Outputting subfile using QTP

[Murray Scholz]

John,

We did it this way in complex QTP's where the output subfile might
contain many scores/hundreds of 'fillers' of various sizes.... we didn't
have to individually specify lots and lots of temps or defines for all
the various sizes.  Also there seemed to be some efficiency (probably
very marginal ?) in using temp over define, as the temp just is used,
whereas the define gets 'evaluated'.

Cognos can reproduce it now... (Access Violation on Alpha , works fine
on good old VAX )
and has been reported to Ottawa as a bug.

Cheers
Murray Scholz

"Penney, John" wrote:
> 
> Hey Murray:
> 
> Velly intelesting as Artie Johnson would say. Why not DEFINE OR TEMP the
> variables before including them in a subfile? Never seen a syntax construct
> like that...
> 
> JP a Canuck in Upper NY
> 
> -----Original Message-----
> From: Murray Scholz [mailto:murray.scholz@abri.une.edu.au]
> Sent: Wednesday, March 10, 1999 6:22 PM
> To: POWERHOUSE (new) List Group
> Subject: Outputting subfile using QTP
> 
> Hello All,
> 
> Was wondering whether anyone out there has experienced something similar
> to the following... (Cognos cannot reproduce problem)
> 
> POWERHOUSE v710G1 , VMS 6.2 , on Alpha
> 
> Q T P   (7.10.G1)
> > access *a-subfile
> > temp a-filler char*1
> > subfile another-subfile keep include &
>         fld-a , &
>         a-filler char*6 , &
>         fld-b , &
>         a-filler char*10 , &
>         fld-c
> > go
> 
> Executing request 1 ...
> %SYSTEM-E-ACCVIO, access violation, reason mask=!XB, virtual
> address=!XL, PC=!XL, PS=!XL
> %NONAME-W-NOMSG, Message number 00000000
> %SYSTEM-F-ABORT, abort
> 
> We use the above technique, in order to 'construct up' subfiles, that
> fit record definitions for an existing COBOL (or other) system, and are
> not concerned with the 'contents' of the 'filler', just getting the
> offsets of the real data fields in the correct positions.
> 
> The above code worked "fine" in v710E , also works "fine" in 710G1 on
> VAX,
>  but dies on Alpha ??!!??
> (well, "fine" up to a point... qtp didn't die, but the actual value of
> the
> filler in the subfile would be binary zero in 1st byte, then spaces in
> subsequent bytes... which wasn't a big deal as we are just using the
> temp field as 'filler')
> 
> Changing the temp to a define does fix the problem, but there are LOTS
> of instances of it through our various systems...Any other ideas ?? ...
> 
> Murray Scholz
> Associate Director
> Agricultural Business Research Institute
> University of New England
> Armidale NSW 2351 Australia
> Email: murray.scholz@abri.une.edu.au
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
This list is closed, thus to post to the list, you must be a subscriber.