HL7 questions, after some experience

[OrthoDocSoft]

Folks,

I was literally looking for "<cr>" to know when the end of my first segement (MSH) was. Now that I'm getting real messages, I don't see these four characters anywhere, and I'm betting I was looking at "fake" messages where <cr> means "carriage return", but that's ok. This practice was returning essentially a blank MSH segement (no good), so I'm abandoning it.

My real world MSH segments have 20 pipes > | < in them, so I can count them to find the end, I think,

OR

I could look for "PID" (the next segment) as a signal for the end of the MSH segment.

Does anyone have an opinion which is better?

Does anyone know a better way to parse this portion of the HL7 message? I only need the date and time bits, and the MessageType data, so....

Thanks,

Ortho

"you cain't fix 'stupid'...

 

[dilettante]

I had the idea these were designed to be pretty simple to parse. Are you sure you can't find the end of each message by looking for vbCr characters?

Waiting for a new segment ID like that "PID" has the disadvantage that you may have to wait some time for a new one to arrive, holding the current one buffered for possibly a very long time. Plus these messages probably come as a sequence of messages for each patient, so patient IDs may have other messages following them instead of another PID immediately.

I agree that the <cr> you were seeing in documentation was just a "meta-linquistic variable" standing in for the CR character (value 13 or &H0D, i.e. a vbCr character).