Cloverleaf 5.5 and extra pipes….

[Tom Rioux]

I’m doing testing for conversion to Cloverleaf 5.5.  We have messages from the inbound source that has extra pipes at the end of each segment.  In 5.2, these extra pipes are sent out.  However, in 5.5, these extra pipes are SOMETIMES getting removed and sometimes they aren’t.   I can’t find any rhyme or reason as to why it is happening.

This is happening (SOMETIMES YES AND SOMETIMES NO) on routes with Xlates.

This is happening (SOMETIMES YES AND SOMETIMES NO) on routes that are RAW.  

The ONLY common denominator I can find where it IS happening, is that there are no outbound tclprocs on the routes that this is happening.  But there again, I can’t say it is only happening on routes without outbound tclprocs, because sometimes yes, and sometimes no.

Any ideas?

Thanks….Tom Rioux

Topic

[Jim Kosloskey]

Tom,

When an Xlate is involved I would not expect to see separators for trailing null fields in segments (that is the way the standard is defined).

I would expect to see those when an Xlate is involved only if something special is being done in the Xlate to make that happen.

I do not know what happens when BULKCOPY is in the picture vis`a`vis trailing null field separators.

But a raw route should be different. I would expect to see the same thing coming out that came in unchanged (unless there are some procs involved).

Jim Kosloskey

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Robert Kersemakers]

In case of an Xlate, check the HL7 variant. Maybe it has changed in 5.5 and includes a mandatory field at the end of the segment.

Not sure about the RAW routes though.

I would take some of the inbound messages where the extra pipes are removed in outbound and then put them through the testing tool with all the inbound procs, Xlate and outbound procs. Check which proc/xlate removes these pipes.

If the testing tool is not an option (this is the case when info is generated during Startup of a proc), I would recreate the entire interface and echo the message after each proc/Xlate.

Zuyderland Medisch Centrum; Heerlen/Sittard; The Netherlands

[Tom Rioux]

Thanks for your input guys.  I have found the culprit.  My message file that I’m using to do my testing is from March 5th.  Well, unknown to me, there was an inbound tcl proc put in place on March 11th.  This proc makes a call to the msgs.tcl package in the libs directory.  We have two scripts for this particular interface in the inbound tps.  I noticed that when the script that doesn’t call the package gets run, it doesn’t remove the extra pipes, but when it runs through the script that calls the msgs.tcl package, it strips the extra pipes from the message.

May not sound like a big deal, but we actually do have a couple of vendors that actually want those extra pipes.  It’s good to know what is causing it now.

Thanks….Tom

[Author]

Replies