[HuzaifaChaudary]

CaloriePerHourSquareMeterDegreeCelsius had the kcal/… abbreviations , and since it also has the Kilo prefix , the generated kilocalorie unit got kkcal/… :

CaloriesPerHourSquareMeterDegreeCelsius=kcal/(h·m²·°C),...
KilocaloriesPerHourSquareMeterDegreeCelsius=kkcal/(h·m²·°C),...

so HeatTransferCoefficient.Parse("1 kcal/(h·m²·°C)") returned 1 calorie per hour , 0.00116 W/(m²·K) instead of 1.163 , and a kilocalorie value printed as kkcal/(h·m²·°C) . the conversion factors were right , only the names were off by the prefix .

the calorie unit now has cal/… and the generated kilocalorie one kcal/… , the same pattern HeatFlux uses for kcal/(h·m²) . the change is in HeatTransferCoefficient.json , the rest is generate-code output . the regenerated HeatTransferCoefficientTestsBase had "4.2 kkcal/(h·m²·°C)" as the expected kilocalorie string before , with the old resource file 84 of its 266 tests fail , with the new one all pass . UnitsNet.Tests on net10.0 , macOS : 52558 passed , the 2 FeetInchesTests de-CH failures happen on master too ( ICU group separator ) .

found by converting one of every unit to its base unit with UnitsNet's own formulas and with pint , then looking at what disagreed . the other differences were definitions , Year365 vs pint's 365.25 days and the like , this was the one real bug . I also checked that each FromBaseToUnitFunc inverts its FromUnitToBaseFunc for the 758 units defined in the json files , they all do , up to floating point rounding .

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.