I did not see this in the postings, so i am posting it again.

Hello Roy.

We receive a csv ASCII file from a bank onto a Windows XP computer.  The 
records in the file are of varying length.  The fields in the file are all separated 
by a comma.  At the end of each record is a CR/LF.  However, if one of the 
fields in a record is smaller than the same field in another record, that causes 
the CR/LF in the record to be in a location(column) that is different than other 
records.  Consequently, the records in the file from the bank are of varying 
lengths.

Since we are using Minisoft WS92 as a terminal emulation program to log on to 
the HP3000, we are also using Minisoft to upload the bank's csv ASCII variable 
length record file to the HP.  In the upload to the HP, there is no option in 
Minisoft to specify that, the file being uploaded to the HP, should be created 
on the HP as a variable length file.

Once the file is uploaded from the PC to the HP, I have an HP COBOL II 
program that i want to open the file with and read it sequentially from 
beginning to end.  After reading the first record, which is 256 bytes in length, 
the data in the first 256 bytes, contains all of the first record plus some of the 
second record.  The second record read contains the second part of the 
second record and part of the third record.  I guess what it is doing is giving 
me 256 bytes at a time, but the records are really different lengths.

Please let me know if you need anything else.

Thanks for all of your help.

Tony

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *