david sidlingeru«âvÈ�–)àz¹è²–¦ž‹)j

One of our partners like to do things in quasi-non-standard ways, and I have
not had any problems modifying the schemas for their HL7 docs to support
this. Until now...

In the PR1 segment fields 15 and 16, the first subfield is usually a simple
type (ID). However, the partner interface wants to send multiple fields in
the first subfield. I have modified the CE type to make CE.0 a complex type
with child simple types (CE.0.0_Identifier0, CE.0.1_Identifier1, etc.). I
have performed modifications of this type before (changed IN1.17 to accept
multiple subvalues) and not had a problem. However, with this change I
receive the following message in my event log:

Segment Id: PR1_Procedures
Sequence Number: 1
Field Number: 15
Error Number: 102
Error Description: Data type error
Encoding System: HL7nnnn

I'm getting the same error for PR1.16.

Can anyone give me some pointers to track this down? Reflector spelunking
is not doing the trick.

Any help would be appreciated.

Thanks,
David

wjzhang

Hi David,

As I know, we only support customize and extend HL7 schemas with Z objects.
So I cannot confirm if the change of you could work. I need some time to
research this issue and will update you with the conclusion I get.

For information about using Z objects, please refer to the chapter
'Extending HL7 2.X Schemas with Z Objects' in BTAHL7 help do***ent.

Have a nice day.

Best regards,

WenJun Zhang
Microsoft Online Partner Support

This posting is provided "AS IS" with no warranties, and confers no rights.

ghislain bellemare

Hi David,

I had to do the same thing for one of our partner. Verify that the
"allow trailling delimiter" option is on for this party. If this option
is off, you need to specify a caret for all CE type used.

However, I think it's pretty dangerous to change CE type (such as it's
very, very used). I suggest to create a new datatype (I suggest
Customer_type as name) in your base schema and use it only in the
fields you need.

For my part, it's work very well.

Ghislain

david sidlingeru«âvÈ�–)àz¹è²–¦ž‹)j

Ghislain,

Thanks for the advice. Unfortunately, the problem is not in trailing
delimiters or in the data type definition. They have added a new level of
sub-field to the data type. I've accomodated this on other fields, but this
one (because it is a "sub-sub-field"?) is getting rejected by the parser.

Thanks again,
David

david sidlingeru«âvÈ�–)àz¹è²–¦ž‹)j

As it turns out, I had my separators confused and it was the repeating field
separator that was being passed.

The schema did not allow for repetition of the PR1.15 or PR1.16. Modifying
the schema to allow for multiple occurances fixed the error.

If Microsoft is taking suggestions for future versions/patches of the HL7
parser/disassembler, I would suggest something less vague than "Data type
error" when the error is being caused by an unallowed number of occurances.
It seems that "Data type error" indicates that the actual data is the problem.

Thanks,
David