Support CONVERT expressions (apache#1048)

Author: lovasoa

src/ast/mod.rs

@@ -473,6 +473,17 @@ pub enum Expr {
     },
     /// Unary operation e.g. `NOT foo`
     UnaryOp { op: UnaryOperator, expr: Box<Expr> },
+    /// CONVERT a value to a different data type or character encoding `CONVERT(foo USING utf8mb4)`
+    Convert {
+        /// The expression to convert
+        expr: Box<Expr>,
+        /// The target data type
+        data_type: Option<DataType>,
+        /// The target character encoding
+        charset: Option<ObjectName>,
+        /// whether the target comes before the expr (MSSQL syntax)
+        target_before_value: bool,
+    },
     /// CAST an expression to a different data type e.g. `CAST(foo AS VARCHAR(123))`
     Cast {
         expr: Box<Expr>,
@@ -844,6 +855,28 @@ impl fmt::Display for Expr {
                     write!(f, "{op}{expr}")
                 }
             }
+            Expr::Convert {
+                expr,
+                target_before_value,
+                data_type,
+                charset,
+            } => {
+                write!(f, "CONVERT(")?;
+                if let Some(data_type) = data_type {
+                    if let Some(charset) = charset {
+                        write!(f, "{expr}, {data_type} CHARACTER SET {charset}")
+                    } else if *target_before_value {
+                        write!(f, "{data_type}, {expr}")
+                    } else {
+                        write!(f, "{expr}, {data_type}")
+                    }
+                } else if let Some(charset) = charset {
+                    write!(f, "{expr} USING {charset}")
+                } else {
+                    write!(f, "{expr}") // This should never happen
+                }?;
+                write!(f, ")")
+            }
             Expr::Cast {
                 expr,
                 data_type,

src/dialect/mod.rs

@@ -128,6 +128,11 @@ pub trait Dialect: Debug + Any {
     fn supports_in_empty_list(&self) -> bool {
         false
     }
+    /// Returns true if the dialect has a CONVERT function which accepts a type first
+    /// and an expression second, e.g. `CONVERT(varchar, 1)`
+    fn convert_type_before_value(&self) -> bool {
+        false
+    }
     /// Dialect-specific prefix parser override
     fn parse_prefix(&self, _parser: &mut Parser) -> Option<Result<Expr, ParserError>> {
         // return None to fall back to the default behavior

src/dialect/mssql.rs

@@ -35,6 +35,12 @@ impl Dialect for MsSqlDialect {
             || ch == '_'
     }
 
+    /// SQL Server has `CONVERT(type, value)` instead of `CONVERT(value, type)`
+    /// <https://learn.microsoft.com/en-us/sql/t-sql/functions/cast-and-convert-transact-sql?view=sql-server-ver16>
+    fn convert_type_before_value(&self) -> bool {
+        true
+    }
+
     fn supports_substring_from_for_expr(&self) -> bool {
         false
     }

src/dialect/redshift.rs

@@ -53,4 +53,10 @@ impl Dialect for RedshiftSqlDialect {
         // Extends Postgres dialect with sharp
         PostgreSqlDialect {}.is_identifier_part(ch) || ch == '#'
     }
+
+    /// redshift has `CONVERT(type, value)` instead of `CONVERT(value, type)`
+    /// <https://docs.aws.amazon.com/redshift/latest/dg/r_CONVERT_function.html>
+    fn convert_type_before_value(&self) -> bool {
+        true
+    }
 }

src/parser/mod.rs

@@ -821,6 +821,7 @@ impl<'a> Parser<'a> {
                     self.parse_time_functions(ObjectName(vec![w.to_ident()]))
                 }
                 Keyword::CASE => self.parse_case_expr(),
+                Keyword::CONVERT => self.parse_convert_expr(),
                 Keyword::CAST => self.parse_cast_expr(),
                 Keyword::TRY_CAST => self.parse_try_cast_expr(),
                 Keyword::SAFE_CAST => self.parse_safe_cast_expr(),
@@ -1227,6 +1228,57 @@ impl<'a> Parser<'a> {
         }
     }
 
+    /// mssql-like convert function
+    fn parse_mssql_convert(&mut self) -> Result<Expr, ParserError> {
+        self.expect_token(&Token::LParen)?;
+        let data_type = self.parse_data_type()?;
+        self.expect_token(&Token::Comma)?;
+        let expr = self.parse_expr()?;
+        self.expect_token(&Token::RParen)?;
+        Ok(Expr::Convert {
+            expr: Box::new(expr),
+            data_type: Some(data_type),
+            charset: None,
+            target_before_value: true,
+        })
+    }
+
+    /// Parse a SQL CONVERT function:
+    ///  - `CONVERT('héhé' USING utf8mb4)` (MySQL)
+    ///  - `CONVERT('héhé', CHAR CHARACTER SET utf8mb4)` (MySQL)
+    ///  - `CONVERT(DECIMAL(10, 5), 42)` (MSSQL) - the type comes first
+    pub fn parse_convert_expr(&mut self) -> Result<Expr, ParserError> {
+        if self.dialect.convert_type_before_value() {
+            return self.parse_mssql_convert();
+        }
+        self.expect_token(&Token::LParen)?;
+        let expr = self.parse_expr()?;
+        if self.parse_keyword(Keyword::USING) {
+            let charset = self.parse_object_name()?;
+            self.expect_token(&Token::RParen)?;
+            return Ok(Expr::Convert {
+                expr: Box::new(expr),
+                data_type: None,
+                charset: Some(charset),
+                target_before_value: false,
+            });
+        }
+        self.expect_token(&Token::Comma)?;
+        let data_type = self.parse_data_type()?;
+        let charset = if self.parse_keywords(&[Keyword::CHARACTER, Keyword::SET]) {
+            Some(self.parse_object_name()?)
+        } else {
+            None
+        };
+        self.expect_token(&Token::RParen)?;
+        Ok(Expr::Convert {
+            expr: Box::new(expr),
+            data_type: Some(data_type),
+            charset,
+            target_before_value: false,
+        })
+    }
+
     /// Parse a SQL CAST function e.g. `CAST(expr AS FLOAT)`
     pub fn parse_cast_expr(&mut self) -> Result<Expr, ParserError> {
         self.expect_token(&Token::LParen)?;

tests/sqlparser_mssql.rs

@@ -475,6 +475,13 @@ fn parse_cast_varchar_max() {
     ms_and_generic().verified_expr("CAST('foo' AS VARCHAR(MAX))");
 }
 
+#[test]
+fn parse_convert() {
+    ms().verified_expr("CONVERT(VARCHAR(MAX), 'foo')");
+    ms().verified_expr("CONVERT(VARCHAR(10), 'foo')");
+    ms().verified_expr("CONVERT(DECIMAL(10,5), 12.55)");
+}
+
 #[test]
 fn parse_similar_to() {
     fn chk(negated: bool) {

tests/sqlparser_mysql.rs

@@ -1843,3 +1843,18 @@ fn parse_drop_temporary_table() {
         _ => unreachable!(),
     }
 }
+
+#[test]
+fn parse_convert_using() {
+    // https://dev.mysql.com/doc/refman/8.0/en/cast-functions.html#function_convert
+
+    // CONVERT(expr USING transcoding_name)
+    mysql().verified_only_select("SELECT CONVERT('x' USING latin1)");
+    mysql().verified_only_select("SELECT CONVERT(my_column USING utf8mb4) FROM my_table");
+
+    // CONVERT(expr, type)
+    mysql().verified_only_select("SELECT CONVERT('abc', CHAR(60))");
+    mysql().verified_only_select("SELECT CONVERT(123.456, DECIMAL(5,2))");
+    // with a type + a charset
+    mysql().verified_only_select("SELECT CONVERT('test', CHAR CHARACTER SET utf8mb4)");
+}