LabWindows/CVI

cancel
Showing results for 
Search instead for 
Did you mean: 

Possbile CVI 9.0 bug - build version auto-increments off-by-one for release and debug when batch-building DLLs or EXEs

I think I might have found a minor bug in CVI 9.0 FDS... 

 

If I RTFM first (using the CVI 9.0 help), and search for the page entitled "Version Info Dialog Box", I see that it says that "LabWindows/CVI increments fields only for successful builds of the Release Configuration."  Well, it seems to do it for successful builds of both the release and the debug configuration when batch-building.  That is a good thing, in my opinion, because I sometimes deploy the debug DLLs for a new roll-out of an ATE system for the first time, and versioning them always helps.

 

But the "bug" I think I might have found (or maybe I should call it undefined behavior since it is not defined in the help file Smiley Wink ) is that when I do a batch build, the build number increments for both the release and the debug DLLs on my project, and the numbers are offset-by-one, and the debug version is higher than the release version...

 

Example, if let's say, in my taget settings, under "Version Info...", I say that the file version and the product version should be "0.1.0.20+".  Then if I do a batch build and check off to build both the release and the debug in the batch build pop-up, the release DLL will have the version 0.1.0.20, but the debug DLL will have 0.1.0.21.  I just checked and it does it for EXEs, too.

 

Manual building of the project (right-click on the project name in the upper left pane of the CVI IDE) does not auto-increment the build number of the DLL while in debug configuration, as documented in the CVI help.  Manual building of the project in release mode does auto-increment the build number of the DLL, also as documented in the CVI help.  So the issue must be in the batch build tool, and hopefully easy to fix.

 

So,  if the NI development crew is able to determine that this really is a bug (and especially if it has been in the IDE prior to CVI 9.0), does this mean I earned a free NI CVI T-shirt or something?

 

It's only fair...   Smiley Happy

 

JB

--
To whom it may concern: My alias is also my nickname, I've had it since I was a (very) skinny basketball-playing teen. OK, so I've got a 38 inch waist now, but my hometown friends haven't shaken that appellation for me. I trust that you will someday be OK with that alias, as I have been with that nickname.
0 Kudos
Message 1 of 4
(3,833 Views)

be it a bug or not, i don't care, here is your t-shirt:

 

  /--___--\
 /         \
/ /| N I |\ \
\/ | CVI | \/
   |     |
   |_____|


i hope it will fit you well.
Message 2 of 4
(3,807 Views)

Looks more like a baseball shirt than a T-shirt.  Can you shorten the sleeves a bit?  And what's with those stray flaps on the shoulders?  Are those like the ones that they put on fancy raincoats?  Even though the T-shirt is free, I can still be picky!

 

JB

--
To whom it may concern: My alias is also my nickname, I've had it since I was a (very) skinny basketball-playing teen. OK, so I've got a 38 inch waist now, but my hometown friends haven't shaken that appellation for me. I trust that you will someday be OK with that alias, as I have been with that nickname.
0 Kudos
Message 3 of 4
(3,805 Views)

I suspect what is happening is that the release build is being built first, and on success the version number is bumped. This would result in a subsequent (within the batch) build of the debug binary to have a version number one larger than that of the release binary. Whatever's going on, we'll look into it further and fix it appropriately. As far as a t-shirt goes, it looks like dummy_decoy already supplied you with a lovely one. Also, I'm fairly certain that you were able to snag one of those black CVI shirts at NI Week a year a couple years back, and we haven't had any new shirts since.

 

Mert A.

National Instruments

0 Kudos
Message 4 of 4
(3,783 Views)