differences between teststand sequence file internals in TS2 and TS3.1

[pdobratz]

TestStand 2.0
TestStand 3.1

Does anyone know if there is any documentation available with the changes to the structure of TestStand sequence files from TestStand 2.0 to TestStand 3.1?

In our group, we have a program that loads sequence files and uses the TestStand API to do static checking.  We check things like the existance of steps referenced in Goto Targets and preconditions, and we also check for conformance to certain internal coding conventions.  For some work we are now using TestStand 3.1, and this software tool trips up on certain aspects of the sequence file.  Although this is simple enough to work around on a case by case basis, I was wondering if there is a master list of changes anywhere (since TestStand must have it as it supports saving 2.0 files from 3.1).  Mainly, I am concerned with modifications to the TS container in each built in step type.

Thanks,
Peter

[Eric M]

Hello Peter,

The changes to the sequence file structure are not documented. However, if you take the case of the Step.TS container, you can discover the changes by examining the data type of which TS is an instance. In other words, take a look at the TEInf standard data type in TestStand 3.1 and compare the fields to its predecessor in 2.0.

If you are using lookup strings to check the values of fields in the TS container, you would be better off using the Step class of the TestStand API wherever possible. Many of the properties that are stored in the TS container are properties of the Step class. For example, the Step.TS.PreCond property can be gotten through the API using the Precondition property of the Step class.

Using the Step class in the API to get step properties will make your utility more robust through changing versions of TestStand. While the contents of the TS container may change from one version to the next, the API classes are developed and maintained with backwards compatibility in mind.

Regards,

Eric