[HAPI-devel] avoiding duplicate code

[Alex H. <al...@9p...>]

I’ve been working on a small system to extract demographic data from HL7 ADT
messages. Up until now I’ve only worked with version 2.2. However, in the
future I’d like my application to be flexible enough to move among different
versions of HL7, realizing that there will be differences to be coded
against in each version.

In reviewing some of the class hierarchy in HAPI I have noticed that data
types, segment, groups and messages do not derive from a particular base
class. That is, if I write a function to extract data from the PID segment,
then I cannot write a generic function to accept a PID object across all
versions of HL7.

But perhaps I am missing something? Is there a good reason for this and
should I avoid trying to design for a generic case since each version of HL7
needs to be handled specifically? How do others approach this? I’m new to
HL7, but I’d like to minimize any duplication in my code that deals with
extracting information from HL7 messages across versions.

Thanks,

-Alex Harvey