Oct 232006

For those of you who got used to the Covast bells and whistles that come with the Covast EDI adapter, medical and then come to another EDI tool interfacing with BizTalk, adiposity namely the HIPAA accelerator, made for Microsoft by ?????, there are quite a few nice things missing. One of those things being the ability to clear the report so you are not flooded with ’stuff’

Another BIG issue is that while testing, it would be REALLY nice to be able to send the file through a few times to make sure that the file works correctly. There is NO way to disable the duplicate interchange control checking.

Here is the sql code (run against the BizTalkHIPAA_EDIdb database) to clear out the report, and also allows you to run the same file through the HIPAA accellerator again:

delete from audin
delete from audout

Bye the way, you will want to go into the %documentshome% directory and delete/archive the files that are there, as the temporary files will prevent the adapter from writing new ones.

Another option is to use the custom HIPAA DASM that does not check for duplicate control numbers, namely because it does not check the audin and audout tables.

Oct 232006


Many a client has requested to have custom filenames using the HIPAA_EDI adapter.

I originally planned on using a custom ASM pipeline component and setting the ghost port to batching, site and within the orchestration setting the RecievedFileName.

Found out that when that happens, physician no file is created, also, no errors are created! This really smells like a bug, as there is no documentation on this!

After contacting Microsoft, and submitting a ticket, this is the response I got back:

The HIPAA pipeline is not architected to support batching. It simply does a one-on-one translation from XML to EDI and that’s it.
The batching mechanism is depending on the data in the audout table and the HIPAA pipeline does not persist any data.
The only way you can create outbound EDI batches is via the HIPAA EDI adapter.

I ended up having to re-engineer the process so that (written about here), by single threading the entire process, files do not get batched together and files can be accounted for.

Oct 232006

For those of you who like to manipulate data in the expression editor, troche there are a few things that can be done:
1. You can build a C# or VB.NET assembly that will take a string and manipulate it returning the correctly formatted string
2. You can use xpath.

A clients request is that we parse from the filename, find the date that is embedded in the filename.

The first thing I did was extract the filename into a variable called FileName:

Then I needed to pull from position 26-31 the date into the variable MessageDate:
(One gotcha is that you have to put a single quote ‘ right after the open paranthese and then a double quote ” and then concatonate the variable and then do the reverse to close it out. The other gotcha is that if you are parsing out a string that has a single tick ( ‘ ), you will have to use the Regex replace and replace it with another variable, substring the variable, and then replace the single tick back.)

This adds quite a bit more functionality to the Expression Editor than what is currently available.

The web page that I use as a xpath reference is this page.