Rollup merge of #114193 - crlf0710:lexer_unicode15, r=Manishearth

Update lexer emoji diagnostics to Unicode 15.0

This replaces the `unic-emoji-char` dep tree (which hasn't been updated for a while) with `unicode-properties` crate which contains Unicode 15.0 data.

Improves diagnostics for added emoji characters in recent years. (See tests).

cc #101840

cc ``@Manishearth``

2 files changed, 9 insertions, 8 deletions

diff --git a/compiler/rustc_lexer/Cargo.toml b/compiler/rustc_lexer/Cargo.toml
index 23294dc2e1b..2211ac1c8a7 100644
--- a/compiler/rustc_lexer/Cargo.toml
+++ b/compiler/rustc_lexer/Cargo.toml
@@ -16,7 +16,11 @@ Rust lexer used by rustc. No stability guarantees are provided.
 # Note that this crate purposefully does not depend on other rustc crates
 [dependencies]
 unicode-xid = "0.2.0"
-unic-emoji-char = "0.9.0"
+
+[dependencies.unicode-properties]
+version = "0.1.0"
+default-features = false
+features = ["emoji"]
 
 [dev-dependencies]
 expect-test = "1.4.0"
diff --git a/compiler/rustc_lexer/src/lib.rs b/compiler/rustc_lexer/src/lib.rs
index d511d2b1280..43dfd34a6ff 100644
--- a/compiler/rustc_lexer/src/lib.rs
+++ b/compiler/rustc_lexer/src/lib.rs
@@ -34,6 +34,7 @@ pub use crate::cursor::Cursor;
 use self::LiteralKind::*;
 use self::TokenKind::*;
 use crate::cursor::EOF_CHAR;
+use unicode_properties::UnicodeEmoji;
 
 /// Parsed token.
 /// It doesn't contain information about data that has been parsed,
@@ -428,9 +429,7 @@ impl Cursor<'_> {
                 Literal { kind, suffix_start }
             }
             // Identifier starting with an emoji. Only lexed for graceful error recovery.
-            c if !c.is_ascii() && unic_emoji_char::is_emoji(c) => {
-                self.fake_ident_or_unknown_prefix()
-            }
+            c if !c.is_ascii() && c.is_emoji_char() => self.fake_ident_or_unknown_prefix(),
             _ => Unknown,
         };
         let res = Token::new(token_kind, self.pos_within_token());
@@ -514,9 +513,7 @@ impl Cursor<'_> {
         // we see a prefix here, it is definitely an unknown prefix.
         match self.first() {
             '#' | '"' | '\'' => UnknownPrefix,
-            c if !c.is_ascii() && unic_emoji_char::is_emoji(c) => {
-                self.fake_ident_or_unknown_prefix()
-            }
+            c if !c.is_ascii() && c.is_emoji_char() => self.fake_ident_or_unknown_prefix(),
             _ => Ident,
         }
     }
@@ -525,7 +522,7 @@ impl Cursor<'_> {
         // Start is already eaten, eat the rest of identifier.
         self.eat_while(|c| {
             unicode_xid::UnicodeXID::is_xid_continue(c)
-                || (!c.is_ascii() && unic_emoji_char::is_emoji(c))
+                || (!c.is_ascii() && c.is_emoji_char())
                 || c == '\u{200d}'
         });
         // Known prefixes must have been handled earlier. So if