[Ruth Berge]

Did you ever get a resolution to this problem?

Someone else in my department may have a similar problem.

[Ruth Berge]

The problem was simple.  I was using an inbound file for testing.   I had copied the same file to two folders.  This is because the receiving thread consumes the file and destroys it.  I had one copy to keep resubmitting and review.  What I had also done was fix the carriage return/line feed (0d 0a) in the file.  But the file that I was actually using did not have the fixes.  I kept looking at the wrong file to validate that it was fine.  The actual file was sitting in my Notepad editor and instead of copying from the other file, I was saving that file each time.  It wasn’t until I was working with someone else that I happened to review the right file.

Also Dale Powers (the person who was helping me) suggested that I change the inbound feed from End of File Terminated to New Line terminated.  

Those were the only two changes.

[Ruth Berge]

Lacking a response here (and granted this was 5 years ago now), I recall that we coded a java process to do this.

[Ruth Berge]

Did anyone every respond? I see that the only reply was to tell you to post in another forum.  Did you try that?

I have to do the same thing (different state) and would love to see what anyone says about this.

[Ruth Berge]

The cl_check_ack_modify was edited so that a successful message would write to a specific log file.

under switch – exact

AA-CA{

  set Ambfile [clock clicks]

  set Amb01  [open ambupdate/$Ambfile$disp w]

  fconfigure $Amb01 -translation binary -encoding utf-8

  if {[uplevel #0 {info exists EuroBinary}]}  {

  fconfigure $Amb01 -encoding -eofchar {}

  }

  msgwrite $style $my_mh $Amb01

  close $Amb01

 #Good ack clean up

  set send_cnt 0             :#init counter

  return “{KILLREPLY $mh} {KILL $my_mh}”

  }

The set up is like this.  Lets say that you want a folder that has a file for each message that is successfully sent from thread MyProv01.  MyProv01 is a sending thread.   You have a tcl proc cloned from cl_check_ack_modify as shown above so that the success ack creates a file to a specified file name.

On the thread, on the Outbound tab, enter the name of the tcl proc you created in the TPS Inbound reply.  Also have Send Ok Procs set to cl_save_ob_msg

In this example it create a file in a folder under the exec process.  I have to create the folder when I install the site.

A different process can read the files and do what is needed with them.

Again, I had help from others to get this working.

[Ruth Berge]

I wanted to report back that the method that Charlie Bursell describes appears to work.  I have not fully tested but I do know that if the message is not sent then it will not log it.  This was not a piece of cake for me to get set up and I could only do it with assistance internally.  I’ll try to update again when I have more complete testing but I do want to thank you for the suggestion.  I could not determine which of those tcl procs would be a good starting point and it was invaluable for you to give me that information.

[Ruth Berge]

the report must be generated automatically.  I don’t know how to count SMAT!   Also i want to be able to do something with the message after it is succesfully sent.  For this reason, I am hoping to get a log file with the successful messages.

[Ruth Berge]

I am trying to meet a Meaningful Use lab menu item that requires a count of all messages sent to an Ambulatory provider.  I am assuming that they mean messages that were successfully sent by the way.  For my purposes, a log file of those messages will work.

[Ruth Berge]

I looked again at the messages.  

First, I guess I was leaping in my first statement.  The issue that I see in the CCG testing tool is that even though the XLATE sorts the data into 4 messages, the testing tool labels all of them as Message 1.  I thought this was because there was no line feed after the hex 0d on the last segment.  This may not be true.  It just happens that a file I sent through with a single message does have hex0d and hex0a at the end of the message.

So I will then rephrase the question and ask what I should expect from the CCG testing tool?  If the XLATE being tested is successfully breaking out multiple messages from one, then should be output in the test tool label each message as “Message 1” “Message 2” etc.?

Then I will take the other information as very valuable as I continue (no pun intended).  And thanks for all the responses before and after this.

[Ruth Berge]

In the cases where you have this working, are you using only the xlate or are you using the tcl proc which I have seen posted here?

When I look at the help text, it certainly implies that CONTINUE will create a new message.

[Ruth Berge]

Thanks,  so i do not see a carriage return and line feed (0d, 0a) in the output

I also did change it to CONTINUE with the same result.  I had forgotton about the post processing.  Thanks.

I’m trying to read between the lines of what you are saying.  It appears that CONTINUE and SEND do not add the extra message delimiters.  Is there an easy way to do this?  I want to generate multiple messages from one message.  As it is, the message contents are correctly created but the delimiters between the messages are not there.

[Ruth Berge]

The support for many of the multi component values including the mime type in OBX-5 is only in HL7 v 2.6.   I’m not looking it up right now mind you but if you are strictly looking for correct HL7 then I do not believe that even the ED datatype can be used as described here unless they are declaring the higher version.  If the interface is 2.3.1 then the correct interpretation would be to ignore those sub-components.

[Ruth Berge]

Where can I get tcl file mentioned in this thread?

[Author]

Replies