Test the possibility of using problem matchers for errors/warnings

[lefticus]

No description provided.

[codecov-commenter]

Codecov Report

Merging #195 (cb1b553) into main (a85fe3d) will decrease coverage by 12.35%.
The diff coverage is 0.00%.

@@             Coverage Diff             @@
##             main     #195       +/-   ##
===========================================
- Coverage   80.00%   67.64%   -12.36%     
===========================================
  Files           3        3               
  Lines          35       34        -1     
  Branches       18        0       -18     
===========================================
- Hits           28       23        -5     
- Misses          6       11        +5     
+ Partials        1        0        -1     

Flag	Coverage Δ
Linux	?
Windows	67.64% <0.00%> (-12.36%)	⬇️
macOS	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/main.cpp	21.42% <0.00%> (-24.73%)	⬇️
test/tests.cpp	100.00% <0.00%> (ø)
test/constexpr_tests.cpp	100.00% <0.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a85fe3d...cb1b553. Read the comment docs.

[aminya]

Does clang uses the same report style as gcc?
Related to aminya/setup-cpp#34

[lefticus]

Does clang uses the same report style as gcc? Related to aminya/setup-cpp#34

It is very similar, but isn't matching.

I'm pretty confused, TBH

clang, which doesn't match:

https://github.com/cpp-best-practices/cpp_starter_project/runs/5164418830?check_suite_focus=true#step:6:20

/home/runner/work/cpp_starter_project/cpp_starter_project/src/main.cpp:47:13: warning: variable 'j' is uninitialized when used here [-Wuninitialized]
    int l = j + k;

clang-tidy, which does match:

https://github.com/cpp-best-practices/cpp_starter_project/runs/5164419289?check_suite_focus=true#step:6:23

/home/runner/work/cpp_starter_project/cpp_starter_project/src/main.cpp:43:10: error: 42 is a magic number; consider replacing it with a named constant [cppcoreguidelines-avoid-magic-numbers,readability-magic-numbers,-warnings-as-errors]

gcc, which does match:

https://github.com/cpp-best-practices/cpp_starter_project/runs/5164419322?check_suite_focus=true#step:6:16

/Users/runner/work/cpp_starter_project/cpp_starter_project/src/main.cpp:47:15: warning: 'j' may be used uninitialized [-Wmaybe-uninitialized]
   47 |     int l = j + k;
      |             ~~^~~

[lefticus]

I just ran them through a regex tester and don't see what it's not matching

https://github.com/aminya/setup-cpp/pull/35/files#diff-1eb832d204888d4e54321209ab1d61281ea359429af65207b2420b4bdd9f8358R7

[aminya]

The gcc matcher is only enabled if the compiler is gcc. If the regex is the same as clang, I can add the same thing to clang and clang-tidy.

[aminya]

Here it is for llvm:
aminya/setup-cpp#36

[lefticus]

The gcc matcher is only enabled if the compiler is gcc. If the regex is the same as clang, I can add the same thing to clang and clang-tidy.

clang-tidy is automatically getting picked up, since it's not being run as its own compiler here.

it's on my to-do list to have cppcheck output in the appropriate gcc/msvc so it shows up in IDEs as expected, which should also make it get picked up here.

[lefticus]

aminya/project_options#66 to make sure cppcheck is picked up as expected.

[lefticus]

Confirmed @aminya if we merge aminya/project_options#66 then we get all of the warnings now picked up and this PR becomes an interesting artifact

https://github.com/cpp-best-practices/cpp_starter_project/pull/195/files

[aminya]

Awesome! 🎉.

I wish I could add similar support for other CI services like GitLab. I don't know if other services have such similar capabilities.