EXTOL Integrator can support non-standard (or, proprietary) EDI documents with certain constraints and limitations. These types of files can be considered fixed length or delimited files.
In an earlier post we discussed Flat File Set-up in EXTOL EDI Integrator.
Questions and Answers
I am trying to create a new outbound document using proprietary standards. When I do the outbound translation, I get no data, and no error messages on the outgoing application data log. It says it's enveloped, but when I look inside, there's nothing. I can see that there's data in the file, when I use dspfilseq. So I'm not sure what I'm missing.
I needed to qualify the EXH and EXT segments in the message class by the L and the *STSK and *SESK respectively and in the standards none of the segments had a loop repeat, these needed a value.
Customer was getting a Trans error on inbound proprietary document. the error was indicating the first segment in the message class was not found.
Customer had the EXH and EXT qualified by the L = LoopID with *STSK and *SESK this is not needed on inbund only outbound.
How do we suppress the segment IDs for an outbound proprietary document?
Customer needs to go to data area EXFLGSL and put in the industry group and the Segment ID length.
Below is the data I am using for inbound. Is it correct?
NT001THE PERSON ENTERINGS NAME
NT002THE PERSON ENTERINGS EMAIL
Are the segments in an extol map?:
NT (N Occurring)
DT (N occurring)
The example below has three elements in it, correct?
NT001THE PERSON ENTERINGS NAME
When we set-up the document for this record we are going to use the segment "NT". Does this segment
have any bearing on this inbound data?
Correct. The segment id must be on the inbound data and must be noted by position in the EXB built during unwrapping (it gives the position of the segment id and length -- excluding the EXH).
For outbound, how would a "dummy" loop start segment be removed prior to wrapping. Do you have to flag it to be removed in the map or is it done by omiting the first 1-3 segment positions with the data area?
As with the normal translator, suppressing a loop start segment will also remove all segments within the loop. The suppression in the data area refers only to suppression of the segment id (you use a dummy segment id that is more meaningful to the mapper, with element 01 of that segment containing a constant for the actual segment id value). The segments are still sent but the "standard" segment id is removed from the translated data.
To do batch data and allow multiple TPs to send data in one connection we are able to put all the envelopes at the document level and not in the script. Are we still able to retrieve a unique 'control' number if the envelopes are applied at this level?
In proprietary data processing, you can either have a program that takes the connection data and inserts envelopes (EXB, EXG, EXH) at the appropriate levels or do the same thing with a script. The difference is that by doing it in a script, you can only ever have one interchange, one group, and one document because the script wraps the envelopes around the entire batch -- not on the individual document breaks. This presents the problem that you cannot have multiple tp's in the same connection. The only way of handling this would be through a pre-process program that puts the envs in their proper place and also identifies the trading partner.
By: Sean Hoppe on