Commitment control ...

["Hans Boldt" <hboldt@xxxxxxxxxxxx>]

Hi Ambarish!  What you are seeing is the effects of blocking.  A
sequential output file is blocked.  An input with add file is not.

Blocking provides a performance improvement since one read of
several records is generally faster than several reads of single
records.

Hans Boldt, ILE RPG Development, IBM Toronto Lab, hboldt@vnet.ibm.com


Ambarish Bagali <as4_sup2@sriven.scs.co.in> wrote:
>Hi,
>I have observed one strange thing(??). I have written one RPG. The piece of
>code is like this:
>
>File1 is used in 'O' mode
>
>FFILE1  O                              DISK
>C              DO   5
>C              MOVEL'ABCD'     FLD1  << Data base Field >>
>C              WRITE FILREC
>C              ENDDO
>C              SETON        LR
>C   'ARB'  DSPLY
>
>Compiled RPG normally. (no Commit control). I stopped my program at the
>last line. And found no records written in to file FILE1. (I used RUNQRY to
>check).
>
>Then I changed  code a little bit.  I defined the file in I/A mode.
>
>FFILE1  I                              DISK     A
>C     1      IFEQ 2
>C              READ FILE1      30   << Dummy read>>
>C              ENDIF
>  *
>C              DO   5
>C              MOVEL'ABCD'     FLD1
>C              WRITE FILREC
>C              ENDDO
>C              SETON        LR
>C   'ARB'  DSPLY
>
>I Stopped  my program at the last line, I found the file being populated.
>Could any body please tell me what exactly is happening?
>
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to "MIDRANGE-L@midrange.com".
| To unsubscribe from this list send email to MAJORDOMO@midrange.com
|    and specify 'unsubscribe MIDRANGE-L' in the body of your message.
| Questions should be directed to the list owner/operator: david@midrange.com
+---