|
[EDI-L Mailing List Archive Home]
[Message List]
[Reply To This Message]
UNA and how it's used (formerly: Help with what used to be "Paper free" ASN (DESADV))
That's mostly true, but there are a couple of errors. In fact, there is a
case where the UNA segment is mandatory: If the user is not using the
appropriate default separator characters. EDIFACT's syntax defines two sets
of default separators...but, of course, it's not quite that simple either.
There are officially four versions of the syntax and the rules change a
little with the versions.
As Bernd points out the characters shown are the defaults for UNOA. That's
nearly complete, but there is an omission. Also UNOB gets a different set
of separators depending on the version of the syntax.
Here's how it all breaks out (and it's based on the first two sub-elements
of UNB)...
If the syntax version is 1, 2 or 3 (ex: UNB+UNOA:2...where 2 is the syntax
version), then there are five "service" characters defined:
-> Component separator,
-> Element separator,
-> Decimal mark,
-> Release character, and
-> Segment terminator.
Add to that, if the character set level is UNOA, then the default characters
are defined as follows:
-> Component separator is the colon (:)
-> Element separator is the plus sign (+)
-> Decimal mark is the period/full stop (.) or comma (,)
-> Release character is the question mark (?)
-> Segment terminator is the apostrophe or single quote (')
Still on syntax versions 1, 2, and 3, if the character set level is UNOB
(works with 1, 2, & 3) or UNOC/D/E/F (apply to 3 only), then the default
characters are defined as follows:
-> Component separator is the IS 1 (x1f)
-> Element separator is the IS 3 (x1d)
-> Decimal mark is the period/full stop (.) or comma (,)
-> Release character is _not_defined_
-> Segment terminator is the IS 4 (x1c)
Because the separators are non-printable, the release character is not
defined.
Great, now there's the new syntax version 4 (ex: UNB+UNOB:4...where 4 is the
syntax version). In this case there are all the same "service" characters
as in previous version plus one more: repetition separator. As it turns
out, they "rationalized" the separators a little bit for this version. Now
there is only one default set and it applies to all character set levels:
-> Component separator is the colon (:)
-> Element separator is the plus sign (+)
-> Decimal mark is the period/full stop (.) or comma (,)
-> Release character is the question mark (?)
-> Repetition separator is the asterisk (*)
-> Segment terminator is the apostrophe or single quote (')
So, if you use the appropriate separators according to the entries in UNB,
it is not necessary to send UNA. However, if you wish to use something
other than those defaults, you MUST transmit a UNA before each applicable
interchange. Assuming you use a UNA, any single-byte character can be used
for any separator character. Therefore...
UNB+UNOA:3... is acceptable as is (no UNA needed)
UNB+UNOC:3... is not acceptable. Instead use UNA:+.? 'UNB+UNOC:3...
UNB+UNOB:2... is not acceptable. Instead use UNA:+.? 'UNB+UNOB:2...
UNB+UNOB:4... is acceptable as is.
UNB$UNOX:4... is not acceptable. Instead use UNA:$.?*'UNB$UNOX:4...
It should be noted that before syntax version 4, the second to last
separator space in UNA was not used (officially "reserved for future use").
Therefore, in the examples above, UNAs show a blank after the release
character definition while the syntax 4 example shows the repetition
separator defined in that slot.
Lots of information. Always confusing. Hope that helps clear it up a
little.
Best regards,
Bill Chessman
Inovis(tm), Inc.
-----Original Message-----
From: Eckenfels. Bernd [mailto:
Sent: Wednesday, July 30, 2003 11:09 PM
To:
Subject: [EDI-L] Help with what used to be "Paperfree" ASN (DESADV)
Hello Tim,
> First:
> I'm getting the following at the very beginning of the message:
> UNA:+.?*'
> Where is this coming from?
This is the EDIFACT separation definition. It is common to have this
element, but not mandatory. It lists the separators for sub-, elements,
decimal, quoting, repeating and segments. The values you see here are
default for UNOA and UNOB charset. It should not hurt you, to keep it.
Greetings
Bernd
--
www.seeburger.de
To unsubscribe from this group, send an email to:
Message Identifiers: <SALES>, <JOBS>, <LIST>, <TECH>, <MISC>, <EVENT>,
<OFF-TOPIC>
Access the list online at: http://groups.yahoo.com/group/EDI-L
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|
 |
Subscribe in XML format
| RSS 2.0 |
|
| Atom 0.3 |
|
|
Cart
