Hi,
My question is around the best way to manage variables between sequences. I have found or tried 4 methods to date, but each has its own draw-back. The problem I am trying to solve is to be able to create a container that contains a group of related variables and share these between Sequence Files and Sub-sequences. The problem I run into is when we have 50+ sub-sequence files, what is the best way to use a container that allows for the quickest and easiest way to add, delete, or modify variables in a container that is shared about the 50+ sub-sequences. Here is the current 4 approaches that we have used and the draw-back with each:
1. Station Globals - These worked ok, except when we went to move or share sequence files amongst several set-ups. It was a constant battle to try and keep the station global file containing the variables up to date amongst the various test machines.
2. Parameters - These worked until we needed to add, delete or modify a shared container that was used amongst the 50+ sub-sequences. Anytime 1 variables is adjusted at the top sequence, then we found you had to copy and paste the container to all 50 sub-sequences as well as updating the actual sequence call.
3. Locals (Propagate to Subsequence & Allow Propagation from Caller) - This took care of our adding or removing variables, but made debugging very difficult in that it was not clear as to what was being passed across sub-sequences vs. being used locally.
4. Creating Custom Data Types and using them as parameters - This was as close as we could get to being able to modify containers in sub-sequences, but found that this could be difficult to manage and maintain over time in that the number of data types created as well as the type conflicts when adding or removing a variable that is used in a newer sequence and then trying to run an older sequence which created a type conflict. We also found it very difficult to delete an older type, when we knew it was no longer in use but could not find it listed in older sequence by simply using find/replace features.
In the end, we want to be able to re-use older sub-sequences for testing, but to be able to affectively add, modify or change variables at the very top level that then cascades. Is there another approach that we are not aware of that can address this or help make it easier in the long run to manage and maintain as our number of sub-sequences grow?
Thanks,