Re: [HAPI-devel] Segment Terminator Rules for PIPE format

[Jim K. <ji...@ho...>]

Hi Nico-

Thanks for the quick reply.  I'm using 0.4.3 and I'm 100% certain that there 
are no cr/lf characters anywhere but at the end of each segment.  Hotmail 
garbled the sample message segments that i sent to the mailing list.  Each 
of those segments I sent are contained on one line when sent to the 
interface.

While digging through PipeParser I spotted this code at the end of the 
encode method:

return stripExtraDelimiters(result.toString(), 
encodingChars.getFieldSeparator());

It seems like this method is pretty indescriminate, it essentially wipes out 
all delimiters up to the last populated field.

I have a feeling that this is the source of my problem. However the real 
question is, what is the proper behavior according to the HL7 specification?

I'm working with a CDC ELR implementation guide that specifies how to use an 
ORU for ELR messaging.  In all the examples, if a segment terminates before 
exhausting all possible fields, the segment terminates with pipe+cr/lf. so:

MSH|MSH.1|.....|MSH.12|<cr/lf>

If the segment terminates with the last field filled then it terminates with 
just <cr/lf>:

MSH|MSH.1|....||MSH.20<cr/lf>

I noticed that the examples of HL7 vendors like Chameleon follow a similar 
approach.  Is this approach right?  I'm inclined to think it may be but I 
can't track down the HL7 specification that states it definitively.  If the 
approach is right, then the stripExtraDelimiters method is a problem.