Revert to default 3.0 type versions

[RobBoyer]

I'm having some type version conflict issues and hoping someone can let me know what the best way is to revert everything in the TestStand type palette back to default for TestStand 3.0.  (The version of each type that I'm supposed to have appears to be 3.0.0.12373, but many of the versions somehow have been incremented to 3.0.0.12374.)

I'm not really sure how this happened, but I'm more concerned with how to get everything back to default in the type palette as well as in all of my sequences.  I managed to get some of the type versions back to the right version in the type palette by performing the following steps:

1) Change the affected types in the global type palette to "Always prompt the user to resolve conflict"

2) Open up a sequence with the correct versions of the types.

3) Tell TestStand to use the sequence file's version of the type.

4) Save all the type palette ini files.

One problem with this is that as soon as I open another sequence file that has version 3.0.0.12374 of any of the types, TestStand immediately replaces the newly repaired global type palette.  Also, when going through the above resolution procedure, I noticed that I have some version 3.0.0.12373 copies of type definitions that have a time-stamp of todays date!  So even if I do manage to get all the sequence files repaired, I have no idea if I've managed to modify a type and save it as version 3.0.0.12373...

I'm beginning to wonder if I should just re-install TestStand, but I can't shake the feeling that there's probably a better way to do this.  (There certainly should be.)  Any pointers would be greatly appreciated.

Thanks, -Rob

[AllenP]

Rob,

Unwanted Type Propogation can be nasty to clean up.  Since the Type Differ tool only works for versions 3.1 and higher,  you will have to do something a little more crude to solve the issue.

The first step is to make sure your Type Palettes are clean.  It sounds like you have done this.  Once you have done this, one way to prevent accidental type propogation is to make the files Read Only.  This will make it more difficult to "infect" these files, since you can't overwrite them.  There are also some other files that are loaded automatically, such as the FrontEndCallbacks sequence, and your Process Model.  Make sure all of these files are clean.  If you can start TestStand, and observe all the types are at the proper value, you probably have done this correctly.

The next step is to find the source of the problem.  If you are ever prompted to save your Type Palette files, this usually means that you have opened a file that has a newer version of the Type than you want.  If you know which file(s) you opened, this will help you track down the source of the problem.  One trick that you can do is to change the version of the incorrect type to 0.  Save the sequence file causing the problem, but not the type palettes or Process Model.  If you shut down TestStand, then start it back up, the correct version of the Type should be loaded in your sequence file.  Save the file again.

If you know that a particular version number is incorrect, you can always search in files for the text (for example, if you know you have version 3.0.0.12374 somewhere, search for that text in Windows Explorer).  It should give you a list of any file that uses that type.  This will help you narrow it down.  I still recommend you revert any changes by going through the Sequence Editor, rather than manually editing the files in a text editor.

For others that run into this issue, the Type Differ, located in NI Labs at http://www.ni.com/labs under the TestStand Large Application Toolkit, will help solve these issues.  In the future we are looking at other solutions to solve unwanted type propogation as well.

Allen P.
NI

[RobBoyer]

Thanks for the tips, Allen.  I wound up copying the global type palette inis from a clean PC, and then using your trick of setting the file's versions of each type to 0 to force it to get the correct version from the global type palette.  I think all is well now, (though I've got numerous sequence files to repair still).

Anyway, thanks again.  I'll give you 5 stars if I can figure out how... 🙂