If you use the file system object and textstream object, you can use the
read method on the textstream to read a specified number of characters.  You
would first read enough characters to get the record type.  Then read the
number of characters for that record (minus what you have read for the
record type) to process them.
Coral

>
> ----------------------------------------------------------------------
>
> Subject: reading multi record ebcdic file with different record lengths
>  an d packed data
> From: "Calhoun, Bob" <Bob.Calhoun@M...>
> Date: Mon, 30 Dec 2002 13:47:03 -0500
> X-Message-Number: 8
>
> does anyone have any experience reading an ebcdic file  that is made up of
> multiple record types with different record lengths... these
> records are not
> cariage return line feed delimited either...
>
> example;
>
> record 1 type a length 100
> record 2 type b length 300
> record 3 type c length 50
> record 4 type c length 50
> record 5 length 200
> record 6 length 500
> record 7 length 1100
> record 8 length 1100
>
> then back to record 1
> with varying combinations, and this is in ebcdic with some packed data
> also...
>
> I am trying to use line input and I seem to lose some bytes..
> plus line input seems to try to read the whole file in and run out of
> space...
>
> anyone have any ideas on how to approach this?
>
> thanks,
> jb
>
> Confidentiality Notice: This e-mail message, including any
> attachments, is for the sole use of the intended recipient(s) and
> may contain confidential and privileged information. Any
> unauthorized review, use, disclosure or distribution is
> prohibited. If you are not the intended recipient, please contact
> the sender by reply e-mail and destroy all copies of the original message.
>
>