feat: value_cast<Unit, Representation> added

CHANGELOG.md

@@ -6,6 +6,7 @@
 
 - feat: `fma`, `isfinite`, `isinf`, and `isnan` math function added by @NAThompson
 - feat: `quantity_point` support added for `quantity_cast` and `value_cast`
+- feat: `value_cast<Unit, Representation>` added
 - (!) refactor: `zero_Fahrenheit` renamed to `zeroth_Fahrenheit`
 - refactor: math functions constraints refactored
 - fix: `QuantityLike` conversions required `Q::rep` instead of using one provided by `quantity_like_traits`

docs/users_guide/examples/avg_speed.md

@@ -83,7 +83,7 @@ Next, let's do the same for integral and floating-point representations, but thi
 using US Customary units:
 
 ```cpp title="avg_speed.cpp" linenums="54"
---8<-- "example/avg_speed.cpp:93:125"
+--8<-- "example/avg_speed.cpp:93:124"
 ```
 
 One important difference here is the fact that as it is not possible to make a lossless conversion
@@ -108,8 +108,8 @@ Please note how the first and third results get truncated using integral represe
 
 In the end, we repeat the scenario for CGS units:
 
-```cpp title="avg_speed.cpp" linenums="87"
---8<-- "example/avg_speed.cpp:127:157"
+```cpp title="avg_speed.cpp" linenums="86"
+--8<-- "example/avg_speed.cpp:126:155"
 ```
 
 Again, we observe `value_cast` being used in the same places and consistent truncation errors
@@ -129,6 +129,6 @@ Average speed of a car that makes 2.2e+07 cm in 7200 s is 110 km/h.
 
 The example file ends with a simple `main()` function:
 
-```cpp title="avg_speed.cpp" linenums="118"
---8<-- "example/avg_speed.cpp:159:"
+```cpp title="avg_speed.cpp" linenums="116"
+--8<-- "example/avg_speed.cpp:157:"
 ```

docs/users_guide/framework_basics/value_conversions.md

@@ -77,3 +77,76 @@ quantity<si::metre, int> q3 = value_cast<int>(3.14 * m);
     It is often fine to use an integral as a representation type, but in general, floating-point
     types provide better precision and are privileged in the library as they are considered
     to be value-preserving.
+
+In some cases, a unit and a representation type should be changed simultaneously. Moreover,
+sometimes, the order of doing those operations matters. In such cases, the library provides
+the `value_cast<U, Rep>(q)` which always returns the most precise result:
+
+=== "C++23"
+
+    ```cpp
+    inline constexpr struct dim_currency : base_dimension<"$"> {} dim_currency;
+    inline constexpr struct currency : quantity_spec<dim_currency> {} currency;
+
+    inline constexpr struct us_dollar : named_unit<"USD", kind_of<currency>> {} us_dollar;
+    inline constexpr struct scaled_us_dollar : named_unit<"USD_s", mag_power<10, -8> * us_dollar> {} scaled_us_dollar;
+
+    namespace unit_symbols {
+
+    inline constexpr auto USD = us_dollar;
+    inline constexpr auto USD_s = scaled_us_dollar;
+
+    }  // namespace unit_symbols
+
+    using Price = quantity<currency[us_dollar], double>;
+    using Scaled = quantity<currency[scaled_us_dollar], std::int64_t>;
+
+    Price price = 12.95 * USD;
+    Scaled spx = value_cast<USD_s, std::int64_t>(price);
+    ```
+
+=== "C++20"
+
+    ```cpp
+    inline constexpr struct dim_currency : base_dimension<"$"> {} dim_currency;
+    inline constexpr struct currency : quantity_spec<currency, dim_currency> {} currency;
+
+    inline constexpr struct us_dollar : named_unit<"USD", kind_of<currency>> {} us_dollar;
+    inline constexpr struct scaled_us_dollar : named_unit<"USD_s", mag_power<10, -8> * us_dollar> {} scaled_us_dollar;
+
+    namespace unit_symbols {
+
+    inline constexpr auto USD = us_dollar;
+    inline constexpr auto USD_s = scaled_us_dollar;
+
+    }  // namespace unit_symbols
+
+    using Price = quantity<currency[us_dollar], double>;
+    using Scaled = quantity<currency[scaled_us_dollar], std::int64_t>;
+
+    Price price = 12.95 * USD;
+    Scaled spx = value_cast<USD_s, std::int64_t>(price);
+    ```
+
+=== "Portable"
+
+    ```cpp
+    inline constexpr struct dim_currency : base_dimension<"$"> {} dim_currency;
+    QUANTITY_SPEC(currency, dim_currency);
+
+    inline constexpr struct us_dollar : named_unit<"USD", kind_of<currency>> {} us_dollar;
+    inline constexpr struct scaled_us_dollar : named_unit<"USD_s", mag_power<10, -8> * us_dollar> {} scaled_us_dollar;
+
+    namespace unit_symbols {
+
+    inline constexpr auto USD = us_dollar;
+    inline constexpr auto USD_s = scaled_us_dollar;
+
+    }  // namespace unit_symbols
+
+    using Price = quantity<currency[us_dollar], double>;
+    using Scaled = quantity<currency[scaled_us_dollar], std::int64_t>;
+
+    Price price = 12.95 * USD;
+    Scaled spx = value_cast<USD_s, std::int64_t>(price);
+    ```

example/avg_speed.cpp

@@ -118,8 +118,7 @@ void example()
     // conversion from a floating-point to an integral type is a truncating one so an explicit cast is needed
     // also it is not possible to make a lossless conversion of miles to meters on an integral type
     // (explicit cast needed)
-    print_result(distance, duration,
-                 fixed_int_si_avg_speed(value_cast<int>(distance.force_in(m)), value_cast<int>(duration)));
+    print_result(distance, duration, fixed_int_si_avg_speed(value_cast<m, int>(distance), value_cast<int>(duration)));
     print_result(distance, duration, fixed_double_si_avg_speed(distance, duration));
     print_result(distance, duration, avg_speed(distance, duration));
   }
@@ -148,8 +147,7 @@ void example()
     // conversion from a floating-point to an integral type is a truncating one so an explicit cast is needed
     // it is not possible to make a lossless conversion of centimeters to meters on an integral type
     // (explicit cast needed)
-    print_result(distance, duration,
-                 fixed_int_si_avg_speed(value_cast<int>(distance.force_in(m)), value_cast<int>(duration)));
+    print_result(distance, duration, fixed_int_si_avg_speed(value_cast<m, int>(distance), value_cast<int>(duration)));
 
     print_result(distance, duration, fixed_double_si_avg_speed(distance, duration));
     print_result(distance, duration, avg_speed(distance, duration));

src/core/include/mp-units/bits/value_cast.h

@@ -75,6 +75,26 @@ template<Representation ToRep, typename Q>
   return detail::sudo_cast<quantity<std::remove_reference_t<Q>::reference, ToRep>>(std::forward<Q>(q));
 }
 
+/**
+ * @brief Explicit cast of a quantity's unit and representation type
+ *
+ * Implicit conversions between quantities of different types are allowed only for "safe"
+ * (e.g. non-truncating) conversion. In truncating cases an explicit cast have to be used.
+ *
+ * auto q = value_cast<us, int>(1.23 * ms);
+ *
+ * @tparam ToRep a representation type to use for a target quantity
+ */
+template<Unit auto ToU, Representation ToRep, typename Q>
+  requires Quantity<std::remove_cvref_t<Q>> && (convertible(std::remove_reference_t<Q>::reference, ToU)) &&
+           RepresentationOf<ToRep, std::remove_reference_t<Q>::quantity_spec.character> &&
+           std::constructible_from<ToRep, typename std::remove_reference_t<Q>::rep>
+[[nodiscard]] constexpr Quantity auto value_cast(Q&& q)
+{
+  using q_type = std::remove_reference_t<Q>;
+  return detail::sudo_cast<quantity<q_type::quantity_spec[ToU], ToRep>>(std::forward<Q>(q));
+}
+
 /**
  * @brief Explicit cast of a quantity point's unit
  *
@@ -112,4 +132,24 @@ template<Representation ToRep, typename QP>
   return {value_cast<ToRep>(std::forward<QP>(qp).quantity_from_origin_is_an_implementation_detail_), qp.point_origin};
 }
 
+/**
+ * @brief Explicit cast of a quantity's unit and representation type
+ *
+ * Implicit conversions between quantities of different types are allowed only for "safe"
+ * (e.g. non-truncating) conversion. In truncating cases an explicit cast have to be used.
+ *
+ * auto q = value_cast<us, int>(1.23 * ms);
+ *
+ * @tparam ToRep a representation type to use for a target quantity
+ */
+template<Unit auto ToU, Representation ToRep, typename QP>
+  requires QuantityPoint<std::remove_cvref_t<QP>> && (convertible(std::remove_reference_t<QP>::reference, ToU)) &&
+           RepresentationOf<ToRep, std::remove_reference_t<QP>::quantity_spec.character> &&
+           std::constructible_from<ToRep, typename std::remove_reference_t<QP>::rep>
+[[nodiscard]] constexpr QuantityPoint auto value_cast(QP&& qp)
+{
+  return quantity_point{value_cast<ToU, ToRep>(std::forward<QP>(qp).quantity_from_origin_is_an_implementation_detail_),
+                        qp.point_origin};
+}
+
 }  // namespace mp_units