[PATCH] D107292: [clang] adds warning to alert user when they use alternative tokens as references

Tue Aug 3 08:42:06 PDT 2021

dblaikie added a comment.

In D107292#2921901 <https://reviews.llvm.org/D107292#2921901>, @aaron.ballman wrote:

> In D107292#2920774 <https://reviews.llvm.org/D107292#2920774>, @dblaikie wrote:
>
>> In D107292#2920746 <https://reviews.llvm.org/D107292#2920746>, @cjdb wrote:
>>
>>> In D107292#2920637 <https://reviews.llvm.org/D107292#2920637>, @dblaikie wrote:
>>>
>>>> Not a huge objection - but minor quandry: What's the motivation for this patch? I don't know of any codebase that encourages/uses the alternative tokens and I wonder if adding more usability to them is a worthwhile investment in clang's codebase complexity, etc.
>>>
>>> There are codebases that use them (all of my non-Google, non-LLVM code does, for example, and I'm not the sole user: just a loud one who's also in a position to patch tooling).
>>
>> Ah, any pointers to large open source projects that use this?
>
> https://codesearch.isocpp.org/cgi-bin/cgi_ppsearch?q=bitand&search=Search
>
> (Searching for 'and' is a bit less useful because of how much it shows up in assembly, comments, etc.)

Ah, cool. Only case I could find there (that wasn't C code or compiler test cases) was something called FuzzyLite (which looks like it hasn't been touched in 4 years or so).

I don't fundamentally object to this now it's being proposed as a clang-tidy thing - bar should be low/easy for experiments, getting user experience, adoption, etc.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107292/new/

https://reviews.llvm.org/D107292