Add support for DuckDB struct literal syntax (apache#1194)

Co-authored-by: Andrew Lamb <[email protected]>

Author: 2 people

src/ast/mod.rs

@@ -347,6 +347,23 @@ impl fmt::Display for StructField {
     }
 }
 
+/// A dictionary field within a dictionary.
+///
+/// [duckdb]: https://duckdb.org/docs/sql/data_types/struct#creating-structs
+#[derive(Debug, Clone, PartialEq, PartialOrd, Eq, Ord, Hash)]
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "visitor", derive(Visit, VisitMut))]
+pub struct DictionaryField {
+    pub key: Ident,
+    pub value: Box<Expr>,
+}
+
+impl fmt::Display for DictionaryField {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        write!(f, "{}: {}", self.key, self.value)
+    }
+}
+
 /// Options for `CAST` / `TRY_CAST`
 /// BigQuery: <https://cloud.google.com/bigquery/docs/reference/standard-sql/format-elements#formatting_syntax>
 #[derive(Debug, Clone, PartialEq, PartialOrd, Eq, Ord, Hash)]
@@ -687,6 +704,14 @@ pub enum Expr {
         expr: Box<Expr>,
         name: Ident,
     },
+    /// `DuckDB` specific `Struct` literal expression [1]
+    ///
+    /// Syntax:
+    /// ```sql
+    /// syntax: {'field_name': expr1[, ... ]}
+    /// ```
+    /// [1]: https://duckdb.org/docs/sql/data_types/struct#creating-structs
+    Dictionary(Vec<DictionaryField>),
     /// An array index expression e.g. `(ARRAY[1, 2])[1]` or `(current_schemas(FALSE))[1]`
     ArrayIndex {
         obj: Box<Expr>,
@@ -1146,6 +1171,9 @@ impl fmt::Display for Expr {
             Expr::Named { expr, name } => {
                 write!(f, "{} AS {}", expr, name)
             }
+            Expr::Dictionary(fields) => {
+                write!(f, "{{{}}}", display_comma_separated(fields))
+            }
             Expr::ArrayIndex { obj, indexes } => {
                 write!(f, "{obj}")?;
                 for i in indexes {

src/parser/mod.rs

@@ -1117,6 +1117,10 @@ impl<'a> Parser<'a> {
                 self.prev_token();
                 Ok(Expr::Value(self.parse_value()?))
             }
+            Token::LBrace if dialect_of!(self is DuckDbDialect | GenericDialect) => {
+                self.prev_token();
+                self.parse_duckdb_struct_literal()
+            }
             _ => self.expected("an expression:", next_token),
         }?;
 
@@ -2127,6 +2131,45 @@ impl<'a> Parser<'a> {
         ))
     }
 
+    /// DuckDB specific: Parse a duckdb dictionary [1]
+    ///
+    /// Syntax:
+    ///
+    /// ```sql
+    /// {'field_name': expr1[, ... ]}
+    /// ```
+    ///
+    /// [1]: https://duckdb.org/docs/sql/data_types/struct#creating-structs
+    fn parse_duckdb_struct_literal(&mut self) -> Result<Expr, ParserError> {
+        self.expect_token(&Token::LBrace)?;
+
+        let fields = self.parse_comma_separated(Self::parse_duckdb_dictionary_field)?;
+
+        self.expect_token(&Token::RBrace)?;
+
+        Ok(Expr::Dictionary(fields))
+    }
+
+    /// Parse a field for a duckdb dictionary [1]
+    /// Syntax
+    /// ```sql
+    /// 'name': expr
+    /// ```
+    ///
+    /// [1]: https://duckdb.org/docs/sql/data_types/struct#creating-structs
+    fn parse_duckdb_dictionary_field(&mut self) -> Result<DictionaryField, ParserError> {
+        let key = self.parse_identifier(false)?;
+
+        self.expect_token(&Token::Colon)?;
+
+        let expr = self.parse_expr()?;
+
+        Ok(DictionaryField {
+            key,
+            value: Box::new(expr),
+        })
+    }
+
     /// For nested types that use the angle bracket syntax, this matches either
     /// `>`, `>>` or nothing depending on which variant is expected (specified by the previously
     /// matched `trailing_bracket` argument). It returns whether there is a trailing

tests/sqlparser_duckdb.rs

@@ -314,4 +314,91 @@ fn test_drop_secret_simple() {
             ,
         stmt
     );
-}
+}
+
+#[test]
+fn test_duckdb_struct_literal() {
+    //struct literal syntax https://duckdb.org/docs/sql/data_types/struct#creating-structs
+    //syntax: {'field_name': expr1[, ... ]}
+    let sql = "SELECT {'a': 1, 'b': 2, 'c': 3}, [{'a': 'abc'}], {'a': 1, 'b': [t.str_col]}, {'a': 1, 'b': 'abc'}, {'abc': str_col}, {'a': {'aa': 1}}";
+    let select = duckdb_and_generic().verified_only_select(sql);
+    assert_eq!(6, select.projection.len());
+    assert_eq!(
+        &Expr::Dictionary(vec![
+            DictionaryField {
+                key: Ident::with_quote('\'', "a"),
+                value: Box::new(Expr::Value(number("1"))),
+            },
+            DictionaryField {
+                key: Ident::with_quote('\'', "b"),
+                value: Box::new(Expr::Value(number("2"))),
+            },
+            DictionaryField {
+                key: Ident::with_quote('\'', "c"),
+                value: Box::new(Expr::Value(number("3"))),
+            },
+        ],),
+        expr_from_projection(&select.projection[0])
+    );
+
+    assert_eq!(
+        &Expr::Array(Array {
+            elem: vec![Expr::Dictionary(vec![DictionaryField {
+                key: Ident::with_quote('\'', "a"),
+                value: Box::new(Expr::Value(Value::SingleQuotedString("abc".to_string()))),
+            },],)],
+            named: false
+        }),
+        expr_from_projection(&select.projection[1])
+    );
+    assert_eq!(
+        &Expr::Dictionary(vec![
+            DictionaryField {
+                key: Ident::with_quote('\'', "a"),
+                value: Box::new(Expr::Value(number("1"))),
+            },
+            DictionaryField {
+                key: Ident::with_quote('\'', "b"),
+                value: Box::new(Expr::Array(Array {
+                    elem: vec![Expr::CompoundIdentifier(vec![
+                        Ident::from("t"),
+                        Ident::from("str_col")
+                    ])],
+                    named: false
+                })),
+            },
+        ],),
+        expr_from_projection(&select.projection[2])
+    );
+    assert_eq!(
+        &Expr::Dictionary(vec![
+            DictionaryField {
+                key: Ident::with_quote('\'', "a"),
+                value: Expr::Value(number("1")).into(),
+            },
+            DictionaryField {
+                key: Ident::with_quote('\'', "b"),
+                value: Expr::Value(Value::SingleQuotedString("abc".to_string())).into(),
+            },
+        ],),
+        expr_from_projection(&select.projection[3])
+    );
+    assert_eq!(
+        &Expr::Dictionary(vec![DictionaryField {
+            key: Ident::with_quote('\'', "abc"),
+            value: Expr::Identifier(Ident::from("str_col")).into(),
+        }],),
+        expr_from_projection(&select.projection[4])
+    );
+    assert_eq!(
+        &Expr::Dictionary(vec![DictionaryField {
+            key: Ident::with_quote('\'', "a"),
+            value: Expr::Dictionary(vec![DictionaryField {
+                key: Ident::with_quote('\'', "aa"),
+                value: Expr::Value(number("1")).into(),
+            }],)
+            .into(),
+        }],),
+        expr_from_projection(&select.projection[5])
+    );
+}