Subject: | |
From: | |
Reply To: | Glenn A. Mitchell |
Date: | Tue, 28 Sep 1999 15:23:36 -0400 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
In a prior life, I had to read IBM mainframe data which included packed data
fields (in fact I once had a mainframe tape that mixed ASCII and EBCDIC in
the same record!). If the packed data always resides in the same positions
on each record, FCOPY can easily process the tape converting only the
non-packed data. If the packed data resides in different positions in the
records depending on the type of record being processed, then you have to
have a program do the translation (typically using the CTRANSLATE
intrinsic). Generally, in those sorts of files, there is a record type
indicator of one or two characters at or near the front of the record. Just
remember you either have to translate it to ASCII first before comparing it,
or compare it to the EBCDIC value (e.g. IF RECORD-TYPE='A' won't work until
RECORD-TYPE has been translated; IF RECORD-TYPE=%301 will work OK w/o
translation as %301 is an EBCDIC encoded 'A'.)
Good luck!
--
Glenn A. Mitchell Mailto:[log in to unmask]
3GM Associates, Inc
Portland, ME 04102
207-772-9370
|
|
|