XLT Test tool not showing messages

[Daniel Lee]

Running Cloverleaf 6.0.2.0P on AIX 6.1 and having some odd behavior in the test tool.  For some reason the test tool isn’t displaying the outbound message when I run a test on a test message.

If I put a debug echo statement at the very end of the Xlate I can see that it writes to output so it is getting to the very end of the Xlate.  However, it doesn’t show the message and just shows a “MESSAGE 1” with nothing underneath it.  Whats just as strange is if I put a “SEND” as the very last line of the Xlate and run the test then I do see the message in the test tool.

Has anyone seen this type of behavior before?  Do you think there’s an issue with my Xlate or is the testing tool just being quirky?

Topic

[Daniel Lee]

I’m not going to be able to look at this until tonight but I’m thinking I may have a suppress way at the top of my script.  However, I thought that if the Cloverleaf Xlate code hit a suppress then it stopped processing the Xlate.

I also was able to test this in a route on the test system and saw that it’s working the same way as it is in the test tool so the test tool is working fine.  It’s in my Xlate somewhere.

[Jim Kosloskey]

SUPPRESS does not stop the processing of the Xlate.

The Xlate will always attempt to produce a message unless told not to (this is what the SUPPRESS does).

So SUPRESS can be used unilaterally (primarily used when your Xlate is creating multiple messages from one message) or conditionally (such as when determining a message should NOT be created if certain conditions are true).

So a unilateral SUPPRESS no matter where it is placed in the Xlate without the use of a SEND or CONTINUE will result in no message being produced – but the Xlate will have gone through all of the efforts to build the message just not produce one.

You have experienced the above.

I prefer to use the CONTINUE over the SEND in normal circumstances. The CONTINUE says place the created message(s) next in the engine for any post Xlate processing. The SEND says place the message(s) at the OB protocol bypassing post Xlate processing.

This can become a problem if the Xlate has been running for some time and now one needs to create a post Xlate proc. The proc is placed properly but in engine testing does not show the effect of the proc. That is because the SEND bypasses that proc.

One can bang their head against the wall for quite a while figuring that one out.

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

[Daniel Lee]

Thanks Jim.  I bet it’s as simple as a suppress that I have at the top of my Xlate. (I have like 5 of them in my Xlate)  I’ll take a look at it tonight and post a reply.  There is not a SEND in my Xlate, I was only using the SEND as a debug tool.  But now that you’ve explained what happens in the Xlate when a Suppress is called it makes perfect sense that the SEND would generate the message.

[Author]

Replies