From c5e6ba5e7d746161a59d6fd5e09846262a2fc01e Mon Sep 17 00:00:00 2001
From: etgarperets <etgar.perets@satoricyber.com>
Date: Mon, 14 Jul 2025 11:24:13 +0300
Subject: [PATCH] Add identifier unicode support in Mysql, Postgres and
 Redshift (#1933)

---
 src/dialect/mysql.rs      |  4 +++-
 src/dialect/postgresql.rs |  4 +++-
 src/dialect/redshift.rs   |  4 ++--
 tests/sqlparser_common.rs | 11 +++++++++++
 4 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/src/dialect/mysql.rs b/src/dialect/mysql.rs
index f69e4243..b50c8df5 100644
--- a/src/dialect/mysql.rs
+++ b/src/dialect/mysql.rs
@@ -51,7 +51,9 @@ impl Dialect for MySqlDialect {
     }
 
     fn is_identifier_part(&self, ch: char) -> bool {
-        self.is_identifier_start(ch) || ch.is_ascii_digit()
+        self.is_identifier_start(ch) || ch.is_ascii_digit() ||
+        // MySQL implements Unicode characters in identifiers.
+        !ch.is_ascii()
     }
 
     fn is_delimited_identifier_start(&self, ch: char) -> bool {
diff --git a/src/dialect/postgresql.rs b/src/dialect/postgresql.rs
index b2d4014c..c1f02557 100644
--- a/src/dialect/postgresql.rs
+++ b/src/dialect/postgresql.rs
@@ -72,7 +72,9 @@ impl Dialect for PostgreSqlDialect {
     }
 
     fn is_identifier_part(&self, ch: char) -> bool {
-        ch.is_alphabetic() || ch.is_ascii_digit() || ch == '$' || ch == '_'
+        ch.is_alphabetic() || ch.is_ascii_digit() || ch == '$' || ch == '_'  ||
+        // PostgreSQL implements Unicode characters in identifiers.
+        !ch.is_ascii()
     }
 
     fn supports_unicode_string_literal(&self) -> bool {
diff --git a/src/dialect/redshift.rs b/src/dialect/redshift.rs
index 8ffed98a..c910e4c7 100644
--- a/src/dialect/redshift.rs
+++ b/src/dialect/redshift.rs
@@ -86,9 +86,9 @@ impl Dialect for RedshiftSqlDialect {
     }
 
     fn is_identifier_part(&self, ch: char) -> bool {
-        // Extends Postgres dialect with sharp and UTF-8 multibyte chars
+        // UTF-8 multibyte characters are supported in identifiers via the PostgreSqlDialect.
         // https://docs.aws.amazon.com/redshift/latest/dg/r_names.html
-        PostgreSqlDialect {}.is_identifier_part(ch) || ch == '#' || !ch.is_ascii()
+        PostgreSqlDialect {}.is_identifier_part(ch) || ch == '#'
     }
 
     /// redshift has `CONVERT(type, value)` instead of `CONVERT(value, type)`
diff --git a/tests/sqlparser_common.rs b/tests/sqlparser_common.rs
index 2d2008c8..ba72399f 100644
--- a/tests/sqlparser_common.rs
+++ b/tests/sqlparser_common.rs
@@ -16136,3 +16136,14 @@ SELECT * FROM tbl2
     assert_eq!(stmts.len(), 2);
     assert!(stmts.iter().all(|s| matches!(s, Statement::Query { .. })));
 }
+
+#[test]
+fn test_identifier_unicode_support() {
+    let sql = r#"SELECT phoneǤЖשचᎯ⻩☯♜🦄⚛🀄ᚠ⌛🌀 AS tbl FROM customers"#;
+    let dialects = TestedDialects::new(vec![
+        Box::new(MySqlDialect {}),
+        Box::new(RedshiftSqlDialect {}),
+        Box::new(PostgreSqlDialect {}),
+    ]);
+    let _ = dialects.verified_stmt(sql);
+}