Implement Joining_Type property (#4599)

CHANGELOG.md

@@ -12,6 +12,7 @@
   - `icu_properties`
     - Add `Aran` script code (https://github.com/unicode-org/icu4x/pull/4426)
     - Mark `BidiClassAdapter::new` as const (https://github.com/unicode-org/icu4x/pull/4584)
+    - Implement Joining_Type property (https://github.com/unicode-org/icu4x/pull/4599)
   - `icu_segmenter`
     - Fix Unicode 15.0 line breaking (https://github.com/unicode-org/icu4x/pull/4389)
 - Data model and providers

components/properties/src/lib.rs

@@ -92,7 +92,8 @@ mod trievalue;
 
 pub use props::{
     BidiClass, CanonicalCombiningClass, EastAsianWidth, GeneralCategory, GeneralCategoryGroup,
-    GraphemeClusterBreak, IndicSyllabicCategory, LineBreak, Script, SentenceBreak, WordBreak,
+    GraphemeClusterBreak, IndicSyllabicCategory, JoiningType, LineBreak, Script, SentenceBreak,
+    WordBreak,
 };
 
 /// Module for working with the names of property values

components/properties/src/maps.rs

@@ -600,3 +600,28 @@ make_map_property! {
     pub const indic_syllabic_category => SINGLETON_PROPS_INSC_V1;
     pub fn load_indic_syllabic_category();
 }
+
+make_map_property! {
+    property: "Joining_Type";
+    marker: JoiningTypeProperty;
+    value: crate::JoiningType;
+    keyed_data_marker: JoiningTypeV1Marker;
+    func:
+    /// Return a [`CodePointMapDataBorrowed`] for the Joining_Type Unicode enumerated
+    /// property. See [`JoiningType`].
+    ///
+    /// ✨ *Enabled with the `compiled_data` Cargo feature.*
+    ///
+    /// [📚 Help choosing a constructor](icu_provider::constructors)
+    ///
+    /// # Example
+    ///
+    /// ```
+    /// use icu::properties::{maps, JoiningType};
+    ///
+    /// assert_eq!(maps::joining_type().get('ؠ'), JoiningType::DualJoining); // U+0620: Arabic Letter Kashmiri Yeh
+    /// assert_eq!(maps::joining_type().get('𐫍'), JoiningType::LeftJoining); // U+10ACD: Manichaean Letter Heth
+    /// ```
+    pub const joining_type => SINGLETON_PROPS_JT_V1;
+    pub fn load_joining_type();
+}

components/properties/src/props.rs

@@ -2364,3 +2364,95 @@ impl_value_getter! {
         pub fn get_enum_to_long_name_mapper() / enum_to_long_name_mapper() -> PropertyEnumToValueNameLinearMapper / PropertyEnumToValueNameLinearMapperBorrowed;
     }
 }
+/// Enumerated property Joining_Type.
+/// See Section 9.2, Arabic Cursive Joining in The Unicode Standard for the summary of
+/// each property value.
+///
+/// The numeric value is compatible with `UJoiningType` in ICU4C.
+#[derive(Copy, Clone, Debug, Eq, PartialEq, Ord, PartialOrd)]
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "datagen", derive(databake::Bake))]
+#[cfg_attr(feature = "datagen", databake(path = icu_properties))]
+#[allow(clippy::exhaustive_structs)] // newtype
+#[repr(transparent)]
+#[zerovec::make_ule(JoiningTypeULE)]
+pub struct JoiningType(pub u8);
+
+#[allow(missing_docs)] // These constants don't need individual documentation.
+#[allow(non_upper_case_globals)]
+impl JoiningType {
+    pub const NonJoining: JoiningType = JoiningType(0); // name="U"
+    pub const JoinCausing: JoiningType = JoiningType(1); // name="C"
+    pub const DualJoining: JoiningType = JoiningType(2); // name="D"
+    pub const LeftJoining: JoiningType = JoiningType(3); // name="L"
+    pub const RightJoining: JoiningType = JoiningType(4); // name="R"
+    pub const Transparent: JoiningType = JoiningType(5); // name="T"
+}
+
+impl_value_getter! {
+    markers: JoiningTypeNameToValueV1Marker / SINGLETON_PROPNAMES_FROM_JT_V1, JoiningTypeValueToShortNameV1Marker / SINGLETON_PROPNAMES_TO_SHORT_LINEAR_JT_V1, JoiningTypeValueToLongNameV1Marker / SINGLETON_PROPNAMES_TO_LONG_LINEAR_JT_V1;
+    impl JoiningType {
+        /// Return a [`PropertyValueNameToEnumMapper`], capable of looking up values
+        /// from strings for the `Joining_Type` enumerated property.
+        ///
+        /// ✨ *Enabled with the `compiled_data` Cargo feature.*
+        ///
+        /// [📚 Help choosing a constructor](icu_provider::constructors)
+        ///
+        /// # Example
+        ///
+        /// ```
+        /// use icu::properties::JoiningType;
+        ///
+        /// let lookup = JoiningType::name_to_enum_mapper();
+        /// // short name for value
+        /// assert_eq!(lookup.get_strict("T"), Some(JoiningType::Transparent));
+        /// assert_eq!(lookup.get_strict("D"), Some(JoiningType::DualJoining));
+        /// // long name for value
+        /// assert_eq!(lookup.get_strict("Join_Causing"), Some(JoiningType::JoinCausing));
+        /// assert_eq!(lookup.get_strict("Non_Joining"), Some(JoiningType::NonJoining));
+        /// // name has incorrect casing
+        /// assert_eq!(lookup.get_strict("LEFT_JOINING"), None);
+        /// // loose matching of name
+        /// assert_eq!(lookup.get_loose("LEFT_JOINING"), Some(JoiningType::LeftJoining));
+        /// // fake property
+        /// assert_eq!(lookup.get_strict("Inner_Joining"), None);
+        /// ```
+        pub fn get_name_to_enum_mapper() / name_to_enum_mapper();
+        /// Return a [`PropertyEnumToValueNameLinearMapper`], capable of looking up short names
+        /// for values of the `Joining_Type` enumerated property.
+        ///
+        /// ✨ *Enabled with the `compiled_data` Cargo feature.*
+        ///
+        /// [📚 Help choosing a constructor](icu_provider::constructors)
+        ///
+        /// # Example
+        ///
+        /// ```
+        /// use icu::properties::JoiningType;
+        ///
+        /// let lookup = JoiningType::enum_to_short_name_mapper();
+        /// assert_eq!(lookup.get(JoiningType::JoinCausing), Some("C"));
+        /// assert_eq!(lookup.get(JoiningType::LeftJoining), Some("L"));
+        /// ```
+        pub fn get_enum_to_short_name_mapper() / enum_to_short_name_mapper() -> PropertyEnumToValueNameLinearMapper / PropertyEnumToValueNameLinearMapperBorrowed;
+        /// Return a [`PropertyEnumToValueNameLinearMapper`], capable of looking up long names
+        /// for values of the `Joining_Type` enumerated property.
+        ///
+        /// ✨ *Enabled with the `compiled_data` Cargo feature.*
+        ///
+        /// [📚 Help choosing a constructor](icu_provider::constructors)
+        ///
+        /// # Example
+        ///
+        /// ```
+        /// use icu::properties::JoiningType;
+        ///
+        /// let lookup = JoiningType::enum_to_long_name_mapper();
+        /// assert_eq!(lookup.get(JoiningType::Transparent), Some("Transparent"));
+        /// assert_eq!(lookup.get(JoiningType::NonJoining), Some("Non_Joining"));
+        /// assert_eq!(lookup.get(JoiningType::RightJoining), Some("Right_Joining"));
+        /// ```
+        pub fn get_enum_to_long_name_mapper() / enum_to_long_name_mapper() -> PropertyEnumToValueNameLinearMapper / PropertyEnumToValueNameLinearMapperBorrowed;
+    }
+}

components/properties/src/provider.rs

@@ -58,6 +58,7 @@ const _: () = {
     icu_properties_data::impl_propnames_from_gc_v1!(Baked);
     icu_properties_data::impl_propnames_from_gcm_v1!(Baked);
     icu_properties_data::impl_propnames_from_insc_v1!(Baked);
+    icu_properties_data::impl_propnames_from_jt_v1!(Baked);
     icu_properties_data::impl_propnames_from_lb_v1!(Baked);
     icu_properties_data::impl_propnames_from_sb_v1!(Baked);
     icu_properties_data::impl_propnames_from_sc_v1!(Baked);
@@ -67,6 +68,7 @@ const _: () = {
     icu_properties_data::impl_propnames_to_long_linear_gc_v1!(Baked);
     icu_properties_data::impl_propnames_to_long_linear_gcb_v1!(Baked);
     icu_properties_data::impl_propnames_to_long_linear_insc_v1!(Baked);
+    icu_properties_data::impl_propnames_to_long_linear_jt_v1!(Baked);
     icu_properties_data::impl_propnames_to_long_linear_lb_v1!(Baked);
     icu_properties_data::impl_propnames_to_long_linear_sb_v1!(Baked);
     icu_properties_data::impl_propnames_to_long_linear_sc_v1!(Baked);
@@ -77,6 +79,7 @@ const _: () = {
     icu_properties_data::impl_propnames_to_short_linear_gc_v1!(Baked);
     icu_properties_data::impl_propnames_to_short_linear_gcb_v1!(Baked);
     icu_properties_data::impl_propnames_to_short_linear_insc_v1!(Baked);
+    icu_properties_data::impl_propnames_to_short_linear_jt_v1!(Baked);
     icu_properties_data::impl_propnames_to_short_linear_lb_v1!(Baked);
     icu_properties_data::impl_propnames_to_short_linear_sb_v1!(Baked);
     icu_properties_data::impl_propnames_to_short_linear_wb_v1!(Baked);
@@ -133,6 +136,7 @@ const _: () = {
     icu_properties_data::impl_props_idst_v1!(Baked);
     icu_properties_data::impl_props_insc_v1!(Baked);
     icu_properties_data::impl_props_join_c_v1!(Baked);
+    icu_properties_data::impl_props_jt_v1!(Baked);
     icu_properties_data::impl_props_lb_v1!(Baked);
     icu_properties_data::impl_props_loe_v1!(Baked);
     icu_properties_data::impl_props_lower_v1!(Baked);
@@ -896,6 +900,16 @@ expand!(
             "InSC",
             IndicSyllabicCategory
         ),
+        (
+            JoiningTypeV1Marker,
+            JoiningTypeNameToValueV1Marker,
+            (
+                linear: JoiningTypeValueToShortNameV1Marker,
+                JoiningTypeValueToLongNameV1Marker
+            ),
+            "jt",
+            JoiningType
+        ),
         // note: the names key for the GCM mask is handled above
     )
 );

components/properties/src/trievalue.rs

@@ -8,7 +8,8 @@ use crate::provider::bidi_data::{
 use crate::script::ScriptWithExt;
 use crate::{
     BidiClass, CanonicalCombiningClass, EastAsianWidth, GeneralCategory, GeneralCategoryGroup,
-    GraphemeClusterBreak, IndicSyllabicCategory, LineBreak, Script, SentenceBreak, WordBreak,
+    GraphemeClusterBreak, IndicSyllabicCategory, JoiningType, LineBreak, Script, SentenceBreak,
+    WordBreak,
 };
 use core::convert::TryInto;
 use core::num::TryFromIntError;
@@ -246,3 +247,15 @@ impl TrieValue for MirroredPairedBracketData {
         Self::try_from(i)
     }
 }
+
+impl TrieValue for JoiningType {
+    type TryFromU32Error = TryFromIntError;
+
+    fn try_from_u32(i: u32) -> Result<Self, Self::TryFromU32Error> {
+        u8::try_from(i).map(Self)
+    }
+
+    fn to_u32(self) -> u32 {
+        u32::from(self.0)
+    }
+}

ffi/capi/src/properties_maps.rs

@@ -225,6 +225,18 @@ pub mod ffi {
                 provider,
             )?))
         }
+
+        #[diplomat::rust_link(icu::properties::maps::joining_type, Fn)]
+        #[diplomat::rust_link(icu::properties::maps::load_joining_type, Fn, hidden)]
+        pub fn load_joining_type(
+            provider: &ICU4XDataProvider,
+        ) -> Result<Box<ICU4XCodePointMapData8>, ICU4XError> {
+            Ok(convert_8(call_constructor_unstable!(
+                maps::joining_type [r => Ok(r.static_to_owned())],
+                maps::load_joining_type,
+                provider,
+            )?))
+        }
     }
 
     #[diplomat::opaque]