JonasToth added inline comments.
================
Comment at: test/clang-tidy/abseil-duration-comparison.cpp:89
+ // CHECK-FIXES: d1 > d2;
+
+ // Check against the LHS
----------------
JonasToth wrote:
> What would happen for a type, that can implicitly convert to a duration or
> double/int.
>
> ```
> struct MetaBenchmarkResults {
> int64_t RunID;
> absl::Duration D;
>
> operator absl::Duration() { return D; }
> };
>
> auto R = MetaBenchmarkResults { /* Foo */ };
> bool WithinReason = Threshold < R;
> ```
>
> What is the correct behaviour there? I think this should be diagnosed too.
Hmmmm, I fiddled around in godbolt, and i think there is nothing to trick
around with conversions, as they wont compile.
If you find something along the lines, please tell me. But i guess this is a
non-issue.
https://godbolt.org/z/gqhhHo
================
Comment at: test/clang-tidy/abseil-duration-comparison.cpp:153
+ // Other abseil-duration checks folded into this one
+ b = static_cast<double>(5) > absl::ToDoubleSeconds(d1);
+ // CHECK-MESSAGES: [[@LINE-1]]:7: warning: perform duration comparison in
the duration domain [abseil-duration-comparison]
----------------
Please test the other casts you match on as well (especially c-style)
Repository:
rCTE Clang Tools Extra
https://reviews.llvm.org/D54737
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
