[char_property] Add NumericCharProperty

unic/ucd/normal/src/canonical_combining_class.rs

@@ -17,7 +17,7 @@
 
 use std::fmt;
 
-use unic_utils::{CharDataTable, CharProperty};
+use unic_utils::{CharDataTable, CharProperty, NumericCharProperty};
 
 
 /// Represents *Canonical_Combining_Class* property of a Unicode character.
@@ -27,6 +27,21 @@ use unic_utils::{CharDataTable, CharProperty};
 pub struct CanonicalCombiningClass(u8);
 
 
+impl CharProperty for CanonicalCombiningClass {
+    fn of(ch: char) -> Self {
+        Self::of(ch)
+    }
+}
+
+
+impl NumericCharProperty<u8> for CanonicalCombiningClass {
+    /// Get numeric value for character property value
+    fn number(&self) -> u8 {
+        self.number()
+    }
+}
+
+
 // TODO: Once we fully adopt 1.20 change these to associated consts on CanonicalCombiningClass
 /// *Canonical_Combining_Class*es by their name
 #[allow(non_upper_case_globals)]
@@ -84,13 +99,6 @@ pub mod values {
 }
 
 
-impl CharProperty for CanonicalCombiningClass {
-    fn of(ch: char) -> Self {
-        Self::of(ch)
-    }
-}
-
-
 impl CanonicalCombiningClass {
     /// Find the character *Canonical_Combining_Class* property value.
     pub fn of(ch: char) -> CanonicalCombiningClass {

unic/ucd/normal/src/composition.rs

@@ -45,8 +45,7 @@ pub fn canonical_composition(c: char) -> Option<&'static ([(char, char)])> {
 pub fn canonical_decomposition(c: char) -> Option<&'static [char]> {
     const LOOKUP: &'static [(char, Slice)] =
         include!("tables/canonical_decomposition_mapping_lookup.rsv");
-    const VALUES: &'static [char] =
-        include!("tables/canonical_decomposition_mapping_values.rsv");
+    const VALUES: &'static [char] = include!("tables/canonical_decomposition_mapping_values.rsv");
     bsearch_lookup_table(c, LOOKUP, VALUES)
 }
 

unic/utils/README.md

@@ -3,4 +3,4 @@
 [![Crates.io](https://img.shields.io/crates/v/unic-utils.svg)](https://crates.io/crates/unic-utils)
 [![Documentation](https://docs.rs/unic-utils/badge.svg)](https://docs.rs/unic-utils/)
 
-This UNIC component provides utility libraries that do not depend on Unicode data.
+This UNIC component provides utility libraries that do not depend on any data.

unic/utils/src/char_property.rs

@@ -9,14 +9,20 @@
 // except according to those terms.
 
 
-//! TBD.
+//! Taxonomy and Contracts for Character Property types.
+//!
+//! See also the list of types of character properties defined in the UCD:
+//! <http://unicode.org/reports/tr44/#About_Property_Table>, in [Unicode® Standard Annex #44 —
+//! Unicode Character Database](http://unicode.org/reports/tr44/#About_Property_Table)
 
 
 use std::fmt::{Debug, Display};
 use std::hash::Hash;
 
 
-/// TBD.
+/// A Character Property defined on all characters.
+///
+/// Examples: *Age*, *Name*, *General_Category*, *Bidi_Class*
 pub trait CharProperty
 where
     Self: Copy + Debug + Default + Display + Eq + Hash,
@@ -26,7 +32,9 @@ where
 }
 
 
-/// TBD.
+/// A Character Property defined for some characters.
+///
+/// Examples: *Decomposition_Type*, *Numeric_Type*
 pub trait OptionCharProperty
 where
     Self: Copy + Debug + Display + Eq + Hash,
@@ -36,11 +44,36 @@ where
 }
 
 
-/// TBD.
+/// A Character Property with enumerated values.
+///
+/// This is similar to types *Catalog* and *Enumeration*, as defined in UAX#44.
+///
+/// Usage Note: If the property is of type *Catalog* (as defined by Unicode), it's recommended to
+/// (in some way) mark the type as *non-exhaustive*, so that adding new variants to the `enum` type
+/// won't result in API breakage.
 pub trait EnumeratedCharProperty
 where
     Self: Copy + Debug + Display + Eq + Hash,
 {
     /// TBD
     fn all_values() -> &'static [Self];
 }
+
+
+/// Marker for numeric types accepted by `NumericCharProperty`.
+pub trait NumericCharPropertyValue {}
+
+impl NumericCharPropertyValue for u8 {}
+
+
+/// A Character Property with numeric values.
+///
+/// Examples: *Numeric_Value*, *Canonical_Combining_Class*
+pub trait NumericCharProperty<Value>
+where
+    Self: Copy + Debug + Default + Display + Eq + Hash,
+    Value: NumericCharPropertyValue,
+{
+    /// Get numeric value for character property value
+    fn number(&self) -> Value;
+}

unic/utils/src/lib.rs

@@ -27,11 +27,16 @@ pub const PKG_NAME: &'static str = env!("CARGO_PKG_NAME");
 pub const PKG_DESCRIPTION: &'static str = env!("CARGO_PKG_DESCRIPTION");
 
 
+pub mod char_property;
 pub mod codepoints;
 pub mod tables;
-pub mod char_property;
 
 
-pub use char_property::{CharProperty, EnumeratedCharProperty, OptionCharProperty};
+pub use char_property::{
+    CharProperty,
+    EnumeratedCharProperty,
+    NumericCharProperty,
+    OptionCharProperty,
+};
 pub use codepoints::iter_all_chars;
 pub use tables::CharDataTable;