Add an AST visitor

Cargo.toml

@@ -5,7 +5,7 @@ version = "0.27.0"
 authors = ["Andy Grove <andygrove73@gmail.com>"]
 homepage = "https://github.com/sqlparser-rs/sqlparser-rs"
 documentation = "https://docs.rs/sqlparser/"
-keywords = [ "ansi", "sql", "lexer", "parser" ]
+keywords = ["ansi", "sql", "lexer", "parser"]
 repository = "https://github.com/sqlparser-rs/sqlparser-rs"
 license = "Apache-2.0"
 include = [
@@ -32,6 +32,7 @@ serde = { version = "1.0", features = ["derive"], optional = true }
 # of dev-dependencies because of
 # https://github.com/rust-lang/cargo/issues/1596
 serde_json = { version = "1.0", optional = true }
+derive-visitor = { version = "0.3.0", optional = true }
 
 [dev-dependencies]
 simple_logger = "4.0"

README.md

@@ -40,6 +40,25 @@ This outputs
 AST: [Query(Query { ctes: [], body: Select(Select { distinct: false, projection: [UnnamedExpr(Identifier("a")), UnnamedExpr(Identifier("b")), UnnamedExpr(Value(Long(123))), UnnamedExpr(Function(Function { name: ObjectName(["myfunc"]), args: [Identifier("b")], over: None, distinct: false }))], from: [TableWithJoins { relation: Table { name: ObjectName(["table_1"]), alias: None, args: [], with_hints: [] }, joins: [] }], selection: Some(BinaryOp { left: BinaryOp { left: Identifier("a"), op: Gt, right: Identifier("b") }, op: And, right: BinaryOp { left: Identifier("b"), op: Lt, right: Value(Long(100)) } }), group_by: [], having: None }), order_by: [OrderByExpr { expr: Identifier("a"), asc: Some(false) }, OrderByExpr { expr: Identifier("b"), asc: None }], limit: None, offset: None, fetch: None })]
 ```
 
+### Analyzing and transforming the AST
+
+Once you have an abstract syntax tree, you can analyze and transform it 
+using the optional [`derive-visitor`](https://github.com/nikis05/derive-visitor) feature.
+
+For instance, if you want to rename all identifiers in a query:
+
+```rust
+use sqlparser::{dialect::GenericDialect, parser::Parser, ast::Ident};
+ use derive_visitor::{visitor_enter_fn_mut, DriveMut};
+
+let mut statements = Parser::parse_sql(&GenericDialect, "select xxx").unwrap();
+statements[0].drive_mut(&mut visitor_enter_fn_mut(|ident: &mut Ident| {
+    ident.value = ident.value.replace("xxx", "yyy");
+}));
+assert_eq!(statements[0].to_string(), "SELECT yyy");
+```
+
+
 ## Command line
 To parse a file and dump the results as JSON:
 ```

src/ast/data_type.rs

@@ -17,13 +17,17 @@ use core::fmt;
 #[cfg(feature = "serde")]
 use serde::{Deserialize, Serialize};
 
+#[cfg(feature = "derive-visitor")]
+use derive_visitor::{Drive, DriveMut};
+
 use crate::ast::ObjectName;
 
 use super::value::escape_single_quote_string;
 
 /// SQL data types
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum DataType {
     /// Fixed-length character type e.g. CHARACTER(10)
     Character(Option<CharacterLength>),
@@ -36,37 +40,37 @@ pub enum DataType {
     /// Variable-length character type e.g. VARCHAR(10)
     Varchar(Option<CharacterLength>),
     /// Variable-length character type e.g. NVARCHAR(10)
-    Nvarchar(Option<u64>),
+    Nvarchar(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Uuid type
     Uuid,
     /// Large character object with optional length e.g. CHARACTER LARGE OBJECT, CHARACTER LARGE OBJECT(1000), [standard]
     ///
     /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#character-large-object-type
-    CharacterLargeObject(Option<u64>),
+    CharacterLargeObject(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Large character object with optional length e.g. CHAR LARGE OBJECT, CHAR LARGE OBJECT(1000), [standard]
     ///
     /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#character-large-object-type
-    CharLargeObject(Option<u64>),
+    CharLargeObject(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Large character object with optional length e.g. CLOB, CLOB(1000), [standard]
     ///
     /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#character-large-object-type
     /// [Oracle]: https://docs.oracle.com/javadb/10.10.1.2/ref/rrefclob.html
-    Clob(Option<u64>),
+    Clob(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Fixed-length binary type with optional length e.g.  [standard], [MS SQL Server]
     ///
     /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#binary-string-type
     /// [MS SQL Server]: https://learn.microsoft.com/pt-br/sql/t-sql/data-types/binary-and-varbinary-transact-sql?view=sql-server-ver16
-    Binary(Option<u64>),
+    Binary(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Variable-length binary with optional length type e.g. [standard], [MS SQL Server]
     ///
     /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#binary-string-type
     /// [MS SQL Server]: https://learn.microsoft.com/pt-br/sql/t-sql/data-types/binary-and-varbinary-transact-sql?view=sql-server-ver16
-    Varbinary(Option<u64>),
+    Varbinary(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Large binary object with optional length e.g. BLOB, BLOB(1000), [standard], [Oracle]
     ///
     /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#binary-large-object-string-type
     /// [Oracle]: https://docs.oracle.com/javadb/10.8.3.0/ref/rrefblob.html
-    Blob(Option<u64>),
+    Blob(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Numeric type with optional precision and scale e.g. NUMERIC(10,2), [standard][1]
     ///
     /// [1]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#exact-numeric-type
@@ -80,35 +84,35 @@ pub enum DataType {
     /// [1]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#exact-numeric-type
     Dec(ExactNumberInfo),
     /// Floating point with optional precision e.g. FLOAT(8)
-    Float(Option<u64>),
+    Float(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Tiny integer with optional display width e.g. TINYINT or TINYINT(3)
-    TinyInt(Option<u64>),
+    TinyInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Unsigned tiny integer with optional display width e.g. TINYINT UNSIGNED or TINYINT(3) UNSIGNED
-    UnsignedTinyInt(Option<u64>),
+    UnsignedTinyInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Small integer with optional display width e.g. SMALLINT or SMALLINT(5)
-    SmallInt(Option<u64>),
+    SmallInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Unsigned small integer with optional display width e.g. SMALLINT UNSIGNED or SMALLINT(5) UNSIGNED
-    UnsignedSmallInt(Option<u64>),
+    UnsignedSmallInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// MySQL medium integer ([1]) with optional display width e.g. MEDIUMINT or MEDIUMINT(5)
     ///
     /// [1]: https://dev.mysql.com/doc/refman/8.0/en/integer-types.html
-    MediumInt(Option<u64>),
+    MediumInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Unsigned medium integer ([1]) with optional display width e.g. MEDIUMINT UNSIGNED or MEDIUMINT(5) UNSIGNED
     ///
     /// [1]: https://dev.mysql.com/doc/refman/8.0/en/integer-types.html
-    UnsignedMediumInt(Option<u64>),
+    UnsignedMediumInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Integer with optional display width e.g. INT or INT(11)
-    Int(Option<u64>),
+    Int(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Integer with optional display width e.g. INTEGER or INTEGER(11)
-    Integer(Option<u64>),
+    Integer(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Unsigned integer with optional display width e.g. INT UNSIGNED or INT(11) UNSIGNED
-    UnsignedInt(Option<u64>),
+    UnsignedInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Unsigned integer with optional display width e.g. INTGER UNSIGNED or INTEGER(11) UNSIGNED
-    UnsignedInteger(Option<u64>),
+    UnsignedInteger(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Big integer with optional display width e.g. BIGINT or BIGINT(20)
-    BigInt(Option<u64>),
+    BigInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Unsigned big integer with optional display width e.g. BIGINT UNSIGNED or BIGINT(20) UNSIGNED
-    UnsignedBigInt(Option<u64>),
+    UnsignedBigInt(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Floating point e.g. REAL
     Real,
     /// Double
@@ -125,15 +129,21 @@ pub enum DataType {
     /// Time with optional time precision and time zone information e.g. [standard][1].
     ///
     /// [1]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#datetime-type
-    Time(Option<u64>, TimezoneInfo),
+    Time(
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] TimezoneInfo,
+    ),
     /// Datetime with optional time precision e.g. [MySQL][1].
     ///
     /// [1]: https://dev.mysql.com/doc/refman/8.0/en/datetime.html
-    Datetime(Option<u64>),
+    Datetime(#[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>),
     /// Timestamp with optional time precision and time zone information e.g. [standard][1].
     ///
     /// [1]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#datetime-type
-    Timestamp(Option<u64>, TimezoneInfo),
+    Timestamp(
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] Option<u64>,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] TimezoneInfo,
+    ),
     /// Interval
     Interval,
     /// Regclass used in postgresql serial
@@ -145,13 +155,16 @@ pub enum DataType {
     /// Bytea
     Bytea,
     /// Custom type such as enums
-    Custom(ObjectName, Vec<String>),
+    Custom(
+        ObjectName,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] Vec<String>,
+    ),
     /// Arrays
     Array(Option<Box<DataType>>),
     /// Enums
-    Enum(Vec<String>),
+    Enum(#[cfg_attr(feature = "derive-visitor", drive(skip))] Vec<String>),
     /// Set
-    Set(Vec<String>),
+    Set(#[cfg_attr(feature = "derive-visitor", drive(skip))] Vec<String>),
 }
 
 impl fmt::Display for DataType {
@@ -384,13 +397,17 @@ impl fmt::Display for TimezoneInfo {
 /// [standard]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#exact-numeric-type
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum ExactNumberInfo {
     /// No additional information e.g. `DECIMAL`
     None,
     /// Only precision information e.g. `DECIMAL(10)`
-    Precision(u64),
+    Precision(#[cfg_attr(feature = "derive-visitor", drive(skip))] u64),
     /// Precision and scale information e.g. `DECIMAL(10,2)`
-    PrecisionAndScale(u64, u64),
+    PrecisionAndScale(
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] u64,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))] u64,
+    ),
 }
 
 impl fmt::Display for ExactNumberInfo {
@@ -414,8 +431,10 @@ impl fmt::Display for ExactNumberInfo {
 /// [1]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#character-length
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub struct CharacterLength {
     /// Default (if VARYING) or maximum (if not VARYING) length
+    #[cfg_attr(feature = "derive-visitor", drive(skip))]
     pub length: u64,
     /// Optional unit. If not informed, the ANSI handles it as CHARACTERS implicitly
     pub unit: Option<CharLengthUnits>,
@@ -436,6 +455,7 @@ impl fmt::Display for CharacterLength {
 /// [1]: https://jakewheat.github.io/sql-overview/sql-2016-foundation-grammar.html#char-length-units
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum CharLengthUnits {
     /// CHARACTERS unit
     Characters,

src/ast/ddl.rs

@@ -20,28 +20,36 @@ use core::fmt;
 #[cfg(feature = "serde")]
 use serde::{Deserialize, Serialize};
 
+#[cfg(feature = "derive-visitor")]
+use derive_visitor::{Drive, DriveMut};
+
 use crate::ast::value::escape_single_quote_string;
 use crate::ast::{display_comma_separated, display_separated, DataType, Expr, Ident, ObjectName};
 use crate::tokenizer::Token;
 
 /// An `ALTER TABLE` (`Statement::AlterTable`) operation
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum AlterTableOperation {
     /// `ADD <table_constraint>`
     AddConstraint(TableConstraint),
     /// `ADD [ COLUMN ] <column_def>`
     AddColumn { column_def: ColumnDef },
     /// `DROP CONSTRAINT [ IF EXISTS ] <name>`
     DropConstraint {
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         if_exists: bool,
         name: Ident,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         cascade: bool,
     },
     /// `DROP [ COLUMN ] [ IF EXISTS ] <column_name> [ CASCADE ]`
     DropColumn {
         column_name: Ident,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         if_exists: bool,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         cascade: bool,
     },
     /// `DROP PRIMARY KEY`
@@ -55,11 +63,13 @@ pub enum AlterTableOperation {
     },
     /// Add Partitions
     AddPartitions {
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         if_not_exists: bool,
         new_partitions: Vec<Expr>,
     },
     DropPartitions {
         partitions: Vec<Expr>,
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         if_exists: bool,
     },
     /// `RENAME [ COLUMN ] <old_column_name> TO <new_column_name>`
@@ -183,6 +193,7 @@ impl fmt::Display for AlterTableOperation {
 /// An `ALTER COLUMN` (`Statement::AlterTable`) operation
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum AlterColumnOperation {
     /// `SET NOT NULL`
     SetNotNull,
@@ -226,12 +237,14 @@ impl fmt::Display for AlterColumnOperation {
 /// `ALTER TABLE ADD <constraint>` statement.
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum TableConstraint {
     /// `[ CONSTRAINT <name> ] { PRIMARY KEY | UNIQUE } (<columns>)`
     Unique {
         name: Option<Ident>,
         columns: Vec<Ident>,
         /// Whether this is a `PRIMARY KEY` or just a `UNIQUE` constraint
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         is_primary: bool,
     },
     /// A referential integrity constraint (`[ CONSTRAINT <name> ] FOREIGN KEY (<columns>)
@@ -260,6 +273,7 @@ pub enum TableConstraint {
     /// [1]: https://dev.mysql.com/doc/refman/8.0/en/create-table.html
     Index {
         /// Whether this index starts with KEY (true) or INDEX (false), to maintain the same syntax.
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         display_as_key: bool,
         /// Index name.
         name: Option<Ident>,
@@ -284,6 +298,7 @@ pub enum TableConstraint {
     /// [1]: https://dev.mysql.com/doc/refman/8.0/en/fulltext-natural-language.html
     FulltextOrSpatial {
         /// Whether this is a `FULLTEXT` (true) or `SPATIAL` (false) definition.
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         fulltext: bool,
         /// Whether the type is followed by the keyword `KEY`, `INDEX`, or no keyword at all.
         index_type_display: KeyOrIndexDisplay,
@@ -389,6 +404,7 @@ impl fmt::Display for TableConstraint {
 /// [1]: https://dev.mysql.com/doc/refman/8.0/en/create-table.html
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum KeyOrIndexDisplay {
     /// Nothing to display
     None,
@@ -424,6 +440,7 @@ impl fmt::Display for KeyOrIndexDisplay {
 /// [3]: https://www.postgresql.org/docs/14/sql-createindex.html
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum IndexType {
     BTree,
     Hash,
@@ -442,6 +459,7 @@ impl fmt::Display for IndexType {
 /// SQL column definition
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub struct ColumnDef {
     pub name: Ident,
     pub data_type: DataType,
@@ -477,6 +495,7 @@ impl fmt::Display for ColumnDef {
 /// "column options," and we allow any column option to be named.
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub struct ColumnOptionDef {
     pub name: Option<Ident>,
     pub option: ColumnOption,
@@ -492,6 +511,7 @@ impl fmt::Display for ColumnOptionDef {
 /// TABLE` statement.
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum ColumnOption {
     /// `NULL`
     Null,
@@ -501,6 +521,7 @@ pub enum ColumnOption {
     Default(Expr),
     /// `{ PRIMARY KEY | UNIQUE }`
     Unique {
+        #[cfg_attr(feature = "derive-visitor", drive(skip))]
         is_primary: bool,
     },
     /// A referential integrity constraint (`[FOREIGN KEY REFERENCES
@@ -521,7 +542,7 @@ pub enum ColumnOption {
     /// - ...
     DialectSpecific(Vec<Token>),
     CharacterSet(ObjectName),
-    Comment(String),
+    Comment(#[cfg_attr(feature = "derive-visitor", drive(skip))] String),
 }
 
 impl fmt::Display for ColumnOption {
@@ -579,6 +600,7 @@ fn display_constraint_name(name: &'_ Option<Ident>) -> impl fmt::Display + '_ {
 /// Used in foreign key constraints in `ON UPDATE` and `ON DELETE` options.
 #[derive(Debug, Clone, PartialEq, Eq, Hash)]
 #[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
+#[cfg_attr(feature = "derive-visitor", derive(Drive, DriveMut))]
 pub enum ReferentialAction {
     Restrict,
     Cascade,