Merge pull request #6 from unicode-rs/mixed

Manishearth · web-flow · commit f35d6b6f5099 · 2020-01-01T16:05:45.000+05:30
Add mixed-script detection
diff --git a/Cargo.toml b/Cargo.toml
@@ -16,6 +16,7 @@ according to Unicode Technical Standard #39 rules.
 exclude = [ "target/*", "Cargo.lock" ]
 
 [dependencies]
+unicode-script = { git = "https://github.com/unicode-rs/unicode-script", default-features = false }
 std = { version = "1.0", package = "rustc-std-workspace-std", optional = true }
 core = { version = "1.0", package = "rustc-std-workspace-core", optional = true }
 compiler_builtins = { version = "0.1", optional = true }
diff --git a/src/general_security_profile.rs b/src/general_security_profile.rs
@@ -0,0 +1,20 @@
+//! Utilities for working with the [General Security Profile](https://www.unicode.org/reports/tr39/#General_Security_Profile)
+//! for identifiers
+
+use crate::tables::identifier_status as is;
+
+/// Methods for determining characters not restricted from use for identifiers.
+pub trait GeneralSecurityProfile {
+    /// Returns whether the character is not restricted from use for identifiers.
+    fn identifier_allowed(self) -> bool;
+}
+
+impl GeneralSecurityProfile for char {
+    #[inline]
+    fn identifier_allowed(self) -> bool { is::identifier_status_allowed(self) }
+}
+
+impl GeneralSecurityProfile for &'_ str {
+    #[inline]
+    fn identifier_allowed(self) -> bool { self.chars().all(is::identifier_status_allowed) }
+}
diff --git a/src/lib.rs b/src/lib.rs
@@ -15,7 +15,7 @@
 //! ```rust
 //! extern crate unicode_security;
 //!
-//! use unicode_security::IdentifierStatusChar;
+//! use unicode_security::GeneralSecurityProfile;
 //!
 //! fn main() {
 //!     let ch = 'µ'; // U+00B5 MICRO SIGN
@@ -55,21 +55,16 @@ extern crate std;
 #[cfg(feature = "bench")]
 extern crate test;
 
-use tables::identifier_status as is;
 pub use tables::UNICODE_VERSION;
 
-mod tables;
+pub mod mixed_script;
+pub mod general_security_profile;
 
-#[cfg(test)]
-mod tests;
+pub use mixed_script::MixedScript;
+pub use general_security_profile::GeneralSecurityProfile;
 
-/// Methods for determining characters not restricted from use for identifiers.
-pub trait UnicodeIdentifierStatus {
-    /// Returns whether the character is not restricted from use for identifiers.
-    fn identifier_allowed(self) -> bool;
-}
+#[rustfmt::skip]
+pub(crate) mod tables;
 
-impl UnicodeIdentifierStatus for char {
-    #[inline]
-    fn identifier_allowed(self) -> bool { is::identifier_status_allowed(self) }
-}
+#[cfg(test)]
+mod tests;
diff --git a/src/mixed_script.rs b/src/mixed_script.rs
@@ -0,0 +1,129 @@
+//! [Mixed-script detection](https://www.unicode.org/reports/tr39/#Mixed_Script_Detection)
+
+use unicode_script::{Script, ScriptExtension};
+
+/// An Augmented script set, as defined by UTS 39
+///
+/// https://www.unicode.org/reports/tr39/#def-augmented-script-set
+pub struct AugmentedScriptSet {
+    /// The base ScriptExtension value
+    pub base: ScriptExtension,
+    /// Han With Bopomofo
+    pub hanb: bool,
+    /// Japanese
+    pub jpan: bool,
+    /// Korean
+    pub kore: bool,
+}
+
+impl From<ScriptExtension> for AugmentedScriptSet {
+    fn from(ext: ScriptExtension) -> Self {
+        let mut hanb = false;
+        let mut jpan = false;
+        let mut kore = false;
+
+        if ext == ScriptExtension::Single(Script::Common) ||
+           ext == ScriptExtension::Single(Script::Inherited) ||
+           ext.contains_script(Script::Han) {
+            hanb = true;
+            jpan = true;
+            kore = true;
+        } else {
+            if ext.contains_script(Script::Hiragana) || ext.contains_script(Script::Katakana) {
+                jpan = true;
+            }
+
+            if ext.contains_script(Script::Hangul) {
+                kore = true;
+            }
+
+            if ext.contains_script(Script::Bopomofo) {
+                hanb = true;
+            }
+        }
+        Self {
+            base: ext,
+            hanb, jpan, kore
+        }
+    }
+}
+
+impl From<char> for AugmentedScriptSet {
+    fn from(c: char) -> Self {
+        AugmentedScriptSet::for_char(c)
+    }
+}
+
+impl From<&'_ str> for AugmentedScriptSet {
+    fn from(s: &'_ str) -> Self {
+        AugmentedScriptSet::for_str(s)
+    }
+}
+
+impl Default for AugmentedScriptSet {
+    fn default() -> Self {
+        AugmentedScriptSet {
+            base: ScriptExtension::Single(Script::Common),
+            hanb: true,
+            jpan: true,
+            kore: true,
+        }
+    }
+}
+
+impl AugmentedScriptSet {
+    /// Intersect this set with another
+    pub fn intersect(mut self, other: Self) -> Self {
+        self.base = self.base.intersect(other.base);
+        self.hanb = self.hanb && other.hanb;
+        self.jpan = self.jpan && other.jpan;
+        self.kore = self.kore && other.kore;
+        self
+    }
+
+    /// Check if the set is empty
+    pub fn is_empty(&self) -> bool {
+        self.base.is_empty() && ! self.hanb && !self.jpan && !self.kore
+    }
+
+    /// Check if the set is "All" (Common or Inherited)
+    pub fn is_all(&self) -> bool {
+        self.base == ScriptExtension::Single(Script::Common) ||
+        self.base == ScriptExtension::Single(Script::Inherited)
+    }
+
+    /// Construct an AugmentedScriptSet for a given character
+    pub fn for_char(c: char) -> Self {
+        ScriptExtension::from(c).into()
+    }
+
+    /// Find the [resolved script set](https://www.unicode.org/reports/tr39/#def-resolved-script-set) of a given string
+    pub fn for_str(s: &str) -> Self {
+        let mut set = AugmentedScriptSet::default();
+        for ch in s.chars() {
+            set = set.intersect(ch.into())
+        }
+        set
+    }
+}
+
+/// Extension trait for [mixed-script detection](https://www.unicode.org/reports/tr39/#Mixed_Script_Detection)
+pub trait MixedScript {
+    /// Check if a string is [single-script](https://www.unicode.org/reports/tr39/#def-single-script)
+    ///
+    /// Note that a single-script string may still contain multiple Script properties!
+    fn is_single_script(self) -> bool;
+
+    /// Find the [resolved script set](https://www.unicode.org/reports/tr39/#def-resolved-script-set) of a given string
+    fn resolve_script_set(self) -> AugmentedScriptSet;
+}
+
+impl MixedScript for &'_ str {
+    fn is_single_script(self) -> bool {
+        !AugmentedScriptSet::for_str(self).is_empty()
+    }
+
+    fn resolve_script_set(self) -> AugmentedScriptSet {
+        self.into()
+    }
+}
diff --git a/src/tests.rs b/src/tests.rs
@@ -10,19 +10,19 @@
 
 #[test]
 fn test_char() {
-    use super::IdentifierStatusChar;
-    assert_eq!(IdentifierStatusChar::identifier_allowed('A'), true);
+    use crate::GeneralSecurityProfile;
+    assert_eq!(GeneralSecurityProfile::identifier_allowed('A'), true);
     assert_eq!('A'.identifier_allowed(), true);
-    assert_eq!(IdentifierStatusChar::identifier_allowed('0'), true);
+    assert_eq!(GeneralSecurityProfile::identifier_allowed('0'), true);
     assert_eq!('0'.identifier_allowed(), true);
-    assert_eq!(IdentifierStatusChar::identifier_allowed('_'), true);
+    assert_eq!(GeneralSecurityProfile::identifier_allowed('_'), true);
     assert_eq!('_'.identifier_allowed(), true);
-    assert_eq!(IdentifierStatusChar::identifier_allowed('\x00'), false);
+    assert_eq!(GeneralSecurityProfile::identifier_allowed('\x00'), false);
     assert_eq!('\x00'.identifier_allowed(), false);
     // U+00B5 MICRO SIGN
-    assert_eq!(IdentifierStatusChar::identifier_allowed('µ'), false);
+    assert_eq!(GeneralSecurityProfile::identifier_allowed('µ'), false);
     assert_eq!('µ'.identifier_allowed(), false);
     // U+2160 ROMAN NUMERAL ONE
-    assert_eq!(IdentifierStatusChar::identifier_allowed('Ⅰ'), false);
+    assert_eq!(GeneralSecurityProfile::identifier_allowed('Ⅰ'), false);
     assert_eq!('Ⅰ'.identifier_allowed(), false);
 }
