@scpeters Thanks a lot for the review and suggestions!

I took a look at cpplint/cpplint#214. It contains two commits:

1. replacing the deprecated sre_compile usage with re (same as the change in this PR)
2. removes cpplint's own regex cache and relies on Python re's internal cache instead.

There was some discussion in that PR about the second change potentially causing a performance regression, so I did a small local test on gz-sim.

I ran make codecheck 10 times for both versions:

• Current version before dropping sre_compile: 24.3 s average
• Version with both commits from Drop deprecated sre_compile usage cpplint/cpplint#214: 27.6 s average

So there is some slowdown, although I think it is still within a reasonable range for codecheck.

For reference, I ran the test on the following system:

• CPU: Intel Core Ultra 5 125H (14 cores / 18 threads)
• RAM: 30 GiB
• OS: Ubuntu 24.04.4 LTS
• Kernel: 6.8.0-138-generic

Regarding importing a newer cpplint version, I agree that this would probably be a better long-term solution. I noticed that upstream cpplint is now at version 2.0.2, so this may be a good opportunity to update it.

I did a quick comparison and tried running the newer version locally. The changes are fairly large. Some are only formatting changes, but there are also behavioral differences in the checks. I made a few small adjustments to the areas that I thought needed updating, but I still wasn't able to get gz-sim to produce results comparable to the current version. I think this would need a more careful and comprehensive review before updating it.

Given that, perhaps it would make more sense to close this PR and open a separate PR to update cpplint directly to 2.0.2.

@scpeters what do you think would be the preferred direction here?

@scpeters Thanks a lot for the review and suggestions!

I took a look at cpplint/cpplint#214. It contains two commits:

1. replacing the deprecated sre_compile usage with re (same as the change in this PR)
2. removes cpplint's own regex cache and relies on Python re's internal cache instead.

There was some discussion in that PR about the second change potentially causing a performance regression, so I did a small local test on gz-sim.

I ran make codecheck 10 times for both versions:

• Current version before dropping sre_compile: 24.3 s average
• Version with both commits from Drop deprecated sre_compile usage cpplint/cpplint#214: 27.6 s average

So there is some slowdown, although I think it is still within a reasonable range for codecheck.

yeah, this is a small change; I think it makes sense to just merge this and open a separate issue to evaluate updating the cpplint version

thanks for the contribution!

I'm going to let this live on main for a week before backporting