Subject: | |
From: | |
Reply To: | |
Date: | Thu, 7 Feb 2008 10:19:55 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
On Wed, 6 Feb 2008 15:39:00 -0600, Francois Desrochers
<[log in to unmask]> wrote:
>Jim,
>
>I tend to agree what Dave Waroff wrote in a seperate posting.
>
>According to the Cobol Reference manual, file status 39 indicates a fixed file
attribute conflict. Based on the name/extension, this looks like a bytestream
file. You might get this to work by fine-tuning the Cobol definition and turn
into a variable-length file ex. RECORD CONTAINS 1 to 1000 CHARACTERS.
>
>It's tough to be more specific not knowing exactly what the file contains and
how the records are really arranged.
>
>Not sure why you are getting file status 14 when the name starts with a dot
but it's definitely not the right syntax if we assume the file is called FILE.CSV
in the JIM group of the JKP account.
>
>HTH
>François
>
<< snip >>
I second what François suggests, and add that I've had execellent results in
reading (and writing) 'bytestream' files containing CSV data with regular COBOL
verbs - by telling the compiler that it's reading a variable-length file.
Just be sure that the values you specify for maximum number of characters in
the RECORD CONTAINS clause really is large enough, or you'll end up
truncating the records on read!
If the file ISN'T a bytestream file, are you sure the record definition actually
matches the record size of the file? The output of a 'listfile' command could
be very helpful here - to be sure we're not fixing the wrong problem!
Brian Edminster
Applied Technologies
P.O. Box 234
Germantown, MD, 20875-0234 USA
"Open Source Software by Applied Technologies
- Solutions for Today, Solutions for Tomorrow"
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
|
|