converted fractional_part_rounding_thresholds to use an array of u32s#4719

Open

Cazadorro wants to merge 14 commits intofmtlib:mainfrom

Cazadorro:cuda_agnostic_fractional_part_rounding_thresholds

Commits on Mar 18, 2026

converted formafractional_part_rounding_thresholds to use an array of u32s instead of a unicode literal character array for compatibliity with CUDA and non CUDA platforms
Cazadorro
committed

Commits on Mar 24, 2026

fixed clang format issue
Cazadorro
committed

Commits on Mar 28, 2026

Merge branch 'master' of https://github.com/Cazadorro/fmt into cuda_agnostic_fractional_part_rounding_thresholds
Cazadorro
committed
changed naming of rounding thresholds, and changed static constexpr declaration of thresholds to inline constexpr, this should mean same address for every translation unit see: https://stackoverflo…
Cazadorro
committed

Commits on Apr 2, 2026

Commits on Apr 4, 2026

Commits on Apr 5, 2026

should get rid of the implicit conversion warning due to u16's always converting to i32 when any operation applied
Cazadorro
committed