Re: Read/Write records from/to IFS of variable length nature in a reliablemanner

["Jim Franz" <franz400@xxxxxxxxxxxx>]

Jorge,

You mentioned cpyfrmstmf.
What is the CCSID of the file in the IFS - (wrklnk and opt 8 to display 
attributes - 819 might be a normal value.)
Have you looked at the records in IFS in hex - the wrklnk display option?
In ascii, if viewing in IFS, CR LF it's 0D and 0A   (US).

Hopefully you have a reliable end of record character(s)? If not 0D0A - give 
us what you got.
If not same in each record, that's a data problem.

In rpgle - a prev post mentioned Scott Klement's great tutorial and samples,
when you read the data in, it will be in ebcdic so carriage return (CR) is
x'0D' and line feed(LF) is x'25' ( or together crlf = x'0D25'), so you can 
scan for the end of line character(s) to know where your data stops.
Variable length csv (comma separated variable or some other separator 
character like pipe symbol) files are pretty common.

Jim Franz

----- Original Message ----- 
From: <jmoreno@xxxxxxxxxxxxxxxx>
To: <Midrange-l@xxxxxxxxxxxx>
Sent: Monday, November 07, 2011 5:09 PM
Subject: Read/Write records from/to IFS of variable length nature in a 
reliablemanner


> Hello everyone
>
> I have this question and I have been struggling with this most of the day.
>
>
> Is there a way to Read IFS files containing variable record length
> records?
>
> The CPYFRMSTMF is not behaving well against a flat file that has a
> variable length min (2000) and max (6736)
> Using the CPYFRMSTMF  interface and everyone of the EndOfLine formats has
> not allowed me to map the file correctly
>
> I have used a simple RPGLE program to read the IFS file, but the data
> BUFFER length that I used to get to the data was 10,000
> and I could not get a reliable control character for the break ( *CR, *LF,
> *CRLF, *LFCR, *ALL ...)
>
> It should not have to be hard at all. You get an IFS file with or without
> variable length records and we should be able to map that to a flat
> database file.
> If the file does not exist, ... why not create it! ... (whishing for
> much!)
>
> Any recommendations?.
>
> Regards
>
>
> Jorge Moreno
> --
> This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing 
> list
> To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.