Performance tweaks

[edding3000]

what

Range based for loops with references are already used in this project, but in a few places not.
I am not sure why. I changed this code locations and executed unit tests.
Also use '*' for pointers when using auto keyword, makes it more readable.

why

Improve performance a little bit.

questions

"const auto&" could also be used in many places. Is there a reason why it is not used?

[edding3000]

I was not sure if there was any restriction not to use const, thats why i asked. I could extend the PR with more positions for "const auto&" if you like.

[airween]

I was not sure if there was any restriction not to use const, thats why i asked. I could extend the PR with more positions for "const auto&" if you like.

Sure, go! If the tests will be passed, I'll be happy :).

Btw you could review the tests too, may be you will find some lack there too (for eg. you're checking the code...)

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[airween]

Hi @edding3000,

sorry for the long delay, it seems like this PR is ready to merge. Just one question: could you merge the current v3/master into this one?

[airween]

@edding3000,

also please take a look at the few request changes.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[edding3000]

@airween I merged v3/master. cppcheck fails now. Can you see the reason?

[airween]

@airween I merged v3/master. cppcheck fails now. Can you see the reason?

Yes:

2026-02-17T18:48:51.1404560Z warning: src/run_time_string.cc,54,style,functionConst,The member function 'modsecurity::RunTimeString::evaluate' can be made a const function. Making this function 'const' should not cause compiler errors. Even though the function can be made const function technically it may not make sense conceptually. Think about your design and the task of the function first - is it a function that must not change object internal state?

I think cppcheck realized that you changed the loop's variable to const, and figured out the function can't change the object (I don't see the relationship what this const triggered this error), so it expects to make the fn to const - but warns us that it might be wrong.

So what you can try now:

• change the function signature to const: std::string RunTimeString::evaluate(Transaction *t, Rule *r) const {
• if the build will be failed, rollback that change and add an inline suppression to that line: std::string RunTimeString::evaluate(Transaction *t, Rule *r) { // cppcheck-suppress functionConst
• in this case, please add this note to the PR's description

Do you have cppcheck on your local dev environment? It's important to use the latest release, and on Linux only Debian SID contains that version. Here you can see how do we use it in Docker.