I routinely do Lab Results data in PDF format by having the sending system ‘Base 64’ encode the PDF data. The receiving system then decodes the data before storing it in the db.
The crux of the issue is that PDF and Postscript data can contain characters that will cause problems in Cloverleaf. Unexpected interpretation can occur if, for example, characters in the mlp wrapper (0x0b at the beginning and 0x1c, 0x0d at the end) come in the middle of the HL7 message. The ‘Base 64’ encoding makes all the characters in the printable ASCII set allowing the data to pass through the engine w/o unexpected side-effects.
Base 64 encoding is fairly easy to accomplish and is well documented on the web.
You may be able to do FTP of these types of files without encoding the data. We’ll see what others have to say.
It depends on what you want to do with those files. Cloverleaf doesn’t have a problem reading or protocol bridging any file. So, you should be able to read a postscript file and use tcpip to send it to another system.
You should also be able to open these files using TCL inside the engine and manipulate them if you want. The only thing that you can’t do is create a translation to change those files since there isn’t a Cloverleaf recognized layout available.
We potentially could have an interface where the vendor system would FTP a file to CL and then there would be a thread to pick that file up and send over to a Cerner system. That file would most likely be Postscript or PDF and I just wasn’t sure if Cloverleaf could even recognize those file types. Currently, we have only 1 interface that isn’t HL7 to HL7 so my experience with other file types is pretty much nonexistant.
You shouldn’t have a problem with Cloverleaf recognizing any file. The FileSet driver is able to read any file that shows up in a specific directory. So you could set it up to process the directory where the other system is going to ftp the file. Cloverleaf sees it and reads it.
However, if you’re not going to process the message inside of Cloverleaf you may want to think about accomplishing a simple move another way. For instance you could use cron (unix) or the scheduler (windows) to run a script that looks for files in a directory and moves them along. Or you could use the Cloverleaf file driver to ‘see’ the file and then use the directory parse TPS to run a TCL procedure that moves the file.
If you do use Cloverleaf then you’ll want to raw route the transaction.
Cheers,
Author
Replies
Viewing 3 reply threads
The forum ‘Cloverleaf’ is closed to new topics and replies.
.
.