Auto merge of rust-lang#146173 - Kmeakin:km/unicode-data/no-ascii, r=jhpratt

Don't include ASCII characters in Unicode tables

Split off from rust-lang#145219

Author: bors

alloc/src/str.rs

@@ -418,9 +418,8 @@ impl str {
         }
 
         fn case_ignorable_then_cased<I: Iterator<Item = char>>(iter: I) -> bool {
-            use core::unicode::{Case_Ignorable, Cased};
-            match iter.skip_while(|&c| Case_Ignorable(c)).next() {
-                Some(c) => Cased(c),
+            match iter.skip_while(|&c| c.is_case_ignorable()).next() {
+                Some(c) => c.is_cased(),
                 None => false,
             }
         }

core/src/char/methods.rs

@@ -969,7 +969,43 @@ impl char {
     #[must_use]
     #[inline]
     pub(crate) fn is_grapheme_extended(self) -> bool {
-        unicode::Grapheme_Extend(self)
+        !self.is_ascii() && unicode::Grapheme_Extend(self)
+    }
+
+    /// Returns `true` if this `char` has the `Cased` property.
+    ///
+    /// `Cased` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
+    /// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
+    ///
+    /// [Unicode Standard]: https://www.unicode.org/versions/latest/
+    /// [ucd]: https://www.unicode.org/reports/tr44/
+    /// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
+    #[must_use]
+    #[inline]
+    #[doc(hidden)]
+    #[unstable(feature = "char_internals", reason = "exposed only for libstd", issue = "none")]
+    pub fn is_cased(self) -> bool {
+        if self.is_ascii() { self.is_ascii_alphabetic() } else { unicode::Cased(self) }
+    }
+
+    /// Returns `true` if this `char` has the `Case_Ignorable` property.
+    ///
+    /// `Case_Ignorable` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
+    /// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
+    ///
+    /// [Unicode Standard]: https://www.unicode.org/versions/latest/
+    /// [ucd]: https://www.unicode.org/reports/tr44/
+    /// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
+    #[must_use]
+    #[inline]
+    #[doc(hidden)]
+    #[unstable(feature = "char_internals", reason = "exposed only for libstd", issue = "none")]
+    pub fn is_case_ignorable(self) -> bool {
+        if self.is_ascii() {
+            matches!(self, '\'' | '.' | ':' | '^' | '`')
+        } else {
+            unicode::Case_Ignorable(self)
+        }
     }
 
     /// Returns `true` if this `char` has one of the general categories for numbers.