Support bigquery CAST AS x [STRING|DATE] FORMAT syntax (#978)

apache · Oct 20, 2023 · c68e977 · c68e977
1 parent 83cb734
commit c68e977
Show file tree

Hide file tree

Showing 6 changed files with 130 additions and 6 deletions.
diff --git a/src/ast/mod.rs b/src/ast/mod.rs
@@ -322,6 +322,16 @@ impl fmt::Display for JsonOperator {
     }
 }
 
+/// 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)]
+#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "visitor", derive(Visit, VisitMut))]
+pub enum CastFormat {
+    Value(Value),
+    ValueAtTimeZone(Value, Value),
+}
+
 /// An SQL expression of any type.
 ///
 /// The parser does not distinguish between expressions of different types
@@ -437,19 +447,28 @@ pub enum Expr {
     Cast {
         expr: Box<Expr>,
         data_type: DataType,
+        // Optional CAST(string_expression AS type FORMAT format_string_expression) as used by BigQuery
+        // https://cloud.google.com/bigquery/docs/reference/standard-sql/format-elements#formatting_syntax
+        format: Option<CastFormat>,
     },
     /// TRY_CAST an expression to a different data type e.g. `TRY_CAST(foo AS VARCHAR(123))`
     //  this differs from CAST in the choice of how to implement invalid conversions
     TryCast {
         expr: Box<Expr>,
         data_type: DataType,
+        // Optional CAST(string_expression AS type FORMAT format_string_expression) as used by BigQuery
+        // https://cloud.google.com/bigquery/docs/reference/standard-sql/format-elements#formatting_syntax
+        format: Option<CastFormat>,
     },
     /// SAFE_CAST an expression to a different data type e.g. `SAFE_CAST(foo AS FLOAT64)`
     //  only available for BigQuery: https://cloud.google.com/bigquery/docs/reference/standard-sql/functions-and-operators#safe_casting
     //  this works the same as `TRY_CAST`
     SafeCast {
         expr: Box<Expr>,
         data_type: DataType,
+        // Optional CAST(string_expression AS type FORMAT format_string_expression) as used by BigQuery
+        // https://cloud.google.com/bigquery/docs/reference/standard-sql/format-elements#formatting_syntax
+        format: Option<CastFormat>,
     },
     /// AT a timestamp to a different timezone e.g. `FROM_UNIXTIME(0) AT TIME ZONE 'UTC-06:00'`
     AtTimeZone {
@@ -597,6 +616,15 @@ pub enum Expr {
     },
 }
 
+impl fmt::Display for CastFormat {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        match self {
+            CastFormat::Value(v) => write!(f, "{v}"),
+            CastFormat::ValueAtTimeZone(v, tz) => write!(f, "{v} AT TIME ZONE {tz}"),
+        }
+    }
+}
+
 impl fmt::Display for Expr {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         match self {
@@ -753,9 +781,39 @@ impl fmt::Display for Expr {
                     write!(f, "{op}{expr}")
                 }
             }
-            Expr::Cast { expr, data_type } => write!(f, "CAST({expr} AS {data_type})"),
-            Expr::TryCast { expr, data_type } => write!(f, "TRY_CAST({expr} AS {data_type})"),
-            Expr::SafeCast { expr, data_type } => write!(f, "SAFE_CAST({expr} AS {data_type})"),
+            Expr::Cast {
+                expr,
+                data_type,
+                format,
+            } => {
+                if let Some(format) = format {
+                    write!(f, "CAST({expr} AS {data_type} FORMAT {format})")
+                } else {
+                    write!(f, "CAST({expr} AS {data_type})")
+                }
+            }
+            Expr::TryCast {
+                expr,
+                data_type,
+                format,
+            } => {
+                if let Some(format) = format {
+                    write!(f, "TRY_CAST({expr} AS {data_type} FORMAT {format})")
+                } else {
+                    write!(f, "TRY_CAST({expr} AS {data_type})")
+                }
+            }
+            Expr::SafeCast {
+                expr,
+                data_type,
+                format,
+            } => {
+                if let Some(format) = format {
+                    write!(f, "SAFE_CAST({expr} AS {data_type} FORMAT {format})")
+                } else {
+                    write!(f, "SAFE_CAST({expr} AS {data_type})")
+                }
+            }
             Expr::Extract { field, expr } => write!(f, "EXTRACT({field} FROM {expr})"),
             Expr::Ceil { expr, field } => {
                 if field == &DateTimeField::NoDateTime {

diff --git a/src/parser/mod.rs b/src/parser/mod.rs
@@ -1139,16 +1139,34 @@ impl<'a> Parser<'a> {
         })
     }
 
+    pub fn parse_optional_cast_format(&mut self) -> Result<Option<CastFormat>, ParserError> {
+        if self.parse_keyword(Keyword::FORMAT) {
+            let value = self.parse_value()?;
+            if self.parse_keywords(&[Keyword::AT, Keyword::TIME, Keyword::ZONE]) {
+                Ok(Some(CastFormat::ValueAtTimeZone(
+                    value,
+                    self.parse_value()?,
+                )))
+            } else {
+                Ok(Some(CastFormat::Value(value)))
+            }
+        } else {
+            Ok(None)
+        }
+    }
+
     /// 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)?;
         let expr = self.parse_expr()?;
         self.expect_keyword(Keyword::AS)?;
         let data_type = self.parse_data_type()?;
+        let format = self.parse_optional_cast_format()?;
         self.expect_token(&Token::RParen)?;
         Ok(Expr::Cast {
             expr: Box::new(expr),
             data_type,
+            format,
         })
     }
 
@@ -1158,10 +1176,12 @@ impl<'a> Parser<'a> {
         let expr = self.parse_expr()?;
         self.expect_keyword(Keyword::AS)?;
         let data_type = self.parse_data_type()?;
+        let format = self.parse_optional_cast_format()?;
         self.expect_token(&Token::RParen)?;
         Ok(Expr::TryCast {
             expr: Box::new(expr),
             data_type,
+            format,
         })
     }
 
@@ -1171,10 +1191,12 @@ impl<'a> Parser<'a> {
         let expr = self.parse_expr()?;
         self.expect_keyword(Keyword::AS)?;
         let data_type = self.parse_data_type()?;
+        let format = self.parse_optional_cast_format()?;
         self.expect_token(&Token::RParen)?;
         Ok(Expr::SafeCast {
             expr: Box::new(expr),
             data_type,
+            format,
         })
     }
 
@@ -2101,6 +2123,7 @@ impl<'a> Parser<'a> {
         Ok(Expr::Cast {
             expr: Box::new(expr),
             data_type: self.parse_data_type()?,
+            format: None,
         })
     }
 

diff --git a/tests/sqlparser_bigquery.rs b/tests/sqlparser_bigquery.rs
@@ -304,8 +304,39 @@ fn parse_trailing_comma() {
 
 #[test]
 fn parse_cast_type() {
-    let sql = r#"SELECT SAFE_CAST(1 AS INT64)"#;
-    bigquery().verified_only_select(sql);
+    let sql = r"SELECT SAFE_CAST(1 AS INT64)";
+    bigquery_and_generic().verified_only_select(sql);
+}
+
+#[test]
+fn parse_cast_date_format() {
+    let sql =
+        r"SELECT CAST(date_valid_from AS DATE FORMAT 'YYYY-MM-DD') AS date_valid_from FROM foo";
+    bigquery_and_generic().verified_only_select(sql);
+}
+
+#[test]
+fn parse_cast_time_format() {
+    let sql = r"SELECT CAST(TIME '21:30:00' AS STRING FORMAT 'PM') AS date_time_to_string";
+    bigquery_and_generic().verified_only_select(sql);
+}
+
+#[test]
+fn parse_cast_timestamp_format_tz() {
+    let sql = r"SELECT CAST(TIMESTAMP '2008-12-25 00:00:00+00:00' AS STRING FORMAT 'TZH' AT TIME ZONE 'Asia/Kolkata') AS date_time_to_string";
+    bigquery_and_generic().verified_only_select(sql);
+}
+
+#[test]
+fn parse_cast_string_to_bytes_format() {
+    let sql = r"SELECT CAST('Hello' AS BYTES FORMAT 'ASCII') AS string_to_bytes";
+    bigquery_and_generic().verified_only_select(sql);
+}
+
+#[test]
+fn parse_cast_bytes_to_string_format() {
+    let sql = r"SELECT CAST(B'\x48\x65\x6c\x6c\x6f' AS STRING FORMAT 'ASCII') AS bytes_to_string";
+    bigquery_and_generic().verified_only_select(sql);
 }
 
 #[test]

diff --git a/tests/sqlparser_common.rs b/tests/sqlparser_common.rs
@@ -1934,6 +1934,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::BigInt(None),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -1944,6 +1945,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::TinyInt(None),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -1970,6 +1972,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Nvarchar(Some(50)),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -1980,6 +1983,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Clob(None),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -1990,6 +1994,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Clob(Some(50)),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -2000,6 +2005,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Binary(Some(50)),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -2010,6 +2016,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Varbinary(Some(50)),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -2020,6 +2027,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Blob(None),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -2030,6 +2038,7 @@ fn parse_cast() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::Blob(Some(50)),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );
@@ -2043,6 +2052,7 @@ fn parse_try_cast() {
         &Expr::TryCast {
             expr: Box::new(Expr::Identifier(Ident::new("id"))),
             data_type: DataType::BigInt(None),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );

diff --git a/tests/sqlparser_postgres.rs b/tests/sqlparser_postgres.rs
@@ -1782,7 +1782,8 @@ fn parse_array_index_expr() {
                 })),
                 data_type: DataType::Array(Some(Box::new(DataType::Array(Some(Box::new(
                     DataType::Int(None)
-                ))))))
+                )))))),
+                format: None,
             }))),
             indexes: vec![num[1].clone(), num[2].clone()],
         },

diff --git a/tests/sqlparser_snowflake.rs b/tests/sqlparser_snowflake.rs
@@ -167,6 +167,7 @@ fn parse_array() {
         &Expr::Cast {
             expr: Box::new(Expr::Identifier(Ident::new("a"))),
             data_type: DataType::Array(None),
+            format: None,
         },
         expr_from_projection(only(&select.projection))
     );