hive: support for special not expression !a and raise error for a! factorial operator (#1472)

Co-authored-by: Ifeanyi Ubah <[email protected]>

Author: wugeer

src/ast/operator.rs

@@ -51,6 +51,8 @@ pub enum UnaryOperator {
     PGPrefixFactorial,
     /// Absolute value, e.g. `@ -9` (PostgreSQL-specific)
     PGAbs,
+    /// Unary logical not operator: e.g. `! false` (Hive-specific)
+    BangNot,
 }
 
 impl fmt::Display for UnaryOperator {
@@ -65,6 +67,7 @@ impl fmt::Display for UnaryOperator {
             UnaryOperator::PGPostfixFactorial => "!",
             UnaryOperator::PGPrefixFactorial => "!!",
             UnaryOperator::PGAbs => "@",
+            UnaryOperator::BangNot => "!",
         })
     }
 }

src/dialect/hive.rs

@@ -51,4 +51,9 @@ impl Dialect for HiveDialect {
     fn require_interval_qualifier(&self) -> bool {
         true
     }
+
+    /// See Hive <https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=27362061#Tutorial-BuiltInOperators>
+    fn supports_bang_not_operator(&self) -> bool {
+        true
+    }
 }

src/dialect/mod.rs

@@ -575,6 +575,11 @@ pub trait Dialect: Debug + Any {
         false
     }
 
+    /// Returns true if the dialect supports `a!` expressions
+    fn supports_factorial_operator(&self) -> bool {
+        false
+    }
+
     /// Returns true if this dialect supports treating the equals operator `=` within a `SelectItem`
     /// as an alias assignment operator, rather than a boolean expression.
     /// For example: the following statements are equivalent for such a dialect:
@@ -591,6 +596,11 @@ pub trait Dialect: Debug + Any {
         false
     }
 
+    /// Returns true if the dialect supports `!a` syntax for boolean `NOT` expressions.
+    fn supports_bang_not_operator(&self) -> bool {
+        false
+    }
+
     /// Returns true if the dialect supports the `LISTEN` statement
     fn supports_listen(&self) -> bool {
         false

src/dialect/postgresql.rs

@@ -201,6 +201,11 @@ impl Dialect for PostgreSqlDialect {
     fn supports_notify(&self) -> bool {
         true
     }
+
+    /// see <https://www.postgresql.org/docs/13/functions-math.html>
+    fn supports_factorial_operator(&self) -> bool {
+        true
+    }
 }
 
 pub fn parse_comment(parser: &mut Parser) -> Result<Statement, ParserError> {

src/parser/mod.rs

@@ -1194,6 +1194,14 @@ impl<'a> Parser<'a> {
                     ),
                 })
             }
+            Token::ExclamationMark if self.dialect.supports_bang_not_operator() => {
+                Ok(Expr::UnaryOp {
+                    op: UnaryOperator::BangNot,
+                    expr: Box::new(
+                        self.parse_subexpr(self.dialect.prec_value(Precedence::UnaryNot))?,
+                    ),
+                })
+            }
             tok @ Token::DoubleExclamationMark
             | tok @ Token::PGSquareRoot
             | tok @ Token::PGCubeRoot
@@ -1287,7 +1295,6 @@ impl<'a> Parser<'a> {
             }
             _ => self.expected("an expression", next_token),
         }?;
-
         if self.parse_keyword(Keyword::COLLATE) {
             Ok(Expr::Collate {
                 expr: Box::new(expr),
@@ -2818,8 +2825,7 @@ impl<'a> Parser<'a> {
                 data_type: self.parse_data_type()?,
                 format: None,
             })
-        } else if Token::ExclamationMark == tok {
-            // PostgreSQL factorial operation
+        } else if Token::ExclamationMark == tok && self.dialect.supports_factorial_operator() {
             Ok(Expr::UnaryOp {
                 op: UnaryOperator::PGPostfixFactorial,
                 expr: Box::new(expr),

tests/sqlparser_common.rs

@@ -11532,3 +11532,113 @@ fn test_select_top() {
     dialects.verified_stmt("SELECT TOP 3 DISTINCT * FROM tbl");
     dialects.verified_stmt("SELECT TOP 3 DISTINCT a, b, c FROM tbl");
 }
+
+#[test]
+fn parse_bang_not() {
+    let dialects = all_dialects_where(|d| d.supports_bang_not_operator());
+    let sql = "SELECT !a, !(b > 3)";
+    let Select { projection, .. } = dialects.verified_only_select(sql);
+
+    for (i, expr) in [
+        Box::new(Expr::Identifier(Ident::new("a"))),
+        Box::new(Expr::Nested(Box::new(Expr::BinaryOp {
+            left: Box::new(Expr::Identifier(Ident::new("b"))),
+            op: BinaryOperator::Gt,
+            right: Box::new(Expr::Value(Value::Number("3".parse().unwrap(), false))),
+        }))),
+    ]
+    .into_iter()
+    .enumerate()
+    {
+        assert_eq!(
+            SelectItem::UnnamedExpr(Expr::UnaryOp {
+                op: UnaryOperator::BangNot,
+                expr
+            }),
+            projection[i]
+        )
+    }
+
+    let sql_statements = ["SELECT a!", "SELECT a ! b", "SELECT a ! as b"];
+
+    for &sql in &sql_statements {
+        assert_eq!(
+            dialects.parse_sql_statements(sql).unwrap_err(),
+            ParserError::ParserError("No infix parser for token ExclamationMark".to_string())
+        );
+    }
+
+    let sql_statements = ["SELECT !a", "SELECT !a b", "SELECT !a as b"];
+    let dialects = all_dialects_where(|d| !d.supports_bang_not_operator());
+
+    for &sql in &sql_statements {
+        assert_eq!(
+            dialects.parse_sql_statements(sql).unwrap_err(),
+            ParserError::ParserError("Expected: an expression, found: !".to_string())
+        );
+    }
+}
+
+#[test]
+fn parse_factorial_operator() {
+    let dialects = all_dialects_where(|d| d.supports_factorial_operator());
+    let sql = "SELECT a!, (b + c)!";
+    let Select { projection, .. } = dialects.verified_only_select(sql);
+
+    for (i, expr) in [
+        Box::new(Expr::Identifier(Ident::new("a"))),
+        Box::new(Expr::Nested(Box::new(Expr::BinaryOp {
+            left: Box::new(Expr::Identifier(Ident::new("b"))),
+            op: BinaryOperator::Plus,
+            right: Box::new(Expr::Identifier(Ident::new("c"))),
+        }))),
+    ]
+    .into_iter()
+    .enumerate()
+    {
+        assert_eq!(
+            SelectItem::UnnamedExpr(Expr::UnaryOp {
+                op: UnaryOperator::PGPostfixFactorial,
+                expr
+            }),
+            projection[i]
+        )
+    }
+
+    let sql_statements = ["SELECT !a", "SELECT !a b", "SELECT !a as b"];
+
+    for &sql in &sql_statements {
+        assert_eq!(
+            dialects.parse_sql_statements(sql).unwrap_err(),
+            ParserError::ParserError("Expected: an expression, found: !".to_string())
+        );
+    }
+
+    let sql_statements = ["SELECT a!", "SELECT a ! b", "SELECT a ! as b"];
+
+    // Due to the exclamation mark, which is both part of the `bang not` operator
+    // and the `factorial` operator,  additional filtering not supports
+    // `bang not` operator is required here.
+    let dialects =
+        all_dialects_where(|d| !d.supports_factorial_operator() && !d.supports_bang_not_operator());
+
+    for &sql in &sql_statements {
+        assert_eq!(
+            dialects.parse_sql_statements(sql).unwrap_err(),
+            ParserError::ParserError("No infix parser for token ExclamationMark".to_string())
+        );
+    }
+
+    // Due to the exclamation mark, which is both part of the `bang not` operator
+    // and the `factorial` operator,  additional filtering supports
+    // `bang not` operator is required here.
+    let dialects =
+        all_dialects_where(|d| !d.supports_factorial_operator() && d.supports_bang_not_operator());
+
+    for &sql in &sql_statements {
+        assert_eq!(
+            dialects.parse_sql_statements(sql).unwrap_err(),
+            ParserError::ParserError("No infix parser for token ExclamationMark".to_string())
+        );
+    }
+}