Repository navigation
Support a runtime clamped_value<float> for C++<20 #1594
Description
Currently the clamped value utility supports runtime specialization of floats similar to the integral type, however, not for versions that are less than version 20 due to the runtime detection behavior of checking Min and Max non-template types.
Activity
drewr95 commentedon Sep 8, 2026
@jwellbelove @rolandreichweinbmw Any suggestions on what this specialization should be called? I think it should be compiled (ifdef'd out) if >20. I use a lot of floats, especially for control where constantly normalizing on range
-1.0F...1.0For0...1.0Fso this functionality is highly desirable.rolandreichweinbmw commentedon Sep 8, 2026
By C++ overload, the specialization does not need to be called differently?
drewr95 commentedon Sep 8, 2026
@rolandreichweinbmw hmm that was not obvious to me. I'll double check. Thank you 😁