refactor(regular_expression): Misc fixes (#6234)

Preparation for #6141

- Keep `enum` size + add size asserts tests
- Arrange AST related directories
- Renaming

crates/oxc_ast/src/generated/assert_layouts.rs

@@ -1445,7 +1445,7 @@ const _: () = {
     assert!(offset_of!(Alternative, span) == 0usize);
     assert!(offset_of!(Alternative, body) == 8usize);
 
-    assert!(size_of::<Term>() == 24usize);
+    assert!(size_of::<Term>() == 16usize);
     assert!(align_of::<Term>() == 8usize);
 
     assert!(size_of::<BoundaryAssertion>() == 12usize);
@@ -1465,7 +1465,7 @@ const _: () = {
     assert!(size_of::<LookAroundAssertionKind>() == 1usize);
     assert!(align_of::<LookAroundAssertionKind>() == 1usize);
 
-    assert!(size_of::<Quantifier>() == 64usize);
+    assert!(size_of::<Quantifier>() == 56usize);
     assert!(align_of::<Quantifier>() == 8usize);
     assert!(offset_of!(Quantifier, span) == 0usize);
     assert!(offset_of!(Quantifier, min) == 8usize);
@@ -1513,7 +1513,7 @@ const _: () = {
     assert!(size_of::<CharacterClassContentsKind>() == 1usize);
     assert!(align_of::<CharacterClassContentsKind>() == 1usize);
 
-    assert!(size_of::<CharacterClassContents>() == 24usize);
+    assert!(size_of::<CharacterClassContents>() == 16usize);
     assert!(align_of::<CharacterClassContents>() == 8usize);
 
     assert!(size_of::<CharacterClassRange>() == 40usize);
@@ -3000,7 +3000,7 @@ const _: () = {
     assert!(offset_of!(Alternative, span) == 0usize);
     assert!(offset_of!(Alternative, body) == 8usize);
 
-    assert!(size_of::<Term>() == 20usize);
+    assert!(size_of::<Term>() == 12usize);
     assert!(align_of::<Term>() == 4usize);
 
     assert!(size_of::<BoundaryAssertion>() == 12usize);
@@ -3020,7 +3020,7 @@ const _: () = {
     assert!(size_of::<LookAroundAssertionKind>() == 1usize);
     assert!(align_of::<LookAroundAssertionKind>() == 1usize);
 
-    assert!(size_of::<Quantifier>() == 56usize);
+    assert!(size_of::<Quantifier>() == 48usize);
     assert!(align_of::<Quantifier>() == 8usize);
     assert!(offset_of!(Quantifier, span) == 0usize);
     assert!(offset_of!(Quantifier, min) == 8usize);
@@ -3068,7 +3068,7 @@ const _: () = {
     assert!(size_of::<CharacterClassContentsKind>() == 1usize);
     assert!(align_of::<CharacterClassContentsKind>() == 1usize);
 
-    assert!(size_of::<CharacterClassContents>() == 20usize);
+    assert!(size_of::<CharacterClassContents>() == 8usize);
     assert!(align_of::<CharacterClassContents>() == 4usize);
 
     assert!(size_of::<CharacterClassRange>() == 40usize);

crates/oxc_regular_expression/src/ast.rs

@@ -1,7 +1,3 @@
-// NB: `#[span]`, `#[scope(...)]`,`#[visit(...)]` and `#[generate_derive(...)]` do NOT do anything to the code.
-// They are purely markers for codegen used in `tasks/ast_tools` and `crates/oxc_traverse/scripts`. See docs in those crates.
-// Read [`macro@oxc_ast_macros::ast`] for more information.
-
 // Silence erroneous warnings from Rust Analyser for `#[derive(Tsify)]`
 #![allow(non_snake_case)]
 
@@ -76,19 +72,19 @@ pub struct Alternative<'a> {
 #[cfg_attr(feature = "serialize", derive(Serialize, Tsify))]
 pub enum Term<'a> {
     // Assertion, QuantifiableAssertion
-    BoundaryAssertion(BoundaryAssertion) = 0,
+    BoundaryAssertion(Box<'a, BoundaryAssertion>) = 0,
     LookAroundAssertion(Box<'a, LookAroundAssertion<'a>>) = 1,
     // Quantifier
     Quantifier(Box<'a, Quantifier<'a>>) = 2,
     // Atom, ExtendedAtom
-    Character(Character) = 3,
+    Character(Box<'a, Character>) = 3,
     Dot(Dot) = 4,
-    CharacterClassEscape(CharacterClassEscape) = 5,
+    CharacterClassEscape(Box<'a, CharacterClassEscape>) = 5,
     UnicodePropertyEscape(Box<'a, UnicodePropertyEscape<'a>>) = 6,
     CharacterClass(Box<'a, CharacterClass<'a>>) = 7,
     CapturingGroup(Box<'a, CapturingGroup<'a>>) = 8,
     IgnoreGroup(Box<'a, IgnoreGroup<'a>>) = 9,
-    IndexedReference(IndexedReference) = 10,
+    IndexedReference(Box<'a, IndexedReference>) = 10,
     NamedReference(Box<'a, NamedReference<'a>>) = 11,
 }
 
@@ -286,9 +282,9 @@ pub enum CharacterClassContentsKind {
 #[cfg_attr(feature = "serialize", derive(Serialize, Tsify))]
 pub enum CharacterClassContents<'a> {
     CharacterClassRange(Box<'a, CharacterClassRange>) = 0,
-    CharacterClassEscape(CharacterClassEscape) = 1,
+    CharacterClassEscape(Box<'a, CharacterClassEscape>) = 1,
     UnicodePropertyEscape(Box<'a, UnicodePropertyEscape<'a>>) = 2,
-    Character(Character) = 3,
+    Character(Box<'a, Character>) = 3,
     /// `UnicodeSetsMode` only
     NestedCharacterClass(Box<'a, CharacterClass<'a>>) = 4,
     /// `UnicodeSetsMode` only
@@ -404,3 +400,13 @@ pub struct NamedReference<'a> {
     pub span: Span,
     pub name: Atom<'a>,
 }
+
+// See `oxc_ast/src/lib.rs` for the details
+#[cfg(target_pointer_width = "64")]
+#[test]
+fn size_asserts() {
+    use std::mem::size_of;
+
+    assert!(size_of::<Term>() == 16);
+    assert!(size_of::<CharacterClassContents>() == 16);
+}

crates/oxc_regular_expression/src/display.rs → crates/oxc_regular_expression/src/ast_impl/display.rs

@@ -78,17 +78,17 @@ impl<'a> Display for Alternative<'a> {
 impl<'a> Display for Term<'a> {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         match self {
-            Self::BoundaryAssertion(it) => write!(f, "{it}"),
+            Self::BoundaryAssertion(it) => write!(f, "{}", it.as_ref()),
             Self::LookAroundAssertion(it) => write!(f, "{}", it.as_ref()),
             Self::Quantifier(it) => write!(f, "{}", it.as_ref()),
-            Self::Character(it) => write!(f, "{it}"),
+            Self::Character(it) => write!(f, "{}", it.as_ref()),
             Self::Dot(it) => write!(f, "{it}"),
-            Self::CharacterClassEscape(it) => write!(f, "{it}"),
+            Self::CharacterClassEscape(it) => write!(f, "{}", it.as_ref()),
             Self::UnicodePropertyEscape(it) => write!(f, "{}", it.as_ref()),
             Self::CharacterClass(it) => write!(f, "{}", it.as_ref()),
             Self::CapturingGroup(it) => write!(f, "{}", it.as_ref()),
             Self::IgnoreGroup(it) => write!(f, "{}", it.as_ref()),
-            Self::IndexedReference(it) => write!(f, "{it}"),
+            Self::IndexedReference(it) => write!(f, "{}", it.as_ref()),
             Self::NamedReference(it) => write!(f, "{}", it.as_ref()),
         }
     }
@@ -246,9 +246,9 @@ impl<'a> Display for CharacterClassContents<'a> {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         match self {
             Self::CharacterClassRange(it) => write!(f, "{}", it.as_ref()),
-            Self::CharacterClassEscape(it) => write!(f, "{it}"),
+            Self::CharacterClassEscape(it) => write!(f, "{}", it.as_ref()),
             Self::UnicodePropertyEscape(it) => write!(f, "{}", it.as_ref()),
-            Self::Character(it) => write!(f, "{it}"),
+            Self::Character(it) => write!(f, "{}", it.as_ref()),
             Self::NestedCharacterClass(it) => write!(f, "{}", it.as_ref()),
             Self::ClassStringDisjunction(it) => write!(f, "{}", it.as_ref()),
         }

crates/oxc_regular_expression/src/ast_impl/mod.rs

@@ -0,0 +1,2 @@
+mod display;
+pub mod visit;

crates/oxc_regular_expression/src/visit.rs → crates/oxc_regular_expression/src/ast_impl/visit.rs

@@ -1,3 +1,6 @@
+// NOTE: For now, this file is implemented by hand for convenience.
+// But like `oxc_ast`, this should be generated by `tasks/ast_tools` in the future.
+
 #![allow(unused_variables, clippy::wildcard_imports)]
 use oxc_span::{GetSpan, Span};
 use walk::walk_pattern;