Add CREATE TABLE AS support (#206)

We parse it as a regular `CREATE TABLE` statement
followed by an `AS <query>`, which is how BigQuery works:
https://cloud.google.com/bigquery/docs/reference/standard-sql/data-definition-language#create_table_statement


ANSI SQL and PostgreSQL only support a plain list of columns
after the table name in a CTAS
    `CREATE TABLE t (a) AS SELECT a FROM foo`

We currently only allow specifying a full schema with data
types, or omitting it altogether.

https://www.postgresql.org/docs/12/sql-createtableas.html
https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#as-subquery-clause


Finally, when no schema is specified, we print empty parens after a
plain `CREATE TABLE t ();` as required by PostgreSQL, but skip them
in a CTAS: `CREATE TABLE t AS ...`. This affects serialization only,
the parser allows omitting the schema in a regular `CREATE TABLE` too
since the first release of the parser:
https://github.com/benesch/sqlparser-rs/blame/7d27abdfb4ac4a0980d79b12d9d8ef156434e984/src/sqlparser.rs#L325-L332

Co-authored-by: Nickolay Ponomarev <[email protected]>

Author: Dandandan

CHANGELOG.md

@@ -32,6 +32,7 @@ Check https://github.com/andygrove/sqlparser-rs/commits/master for undocumented
 - Add serde support to AST structs and enums (#196) - thanks @panarch!
 - Support `ALTER TABLE ADD COLUMN`, `RENAME COLUMN`, and `RENAME TO` (#203) - thanks @mashuai!
 - Support `ALTER TABLE DROP COLUMN` (#148) - thanks @ivanceras!
+- Support `CREATE TABLE ... AS ...` (#206) - thanks @Dandandan!
 
 ### Fixed
 - Report an error for unterminated string literals (#165)

src/ast/mod.rs

@@ -481,6 +481,7 @@ pub enum Statement {
         external: bool,
         file_format: Option<FileFormat>,
         location: Option<String>,
+        query: Option<Box<Query>>,
     },
     /// CREATE INDEX
     CreateIndex {
@@ -645,19 +646,32 @@ impl fmt::Display for Statement {
                 external,
                 file_format,
                 location,
+                query,
             } => {
+                // We want to allow the following options
+                // Empty column list, allowed by PostgreSQL:
+                //   `CREATE TABLE t ()`
+                // No columns provided for CREATE TABLE AS:
+                //   `CREATE TABLE t AS SELECT a from t2`
+                // Columns provided for CREATE TABLE AS:
+                //   `CREATE TABLE t (a INT) AS SELECT a from t2`
                 write!(
                     f,
-                    "CREATE {}TABLE {}{} ({}",
-                    if *external { "EXTERNAL " } else { "" },
-                    if *if_not_exists { "IF NOT EXISTS " } else { "" },
-                    name,
-                    display_comma_separated(columns)
+                    "CREATE {external}TABLE {if_not_exists}{name}",
+                    external = if *external { "EXTERNAL " } else { "" },
+                    if_not_exists = if *if_not_exists { "IF NOT EXISTS " } else { "" },
+                    name = name,
                 )?;
-                if !constraints.is_empty() {
-                    write!(f, ", {}", display_comma_separated(constraints))?;
+                if !columns.is_empty() || !constraints.is_empty() {
+                    write!(f, " ({}", display_comma_separated(columns))?;
+                    if !columns.is_empty() && !constraints.is_empty() {
+                        write!(f, ", ")?;
+                    }
+                    write!(f, "{})", display_comma_separated(constraints))?;
+                } else if query.is_none() {
+                    // PostgreSQL allows `CREATE TABLE t ();`, but requires empty parens
+                    write!(f, " ()")?;
                 }
-                write!(f, ")")?;
 
                 if *external {
                     write!(
@@ -670,6 +684,9 @@ impl fmt::Display for Statement {
                 if !with_options.is_empty() {
                     write!(f, " WITH ({})", display_comma_separated(with_options))?;
                 }
+                if let Some(query) = query {
+                    write!(f, " AS {}", query)?;
+                }
                 Ok(())
             }
             Statement::CreateIndex {

src/parser.rs

@@ -1020,6 +1020,7 @@ impl Parser {
             external: true,
             file_format: Some(file_format),
             location: Some(location),
+            query: None,
         })
     }
 
@@ -1108,8 +1109,17 @@ impl Parser {
         let table_name = self.parse_object_name()?;
         // parse optional column list (schema)
         let (columns, constraints) = self.parse_columns()?;
+
+        // PostgreSQL supports `WITH ( options )`, before `AS`
         let with_options = self.parse_with_options()?;
 
+        // Parse optional `AS ( query )`
+        let query = if self.parse_keyword(Keyword::AS) {
+            Some(Box::new(self.parse_query()?))
+        } else {
+            None
+        };
+
         Ok(Statement::CreateTable {
             name: table_name,
             columns,
@@ -1119,6 +1129,7 @@ impl Parser {
             external: false,
             file_format: None,
             location: None,
+            query,
         })
     }
 

tests/sqlparser_common.rs

@@ -1044,6 +1044,7 @@ fn parse_create_table() {
             external: false,
             file_format: None,
             location: None,
+            query: _query,
         } => {
             assert_eq!("uk_cities", name.to_string());
             assert_eq!(
@@ -1177,6 +1178,36 @@ fn parse_drop_schema() {
     }
 }
 
+#[test]
+fn parse_create_table_as() {
+    let sql = "CREATE TABLE t AS SELECT * FROM a";
+
+    match verified_stmt(sql) {
+        Statement::CreateTable { name, query, .. } => {
+            assert_eq!(name.to_string(), "t".to_string());
+            assert_eq!(query, Some(Box::new(verified_query("SELECT * FROM a"))));
+        }
+        _ => unreachable!(),
+    }
+
+    // BigQuery allows specifying table schema in CTAS
+    // ANSI SQL and PostgreSQL let you only specify the list of columns
+    // (without data types) in a CTAS, but we have yet to support that.
+    let sql = "CREATE TABLE t (a INT, b INT) AS SELECT 1 AS b, 2 AS a";
+    match verified_stmt(sql) {
+        Statement::CreateTable { columns, query, .. } => {
+            assert_eq!(columns.len(), 2);
+            assert_eq!(columns[0].to_string(), "a INT".to_string());
+            assert_eq!(columns[1].to_string(), "b INT".to_string());
+            assert_eq!(
+                query,
+                Some(Box::new(verified_query("SELECT 1 AS b, 2 AS a")))
+            );
+        }
+        _ => unreachable!(),
+    }
+}
+
 #[test]
 fn parse_create_table_with_on_delete_on_update_2in_any_order() -> Result<(), ParserError> {
     let sql = |options: &str| -> String {
@@ -1245,6 +1276,7 @@ fn parse_create_external_table() {
             external,
             file_format,
             location,
+            query: _query,
         } => {
             assert_eq!("uk_cities", name.to_string());
             assert_eq!(
@@ -1307,12 +1339,6 @@ fn parse_create_external_table_lowercase() {
     assert_matches!(ast, Statement::CreateTable{..});
 }
 
-#[test]
-fn parse_create_table_empty() {
-    // Zero-column tables are weird, but supported by at least PostgreSQL.
-    let _ = verified_stmt("CREATE TABLE t ()");
-}
-
 #[test]
 fn parse_alter_table() {
     let add_column = "ALTER TABLE tab ADD COLUMN foo TEXT";

tests/sqlparser_mysql.rs

@@ -77,7 +77,7 @@ fn parse_show_columns() {
         Statement::ShowColumns {
             extended: false,
             full: false,
-            table_name: table_name,
+            table_name,
             filter: Some(ShowStatementFilter::Where(
                 mysql_and_generic().verified_expr("1 = 2")
             )),

tests/sqlparser_postgres.rs

@@ -43,6 +43,7 @@ fn parse_create_table_with_defaults() {
             external: false,
             file_format: None,
             location: None,
+            query: _query,
         } => {
             assert_eq!("public.customer", name.to_string());
             assert_eq!(
@@ -226,25 +227,47 @@ fn parse_create_table_with_inherit() {
     pg().verified_stmt(sql);
 }
 
+#[test]
+fn parse_create_table_empty() {
+    // Zero-column tables are weird, but supported by at least PostgreSQL.
+    // <https://github.com/andygrove/sqlparser-rs/pull/94>
+    let _ = pg_and_generic().verified_stmt("CREATE TABLE t ()");
+}
+
+#[test]
+fn parse_create_table_constraints_only() {
+    // Zero-column tables can also have constraints in PostgreSQL
+    let sql = "CREATE TABLE t (CONSTRAINT positive CHECK (2 > 1))";
+    let ast = pg_and_generic().verified_stmt(sql);
+    match ast {
+        Statement::CreateTable {
+            name,
+            columns,
+            constraints,
+            ..
+        } => {
+            assert_eq!("t", name.to_string());
+            assert!(columns.is_empty());
+            assert_eq!(
+                only(constraints).to_string(),
+                "CONSTRAINT positive CHECK (2 > 1)"
+            );
+        }
+        _ => unreachable!(),
+    };
+}
+
 #[test]
 fn parse_create_table_if_not_exists() {
     let sql = "CREATE TABLE IF NOT EXISTS uk_cities ()";
-    let ast =
-        pg_and_generic().one_statement_parses_to(sql, "CREATE TABLE IF NOT EXISTS uk_cities ()");
+    let ast = pg_and_generic().verified_stmt(sql);
     match ast {
         Statement::CreateTable {
             name,
-            columns: _columns,
-            constraints,
-            with_options,
             if_not_exists: true,
-            external: false,
-            file_format: None,
-            location: None,
+            ..
         } => {
             assert_eq!("uk_cities", name.to_string());
-            assert!(constraints.is_empty());
-            assert_eq!(with_options, vec![]);
         }
         _ => unreachable!(),
     }