Author Topic: NODX Application (V8.4 on z/OS)  (Read 2926 times)

Gausselmann

  • Guest
NODX Application (V8.4 on z/OS)
« on: July 07, 2011, 01:05:00 AM »
Hi all,

I'm tryind to define an Application with Indexstyle = NODX.
I've a linedata list, which should be archived without indexing (because there aren't any ;-) and without segmentation, just the whole list.

We are working on z/OS with OnDemand Version 8.4

I define the indexerinformation in the application just as it's said in the IBM Indexing Reference.

But whatever I try, I always get as Result:
arsload: Processing file >/opt/prv/ars/tmp/16777470 (DD:IN-TEST.LR.LRD18690.ONDEMAND.PRAX)<

arsload: 07/07/11 07:23:33 -- Loading started, --UNKNOWN-- bytes to process
The field name 'PAGE_NUMBER' found in the index file does not match any expected fields. The field will be ignored.
Row 1: The value ' ' cannot be converted to a valid integer
ArcLoad_Field rc = 1 field = PAGE_NUMBER_END, value =
Loaded 0 rows into the database
arsload: 07/07/11 07:23:34 Loading failed
arsload: Processing failed for file >/opt/prv/ars/tmp/16777470 (DD:IN-TEST.LR.LRD18690.ONDEMAND.PRAX)<


My definitions in the Indexerinformation look like this:
CC=YES
CCTYPE=A
FILEFORMAT=RECORD,350
GROUPMAXPAGES=999
TRIGGER1=*,1,X'F1',(TYPE=GROUP) /* 1 */
FIELD1=2,13,10,(TRIGGER=1
,BASE=0)
INDEX1=X'E2C5C7D4C5D5E36DD5E4D4C2C5D9',FIELD1,(TYPE=GROUP,BREAK=NO) /* SEGMENT_NUMBER */
INDEX2=X'D9C5D7D6D9E36DC4C1E3C5',FIELD1,(TYPE=GROUP,BREAK=NO) /* REPORT_DATE */
INDEX3=X'D7C1C7C56DD5E4D4C2C5D9',FIELD1,(TYPE=GROUPRANGE) /* PAGE_NUMBER */
INDEX4=X'D7C1C7C56DD5E4D4C2C5D96DC5D5C4',FIELD1,(TYPE=GROUPRANGE) /* PAGE_NUMBER_END */
INDEX5=X'D7D6E2E3C9D5C76DC4C1E3C5',FIELD1,(TYPE=GROUP,BREAK=NO) /* POSTING_DATE */
INDEXSTYLE=NODX

Any suggestions what I'm doing wrong?

Kind regards
Stephan Gausselmann

LWagner

  • Guest
Re: NODX Application (V8.4 on z/OS)
« Reply #1 on: November 15, 2011, 04:28:48 PM »
Did you solve this ?

It turns out that NODX is aout as tightly regulated as construction codes in earthquake country.\

The trick to skip the requirements is rename the indexstyle to DOC, and use a page 1 appearing unique field as the TRIGGER2 value, and orient all fields from there.

Here is the working part of one of mine, with comments:

/* INDEXSTYLE=NODX - error in load */
/* NODX has very specific requirements*/
/* this is a classic NODX def, non-NODX reqs */
/* but use  INDEXSTYLE=DOC with a PAGE 1 trigger  */
/* to guarantee a single load object  * /
/*  */
INDEXSTYLE=DOC
TRIGGER1=*,1,'1',(TYPE=GROUP)
TRIGGER2=*,104,'PAGE   1',(TYPE=GROUP)
FIELD1=-2,118,008,(TRIGGER=2,BASE=0)
FIELD2=-2,74,015,(TRIGGER=2,BASE=0)
INDEX1='POSTING_DATE',FIELD1,(TYPE=GROUP,BREAK=NO)
INDEX2='PPE_DATE',FIELD2,(TYPE=GROUP,BREAK=NO)