Support mysql RLIKE and REGEXP binary operators (#1017)

apache · Oct 20, 2023 · c03586b · c03586b
1 parent 88510f6
commit c03586b
Show file tree

Hide file tree

Showing 5 changed files with 53 additions and 3 deletions.
diff --git a/src/ast/mod.rs b/src/ast/mod.rs
@@ -429,6 +429,14 @@ pub enum Expr {
         pattern: Box<Expr>,
         escape_char: Option<char>,
     },
+    /// MySQL: RLIKE regex or REGEXP regex
+    RLike {
+        negated: bool,
+        expr: Box<Expr>,
+        pattern: Box<Expr>,
+        // true for REGEXP, false for RLIKE (no difference in semantics)
+        regexp: bool,
+    },
     /// Any operation e.g. `foo > ANY(bar)`, comparison operator is one of [=, >, <, =>, =<, !=]
     AnyOp {
         left: Box<Expr>,
@@ -740,6 +748,19 @@ impl fmt::Display for Expr {
                     pattern
                 ),
             },
+            Expr::RLike {
+                negated,
+                expr,
+                pattern,
+                regexp,
+            } => write!(
+                f,
+                "{} {}{} {}",
+                expr,
+                if *negated { "NOT " } else { "" },
+                if *regexp { "REGEXP" } else { "RLIKE" },
+                pattern
+            ),
             Expr::SimilarTo {
                 negated,
                 expr,

diff --git a/src/keywords.rs b/src/keywords.rs
@@ -498,6 +498,7 @@ define_keywords!(
     REFERENCES,
     REFERENCING,
     REGCLASS,
+    REGEXP,
     REGR_AVGX,
     REGR_AVGY,
     REGR_COUNT,
@@ -524,6 +525,7 @@ define_keywords!(
     RETURNS,
     REVOKE,
     RIGHT,
+    RLIKE,
     ROLE,
     ROLLBACK,
     ROLLUP,

diff --git a/src/parser/mod.rs b/src/parser/mod.rs
@@ -1932,10 +1932,21 @@ impl<'a> Parser<'a> {
                 | Keyword::BETWEEN
                 | Keyword::LIKE
                 | Keyword::ILIKE
-                | Keyword::SIMILAR => {
+                | Keyword::SIMILAR
+                | Keyword::REGEXP
+                | Keyword::RLIKE => {
                     self.prev_token();
                     let negated = self.parse_keyword(Keyword::NOT);
-                    if self.parse_keyword(Keyword::IN) {
+                    let regexp = self.parse_keyword(Keyword::REGEXP);
+                    let rlike = self.parse_keyword(Keyword::RLIKE);
+                    if regexp || rlike {
+                        Ok(Expr::RLike {
+                            negated,
+                            expr: Box::new(expr),
+                            pattern: Box::new(self.parse_subexpr(Self::LIKE_PREC)?),
+                            regexp,
+                        })
+                    } else if self.parse_keyword(Keyword::IN) {
                         self.parse_in(expr, negated)
                     } else if self.parse_keyword(Keyword::BETWEEN) {
                         self.parse_between(expr, negated)
@@ -2178,6 +2189,8 @@ impl<'a> Parser<'a> {
                 Token::Word(w) if w.keyword == Keyword::BETWEEN => Ok(Self::BETWEEN_PREC),
                 Token::Word(w) if w.keyword == Keyword::LIKE => Ok(Self::LIKE_PREC),
                 Token::Word(w) if w.keyword == Keyword::ILIKE => Ok(Self::LIKE_PREC),
+                Token::Word(w) if w.keyword == Keyword::RLIKE => Ok(Self::LIKE_PREC),
+                Token::Word(w) if w.keyword == Keyword::REGEXP => Ok(Self::LIKE_PREC),
                 Token::Word(w) if w.keyword == Keyword::SIMILAR => Ok(Self::LIKE_PREC),
                 _ => Ok(0),
             },
@@ -2186,6 +2199,8 @@ impl<'a> Parser<'a> {
             Token::Word(w) if w.keyword == Keyword::BETWEEN => Ok(Self::BETWEEN_PREC),
             Token::Word(w) if w.keyword == Keyword::LIKE => Ok(Self::LIKE_PREC),
             Token::Word(w) if w.keyword == Keyword::ILIKE => Ok(Self::LIKE_PREC),
+            Token::Word(w) if w.keyword == Keyword::RLIKE => Ok(Self::LIKE_PREC),
+            Token::Word(w) if w.keyword == Keyword::REGEXP => Ok(Self::LIKE_PREC),
             Token::Word(w) if w.keyword == Keyword::SIMILAR => Ok(Self::LIKE_PREC),
             Token::Word(w) if w.keyword == Keyword::OPERATOR => Ok(Self::BETWEEN_PREC),
             Token::Word(w) if w.keyword == Keyword::DIV => Ok(Self::MUL_DIV_MOD_OP_PREC),

diff --git a/src/test_utils.rs b/src/test_utils.rs
@@ -111,7 +111,7 @@ impl TestedDialects {
     /// 2. re-serializing the result of parsing `sql` produces the same
     /// `canonical` sql string
     pub fn one_statement_parses_to(&self, sql: &str, canonical: &str) -> Statement {
-        let mut statements = self.parse_sql_statements(sql).unwrap();
+        let mut statements = self.parse_sql_statements(sql).expect(sql);
         assert_eq!(statements.len(), 1);
 
         if !canonical.is_empty() && sql != canonical {

diff --git a/tests/sqlparser_mysql.rs b/tests/sqlparser_mysql.rs
@@ -1454,6 +1454,18 @@ fn parse_show_variables() {
     mysql_and_generic().verified_stmt("SHOW VARIABLES WHERE value = '3306'");
 }
 
+#[test]
+fn parse_rlike_and_regexp() {
+    for s in &[
+        "SELECT 1 WHERE 'a' RLIKE '^a$'",
+        "SELECT 1 WHERE 'a' REGEXP '^a$'",
+        "SELECT 1 WHERE 'a' NOT RLIKE '^a$'",
+        "SELECT 1 WHERE 'a' NOT REGEXP '^a$'",
+    ] {
+        mysql_and_generic().verified_only_select(s);
+    }
+}
+
 #[test]
 fn parse_kill() {
     let stmt = mysql_and_generic().verified_stmt("KILL CONNECTION 5");