Support DuckDB struct syntax and support list of struct syntax (#1372)

Signed-off-by: jayzhan211 <[email protected]> Co-authored-by: Andrew Lamb <[email protected]> Co-authored-by: Ifeanyi Ubah <[email protected]>
apache · Aug 15, 2024 · 8c4d30b · 8c4d30b
1 parent fab834d
commit 8c4d30b
Show file tree

Hide file tree

Showing 6 changed files with 209 additions and 33 deletions.
diff --git a/.gitignore b/.gitignore
@@ -16,3 +16,5 @@ Cargo.lock
 .vscode
 
 *.swp
+
+.DS_store
diff --git a/src/ast/data_type.rs b/src/ast/data_type.rs
@@ -302,7 +302,7 @@ pub enum DataType {
  ///
  /// [hive]: https://docs.cloudera.com/cdw-runtime/cloud/impala-sql-reference/topics/impala-struct.html
  /// [bigquery]: https://cloud.google.com/bigquery/docs/reference/standard-sql/data-types#struct_type
- Struct(Vec<StructField>),
+ Struct(Vec<StructField>, StructBracketKind),
  /// Union
  ///
  /// [duckdb]: https://duckdb.org/docs/sql/data_types/union.html
@@ -517,9 +517,16 @@ impl fmt::Display for DataType {
  }
  write!(f, ")")
  }
- DataType::Struct(fields) => {
+ DataType::Struct(fields, bracket) => {
  if !fields.is_empty() {
- write!(f, "STRUCT<{}>", display_comma_separated(fields))
+ match bracket {
+ StructBracketKind::Parentheses => {
+ write!(f, "STRUCT({})", display_comma_separated(fields))
+ }
+ StructBracketKind::AngleBrackets => {
+ write!(f, "STRUCT<{}>", display_comma_separated(fields))
+ }
+ }
  } else {
  write!(f, "STRUCT")
  }
@@ -618,6 +625,17 @@ fn format_clickhouse_datetime_precision_and_timezone(
  Ok(())
 }
 
+/// Type of brackets used for `STRUCT` literals.
+#[derive(Debug, Clone, PartialEq, PartialOrd, Eq, Ord, Hash)]
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "visitor", derive(Visit, VisitMut))]
+pub enum StructBracketKind {
+ /// Example: `STRUCT(a INT, b STRING)`
+ Parentheses,
+ /// Example: `STRUCT<a INT, b STRING>`
+ AngleBrackets,
+}
+
 /// Timestamp and Time data types information about TimeZone formatting.
 ///
 /// This is more related to a display information than real differences between each variant. To

diff --git a/src/ast/mod.rs b/src/ast/mod.rs
@@ -28,7 +28,8 @@ use serde::{Deserialize, Serialize};
 use sqlparser_derive::{Visit, VisitMut};
 
 pub use self::data_type::{
- ArrayElemTypeDef, CharLengthUnits, CharacterLength, DataType, ExactNumberInfo, TimezoneInfo,
+ ArrayElemTypeDef, CharLengthUnits, CharacterLength, DataType, ExactNumberInfo,
+ StructBracketKind, TimezoneInfo,
 };
 pub use self::dcl::{AlterRoleOperation, ResetConfig, RoleOption, SetConfigValue};
 pub use self::ddl::{

diff --git a/src/parser/mod.rs b/src/parser/mod.rs
@@ -2266,6 +2266,23 @@ impl<'a> Parser<'a> {
  ))
  }
 
+ /// Duckdb Struct Data Type <https://duckdb.org/docs/sql/data_types/struct.html#retrieving-from-structs>
+ fn parse_duckdb_struct_type_def(&mut self) -> Result<Vec<StructField>, ParserError> {
+ self.expect_keyword(Keyword::STRUCT)?;
+ self.expect_token(&Token::LParen)?;
+ let struct_body = self.parse_comma_separated(|parser| {
+ let field_name = parser.parse_identifier(false)?;
+ let field_type = parser.parse_data_type()?;
+
+ Ok(StructField {
+ field_name: Some(field_name),
+ field_type,
+ })
+ });
+ self.expect_token(&Token::RParen)?;
+ struct_body
+ }
+
  /// Parse a field definition in a [struct] or [tuple].
  /// Syntax:
  ///
@@ -7495,12 +7512,20 @@ impl<'a> Parser<'a> {
  ))))
  }
  }
+ Keyword::STRUCT if dialect_of!(self is DuckDbDialect) => {
+ self.prev_token();
+ let field_defs = self.parse_duckdb_struct_type_def()?;
+ Ok(DataType::Struct(field_defs, StructBracketKind::Parentheses))
+ }
  Keyword::STRUCT if dialect_of!(self is BigQueryDialect | GenericDialect) => {
  self.prev_token();
  let (field_defs, _trailing_bracket) =
  self.parse_struct_type_def(Self::parse_struct_field_def)?;
  trailing_bracket = _trailing_bracket;
- Ok(DataType::Struct(field_defs))
+ Ok(DataType::Struct(
+ field_defs,
+ StructBracketKind::AngleBrackets,
+ ))
  }
  Keyword::UNION if dialect_of!(self is DuckDbDialect | GenericDialect) => {
  self.prev_token();

diff --git a/tests/sqlparser_bigquery.rs b/tests/sqlparser_bigquery.rs
@@ -489,28 +489,34 @@ fn parse_nested_data_types() {
  vec![
  ColumnDef {
  name: Ident::new("x"),
- data_type: DataType::Struct(vec![
- StructField {
- field_name: Some("a".into()),
- field_type: DataType::Array(ArrayElemTypeDef::AngleBracket(
- Box::new(DataType::Int64,)
- ))
- },
- StructField {
- field_name: Some("b".into()),
- field_type: DataType::Bytes(Some(42))
- },
- ]),
+ data_type: DataType::Struct(
+ vec![
+ StructField {
+ field_name: Some("a".into()),
+ field_type: DataType::Array(ArrayElemTypeDef::AngleBracket(
+ Box::new(DataType::Int64,)
+ ))
+ },
+ StructField {
+ field_name: Some("b".into()),
+ field_type: DataType::Bytes(Some(42))
+ },
+ ],
+ StructBracketKind::AngleBrackets
+ ),
  collation: None,
  options: vec![],
  },
  ColumnDef {
  name: Ident::new("y"),
  data_type: DataType::Array(ArrayElemTypeDef::AngleBracket(Box::new(
- DataType::Struct(vec![StructField {
- field_name: None,
- field_type: DataType::Int64,
- }]),
+ DataType::Struct(
+ vec![StructField {
+ field_name: None,
+ field_type: DataType::Int64,
+ }],
+ StructBracketKind::AngleBrackets
+ ),
  ))),
  collation: None,
  options: vec![],
@@ -708,10 +714,13 @@ fn parse_typed_struct_syntax_bigquery() {
  },
  StructField {
  field_name: Some("str".into()),
- field_type: DataType::Struct(vec![StructField {
- field_name: None,
- field_type: DataType::Bool
- }])
+ field_type: DataType::Struct(
+ vec![StructField {
+ field_name: None,
+ field_type: DataType::Bool
+ }],
+ StructBracketKind::AngleBrackets
+ )
  },
  ]
  },
@@ -730,12 +739,15 @@ fn parse_typed_struct_syntax_bigquery() {
  fields: vec![
  StructField {
  field_name: Some("x".into()),
- field_type: DataType::Struct(Default::default())
+ field_type: DataType::Struct(
+ Default::default(),
+ StructBracketKind::AngleBrackets
+ )
  },
  StructField {
  field_name: Some("y".into()),
  field_type: DataType::Array(ArrayElemTypeDef::AngleBracket(Box::new(
- DataType::Struct(Default::default())
+ DataType::Struct(Default::default(), StructBracketKind::AngleBrackets)
  )))
  },
  ]
@@ -1013,10 +1025,13 @@ fn parse_typed_struct_syntax_bigquery_and_generic() {
  },
  StructField {
  field_name: Some("str".into()),
- field_type: DataType::Struct(vec![StructField {
- field_name: None,
- field_type: DataType::Bool
- }])
+ field_type: DataType::Struct(
+ vec![StructField {
+ field_name: None,
+ field_type: DataType::Bool
+ }],
+ StructBracketKind::AngleBrackets
+ )
  },
  ]
  },
@@ -1035,12 +1050,15 @@ fn parse_typed_struct_syntax_bigquery_and_generic() {
  fields: vec![
  StructField {
  field_name: Some("x".into()),
- field_type: DataType::Struct(Default::default())
+ field_type: DataType::Struct(
+ Default::default(),
+ StructBracketKind::AngleBrackets
+ )
  },
  StructField {
  field_name: Some("y".into()),
  field_type: DataType::Array(ArrayElemTypeDef::AngleBracket(Box::new(
- DataType::Struct(Default::default())
+ DataType::Struct(Default::default(), StructBracketKind::AngleBrackets)
  )))
  },
  ]

diff --git a/tests/sqlparser_duckdb.rs b/tests/sqlparser_duckdb.rs
@@ -32,6 +32,118 @@ fn duckdb_and_generic() -> TestedDialects {
  }
 }
 
+#[test]
+fn test_struct() {
+ // s STRUCT(v VARCHAR, i INTEGER)
+ let struct_type1 = DataType::Struct(
+ vec![
+ StructField {
+ field_name: Some(Ident::new("v")),
+ field_type: DataType::Varchar(None),
+ },
+ StructField {
+ field_name: Some(Ident::new("i")),
+ field_type: DataType::Integer(None),
+ },
+ ],
+ StructBracketKind::Parentheses,
+ );
+
+ // basic struct
+ let statement = duckdb().verified_stmt(r#"CREATE TABLE t1 (s STRUCT(v VARCHAR, i INTEGER))"#);
+ assert_eq!(
+ column_defs(statement),
+ vec![ColumnDef {
+ name: "s".into(),
+ data_type: struct_type1.clone(),
+ collation: None,
+ options: vec![],
+ }]
+ );
+
+ // struct array
+ let statement = duckdb().verified_stmt(r#"CREATE TABLE t1 (s STRUCT(v VARCHAR, i INTEGER)[])"#);
+ assert_eq!(
+ column_defs(statement),
+ vec![ColumnDef {
+ name: "s".into(),
+ data_type: DataType::Array(ArrayElemTypeDef::SquareBracket(
+ Box::new(struct_type1),
+ None
+ )),
+ collation: None,
+ options: vec![],
+ }]
+ );
+
+ // s STRUCT(v VARCHAR, s STRUCT(a1 INTEGER, a2 VARCHAR))
+ let struct_type2 = DataType::Struct(
+ vec![
+ StructField {
+ field_name: Some(Ident::new("v")),
+ field_type: DataType::Varchar(None),
+ },
+ StructField {
+ field_name: Some(Ident::new("s")),
+ field_type: DataType::Struct(
+ vec![
+ StructField {
+ field_name: Some(Ident::new("a1")),
+ field_type: DataType::Integer(None),
+ },
+ StructField {
+ field_name: Some(Ident::new("a2")),
+ field_type: DataType::Varchar(None),
+ },
+ ],
+ StructBracketKind::Parentheses,
+ ),
+ },
+ ],
+ StructBracketKind::Parentheses,
+ );
+
+ // nested struct
+ let statement = duckdb().verified_stmt(
+ r#"CREATE TABLE t1 (s STRUCT(v VARCHAR, s STRUCT(a1 INTEGER, a2 VARCHAR))[])"#,
+ );
+
+ assert_eq!(
+ column_defs(statement),
+ vec![ColumnDef {
+ name: "s".into(),
+ data_type: DataType::Array(ArrayElemTypeDef::SquareBracket(
+ Box::new(struct_type2),
+ None
+ )),
+ collation: None,
+ options: vec![],
+ }]
+ );
+
+ // failing test (duckdb does not support bracket syntax)
+ let sql_list = vec![
+ r#"CREATE TABLE t1 (s STRUCT(v VARCHAR, i INTEGER)))"#,
+ r#"CREATE TABLE t1 (s STRUCT(v VARCHAR, i INTEGER>)"#,
+ r#"CREATE TABLE t1 (s STRUCT<v VARCHAR, i INTEGER>)"#,
+ r#"CREATE TABLE t1 (s STRUCT v VARCHAR, i INTEGER )"#,
+ r#"CREATE TABLE t1 (s STRUCT VARCHAR, i INTEGER )"#,
+ r#"CREATE TABLE t1 (s STRUCT (VARCHAR, INTEGER))"#,
+ ];
+
+ for sql in sql_list {
+ duckdb().parse_sql_statements(sql).unwrap_err();
+ }
+}
+
+/// Returns the ColumnDefinitions from a CreateTable statement
+fn column_defs(statement: Statement) -> Vec<ColumnDef> {
+ match statement {
+ Statement::CreateTable(CreateTable { columns, .. }) => columns,
+ _ => panic!("Expected CreateTable"),
+ }
+}
+
 #[test]
 fn test_select_wildcard_with_exclude() {
  let select = duckdb().verified_only_select("SELECT * EXCLUDE (col_a) FROM data");