Tracy,
A segment terminator can be just about anything a person wants within
reason. The catch is that if the segment terminator appears in the data
either by accident or in general use than your whole day will be ruined
faster than you can say Oh boy.... Some choose the unprintable character
as it's probability of appearing in the data either by accident or on
purpose are real slim. The bottom line is that you really have to try to
put an unprintable character in an application program to get it into a
data item that could be used in an EDI transaction set somewhere down the
road.
The other school is to place an obscure printable character in the data as
the terminator, (i.e. something like a ~ or $ ) These can be entered from
the keyboard, but make it easier for the maintainer to look at the file
when accomplishing those pesky day to day jobs that we love so much, like
trying to figure out where the segment terminator went.
Take the files and make hex dumps of both of them. From the beginning of
the GS segment, the preceding three bytes will be the segment terminator,
sub-element terminator and the element terminator, in that order. This
will give you what is being used on the returning 997 which may help you
to resolve your problem a little quicker. A good rule of thumb is to hex
dump when in doubt because there could be something there that you can't
see.
Based on your situation, it appears that you are looking for the $ as the
segment terminator which locks you into one character and one character
only, (I'm speculating a bit here.) This is wrong as it limits your
ability to go with the flow in EDI. The standard allows the use of any
character desired and it's controlled by the originator of the
transmission as to what that will be.
EDI files are text or character files which for the most part can be
manipulated with any text editor and should be sent as character files be
it through a VAN or the internet. Making them a binary image will
complicate the situation if the other partner is not aware of this and can
pick it apart on the other side.
On FTP, we use it internally between systems and have a data file and a
done file which are passed together. The done file contains the byte
count of the data file. The idea is that the data file is checked by the
receiver against the originated byte count in the done file. If they
don't match, then there is a problem in the movement of the file from one
system to the other. This scheme works great and traps any data
transmission errors where data has been "lost" during the FTP operation.
I hope that this helps you. Good luck....
Mark
----------
From: Tracy Long[SMTP:[EMAIL PROTECTED]]
Sent: Thursday, April 27, 2000 6:46 AM
To: [EMAIL PROTECTED]
Subject: Using FTP instead of VAN losing segment terminator
Hello everyone,
I need some guidance with using FTP instead of a VAN for EDI. We have
decided to go with FTP with one of our trading partners because the data
we are transmitting is so large.
We had some problems with the data transmission. I had to unwrap the data
and request that they recognize a '$' for segment terminator. Evidently,
when they received from the VAN this was handled for them.
I was shocked when the TP told me that he ignored the $ as the segment
terminator. I thought that this was part of the X12 standard. I am not
sure what translator he is using, but I am guessing that the on unwraped
EDI data that the carriage return could be used as the segment terminator.
The problem that I am having is with the 997 FA. It looks identical to
the 997 FA that I am receiving via the VAN, but I get errors stating that
there is no segment terminator. I can request that he sends a '$' back as
the segment terminator, but I would really would like an explanation for
this. I am transferring the file in ASCII to the mainframe. I am
guessing that the segment terminator is non-printable and non-displayable.
Should I be transferring this file in a different format or with some
other parameters ? I have tried other file formats and get the same error
messages. What I am doing wrong ?
Thanks in advance for your comments,
Tracy Long
Sharp Manufacturing
Memphis,TN
=======================================================================
To signoff the EDI-L list, mailto:[EMAIL PROTECTED]
To subscribe, mailto:[EMAIL PROTECTED]
To contact the list owner: mailto:[EMAIL PROTECTED]
Archives at http://www.mail-archive.com/edi-l%40listserv.ucop.edu/
=======================================================================
To signoff the EDI-L list, mailto:[EMAIL PROTECTED]
To subscribe, mailto:[EMAIL PROTECTED]
To contact the list owner: mailto:[EMAIL PROTECTED]
Archives at http://www.mail-archive.com/edi-l%40listserv.ucop.edu/
