HRL Help needed, Please

[Anonymous]

Hi,

I have some idea about the HRL and VRL and I read topics or solutions I was looking for and almost got it, but not yet…

I have file like this..

“H”,”someval”,”thisval”,”four”,”five”

“S”,”one”,”two”,”three”,”four”,”five”,”six”

“T”,”1″,”2″,”3″,”4″,”5″,”6″,”7″,”8″,”comments”

“T”,”11″,”22″,”33″,”44″,”55″,”66″,”77″,”88″,”comments”

“S”,”one1″,”two2″,”three3″,”four4″,”five5″,”six6″

“T”,”1″,”2″,”3″,”4″,”5″,”6″,”7″,”8″,”comments”

“T”,”11″,”22″,”33″,”44″,”55″,”66″,”77″,”88″,”comments”

“T”,”1″,”2″,”3″,”4″,”5″,”6″,”7″,”8″,”comments”

“T”,”11″,”22″,”33″,”44″,”55″,”66″,”77″,”88″,”comments”

“F”,”f1″,”f2″,”f3″

I need to convert this to HL7.

I don’t really care about first line and last line in the file.

“S” line has all info about the patient that you can fill in for MSH and PID/PV1.

“T” has lab results.

As you can see each “S” pertains to one patient followed by “T” lab results (OBX) for that patient.

How do I go about doing this.

I created 2 VRL’s one for the “S” and one for the “T”. I created HRL but my HRL is treating each record as one individual record. I am not sure about the Global definition in VRL. I tried “n”, ‘n”,n in the termination.

How do I go about fixing this.

Any help is greatly greatly appreciated.

Thanks,

Hari

Topic

[Charlie Bursell]

This looks to me like a poorly designed Comma Separated Values (CSV) record format.  Normally with CSV, the double quotes are used only when the field contains special characters, i,e., a comma, etc.  CSV can be difficult to deal with sometime because of all the nuances.

If I were doing this interface, I would deal with it as CSV.  While you can do that with VRL, I find it easier to use the csv package which is part of the tcllib library.  If you don’t have that it can easily be downloaded.  I think the latest version is tcllib1.7 o greater.

Once installed

package require csv

foreach line in your data file

set list [csv::split $line]

This will return a list of values for each file.

I would then normalize it to an HL7 variant.  Just change the first field to TXX, HXX, etc.  Then join the list with a pipe (|).  Now all you have to do is prepend a dummy MSH and you have an easy to deal with HL7 message that can be easily translated to whatever you want.

Just my opinion.  I am sure there are others

[Anonymous]

Well, this is not a CSV file as the example file I quoted certainly have comma separated values in it. Like the name. My bad on that file list.

So this is truly a text file and with in the double quoted, the values can be “ONE, 1”.

So a csv is not an option in this case.

[Author]

Replies