details. |
|
| Back to top |
|
 |
ratnakumar
New User
Joined: 07 May 2009
Posts: 30
Location: Banglore
|
|
|
|
Just an FYI..
The SYNCSORT code (which worked for me earlier) had failed with SOC-0C7, with the instream data Z000 (i.e without F or D at position 6) |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10900
Location: italy
|
|
|
|
so it seems that DFSORT is more forgiving than SYNCSORT
in this case probably the proper snippet would be
| Code: |
****** ***************************** Top of Data ******************************
000001 //ENRICO1 JOB NOTIFY=&SYSUID,
000002 // MSGLEVEL=(1,1),CLASS=A,MSGCLASS=H
000003 //*
000004 //ICE EXEC PGM=SORT
000005 //SYSOUT DD SYSOUT=*
000006 //SYSPRINT DD SYSOUT=*
000007 //SORTIN DD *
000008 A1234F
000009 A1234D
000010 A2345D
000011 A3456F
000012 Z0000
000013 //SORTOUT DD SYSOUT=*
000014 //SYSIN DD *
000015 INREC IFTHEN=(WHEN=(6,1,CH,EQ,C'F'),OVERLAY=(81:C'10')),
000016 IFTHEN=(WHEN=(6,1,CH,EQ,C'D'),OVERLAY=(81:C'01')),
000017 IFTHEN=(WHEN=NONE,OVERLAY=(81:C'00'))
000018 SORT FIELDS=(1,5,CH,A),EQUALS
000019 OUTFILE REMOVECC,NODETAIL,BUILD=(80X),
000020 HEADER1=('CM#',3X,'FOREIGN COUNT',2X,'DOMESTIC COUNT'),
000021 SECTIONS=(1,5,
000022 TRAILER3=(1,5,6X,TOT=(81,1,ZD,M10,LENGTH=8),
000023 8X,TOT=(82,1,ZD,M10,LENGTH=8)))
000024 //
****** **************************** Bottom of Data **************************** |
everything initialized and all types properly defined |
|
| Back to top |
|
|
sqlcode1
Active Member
Joined: 08 Apr 2010
Posts: 577
Location: USA
|
|
|
|
| Quote: |
so it seems that DFSORT is more forgiving than SYNCSORT
in this case probably the proper snippet would be |
For ZD in this scenario,DFSort ignores/doesn't consider zone thus 40 becomes 0.
Thanks, |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10900
Location: italy
|
|
|
|
yes... for the existing fields ...
but then under the covers DFSORT considers undefined fields as x'00'
like when for the "F" case the overlay is on position 81..
but then DFSORT happily sums on position 82
and if position 6 i neither F or D .... nothing gets overlaid, but DFSORT is happy
puzzling  |
|
| Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA
|
|
|
|
Enrico,
I haven't really been following this discussion (and I don't really want to take the time to read through everything).
Can you please show me a good (simple) example of what you're questioning in DFSORT? What does the input look like in hex? What is the DFSORT statement in question? What do you think is puzzling? The simpler you can make it, the better. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10900
Location: italy
|
|
|
|
Hi Frank!
I am not really questioning DFSORT, I am just trying to understand !
do You want to take the idea exchange offline ? |
|
| Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA
|
|
|
|
Enrico,
Yes, I know that you're trying to understand and I'd like to help you do that. I just need to know the details of what we're discussing so I can respond accurately. I'm happy to discuss this here or offline as you prefer. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
Hello,
Would it make sense to split this follow-on to another topic? Maybe in the Moderator part of the forum, or somehow "police" it so it does not become a free-for-all?
Personally, i'd like to be able to "watch" the proceedings 
d |
|
| Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA
|
|
|
|
Enrico sent me a note offline explaining what he is asking and I think I understand it to be:
Why does DFSORT allow a blank input value to be TOTALed as if it were a ZD zero?. For example, if we run this job with DFSORT:
| Code: |
//S1 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD *
01
02
//SORTOUT DD SYSOUT=*
//SYSIN DD *
OPTION COPY
OUTFIL TRAILER1=(TOT=(1,2,ZD,TO=ZD,LENGTH=2))
|
why does it give a valid total of 03 instead of an S0C7?
The answer is that DFSORT treats a ZD value like so:
zdzd...zdsd
d must be a valid digit (0-9).
z is the zone and is ignored. So X'40' for zd is treated as a decimal 0.
s is the sign. If s is F, E, C, A, 8, 6, 4, 2 or 0, it is treated as plus. If s is D, B, 9, 7, 5, 3 or 1, it is treated as minus. So X'40' for sd is treated as +0.
Therefore, X'4040' is treated as a ZD +0.
Note that this is all documented in "z/OS DFSORT Application Programming Guide".
Does that help? |
|
| Back to top |
|
|
dbzTHEdinosauer
Global Moderator
Joined: 20 Oct 2006
Posts: 6965
Location: porcelain throne
|
|
|
|
also, why should IBM's DFSORT
manipulate display numeric any different than IBM COBOL? |
|
| Back to top |
|
|
|
|
  |
All times are GMT + 6 Hours |
|
