refactor(parser): remove TokenValue::RegExp from Token (#1926)

This PR is part of #1880.

`Token` size is reduced from 48 to 40 bytes.

To reconstruct the regex pattern and flags within the parser , the regex
string is
re-parsed from the end by reading all valid flags.

In order to make things work nicely, the lexer will no longer recover
from a invalid regex.

crates/oxc_ast/src/ast_builder.rs

@@ -163,8 +163,13 @@ impl<'a> AstBuilder<'a> {
         TemplateElementValue { raw, cooked }
     }
 
-    pub fn reg_exp_literal(&self, span: Span, pattern: Atom, flags: RegExpFlags) -> RegExpLiteral {
-        RegExpLiteral { span, value: EmptyObject, regex: RegExp { pattern, flags } }
+    pub fn reg_exp_literal(
+        &self,
+        span: Span,
+        pattern: &'a str,
+        flags: RegExpFlags,
+    ) -> RegExpLiteral {
+        RegExpLiteral { span, value: EmptyObject, regex: RegExp { pattern: pattern.into(), flags } }
     }
 
     pub fn literal_string_expression(&self, literal: StringLiteral) -> Expression<'a> {

crates/oxc_parser/src/js/expression.rs

@@ -185,8 +185,8 @@ impl<'a> Parser<'a> {
             Kind::LParen => self.parse_parenthesized_expression(span),
             Kind::Slash | Kind::SlashEq => {
                 self.read_regex();
-                self.parse_literal_regexp()
-                    .map(|literal| self.ast.literal_regexp_expression(literal))
+                let literal = self.parse_literal_regexp();
+                Ok(self.ast.literal_regexp_expression(literal))
             }
             // JSXElement, JSXFragment
             Kind::LAngle if self.source_type.is_jsx() => self.parse_jsx_expression(),
@@ -315,20 +315,28 @@ impl<'a> Parser<'a> {
         Ok(self.ast.bigint_literal(self.end_span(span), value, base))
     }
 
-    pub(crate) fn parse_literal_regexp(&mut self) -> Result<RegExpLiteral> {
+    pub(crate) fn parse_literal_regexp(&mut self) -> RegExpLiteral {
         let span = self.start_span();
-        let r = match self.cur_kind() {
-            Kind::RegExp => self.cur_token().value.as_regex(),
-            _ => return Err(self.unexpected()),
-        };
-        let pattern = Atom::from(r.pattern);
-        let flags = r.flags;
+
+        // split out the flag part of `/regex/flag` by looking for `/` from the end
+        let regex_src = self.cur_src();
+        let mut flags = RegExpFlags::empty();
+
+        let mut split_index = None;
+        for (i, c) in regex_src.char_indices().rev() {
+            if let Ok(flag) = RegExpFlags::try_from(c) {
+                flags |= flag;
+            } else {
+                split_index.replace(i);
+                break;
+            }
+        }
+
+        // `/` are omitted from the pattern
+        let pattern = split_index.map_or(regex_src, |i| regex_src.get(1..i).unwrap_or(""));
+
         self.bump_any();
-        Ok(RegExpLiteral {
-            span: self.end_span(span),
-            value: EmptyObject {},
-            regex: RegExp { pattern, flags },
-        })
+        self.ast.reg_exp_literal(self.end_span(span), pattern, flags)
     }
 
     pub(crate) fn parse_literal_string(&mut self) -> Result<StringLiteral> {

crates/oxc_parser/src/lexer/mod.rs

@@ -24,7 +24,7 @@ use oxc_syntax::{
     },
     unicode_id_start::is_id_start_unicode,
 };
-pub use token::{RegExp, Token, TokenValue};
+pub use token::{Token, TokenValue};
 
 pub use self::{kind::Kind, number::parse_big_int};
 use self::{
@@ -819,7 +819,6 @@ impl<'a> Lexer<'a> {
 
     /// 12.9.5 Regular Expression Literals
     fn read_regex(&mut self) -> Kind {
-        let start = self.current.token.start + 1; // +1 to exclude `/`
         let mut in_escape = false;
         let mut in_character_class = false;
         loop {
@@ -848,40 +847,27 @@ impl<'a> Lexer<'a> {
             }
         }
 
-        let end = self.offset() - 1; // -1 to exclude `/`
-        let pattern = &self.source[start as usize..end as usize];
-
         let mut flags = RegExpFlags::empty();
 
         while let Some(ch @ ('$' | '_' | 'a'..='z' | 'A'..='Z' | '0'..='9')) = self.peek() {
             self.current.chars.next();
             if !ch.is_ascii_lowercase() {
                 self.error(diagnostics::RegExpFlag(ch, self.current_offset()));
-                continue;
+                return Kind::Undetermined;
             }
-            let flag = match ch {
-                'g' => RegExpFlags::G,
-                'i' => RegExpFlags::I,
-                'm' => RegExpFlags::M,
-                's' => RegExpFlags::S,
-                'u' => RegExpFlags::U,
-                'y' => RegExpFlags::Y,
-                'd' => RegExpFlags::D,
-                'v' => RegExpFlags::V,
-                _ => {
-                    self.error(diagnostics::RegExpFlag(ch, self.current_offset()));
-                    continue;
-                }
+            let flag = if let Ok(flag) = RegExpFlags::try_from(ch) {
+                flag
+            } else {
+                self.error(diagnostics::RegExpFlag(ch, self.current_offset()));
+                return Kind::Undetermined;
             };
             if flags.contains(flag) {
                 self.error(diagnostics::RegExpFlagTwice(ch, self.current_offset()));
-                continue;
+                return Kind::Undetermined;
             }
             flags |= flag;
         }
 
-        self.current.token.value = TokenValue::RegExp(RegExp { pattern, flags });
-
         Kind::RegExp
     }
 

crates/oxc_parser/src/lexer/token.rs

@@ -1,6 +1,5 @@
 //! Token
 
-use oxc_ast::ast::RegExpFlags;
 use oxc_span::Span;
 
 use super::kind::Kind;
@@ -29,7 +28,7 @@ pub struct Token<'a> {
 mod size_asserts {
     use oxc_index::assert_eq_size;
 
-    assert_eq_size!(super::Token, [u8; 48]);
+    assert_eq_size!(super::Token, [u8; 40]);
 }
 
 impl<'a> Token<'a> {
@@ -43,13 +42,6 @@ pub enum TokenValue<'a> {
     None,
     Number(f64),
     String(&'a str),
-    RegExp(RegExp<'a>),
-}
-
-#[derive(Debug, Copy, Clone)]
-pub struct RegExp<'a> {
-    pub pattern: &'a str,
-    pub flags: RegExpFlags,
 }
 
 impl<'a> Default for TokenValue<'a> {
@@ -66,13 +58,6 @@ impl<'a> TokenValue<'a> {
         }
     }
 
-    pub fn as_regex(&self) -> &RegExp<'a> {
-        match self {
-            Self::RegExp(regex) => regex,
-            _ => unreachable!("expected regex!"),
-        }
-    }
-
     pub fn get_string(&self) -> Option<&str> {
         match self {
             Self::String(s) => Some(s),

tasks/coverage/parser_babel.snap

@@ -1141,6 +1141,12 @@ Expect to Parse: "typescript/types/const-type-parameters-babel-7/input.ts"
    ╭─[core/uncategorised/380/input.js:1:1]
  1 │ var x = /
    ·         ──
+ 2 │ /
+   ╰────
+
+  × Unexpected token
+   ╭─[core/uncategorised/380/input.js:1:1]
+ 1 │ var x = /
  2 │ /
    ╰────
 
@@ -1523,6 +1529,12 @@ Expect to Parse: "typescript/types/const-type-parameters-babel-7/input.ts"
    ╭─[core/uncategorised/441/input.js:1:1]
  1 │ /a\
    · ────
+ 2 │ /
+   ╰────
+
+  × Unexpected token
+   ╭─[core/uncategorised/441/input.js:1:1]
+ 1 │ /a\
  2 │ /
    ╰────
 
@@ -7967,6 +7979,11 @@ Expect to Parse: "typescript/types/const-type-parameters-babel-7/input.ts"
  2 │ /
    ╰────
 
+  × Unexpected token
+   ╭─[esprima/invalid-syntax/migrated_0040/input.js:2:1]
+ 2 │ /
+   ╰────
+
   × Invalid Unicode escape sequence
    ╭─[esprima/invalid-syntax/migrated_0041/input.js:1:1]
  1 │ var x = /[a-z]/\ux
@@ -8141,6 +8158,11 @@ Expect to Parse: "typescript/types/const-type-parameters-babel-7/input.ts"
  2 │ /
    ╰────
 
+  × Unexpected token
+   ╭─[esprima/invalid-syntax/migrated_0062/input.js:2:1]
+ 2 │ /
+   ╰────
+
   × Unterminated string
    ╭─[esprima/invalid-syntax/migrated_0063/input.js:1:1]
  1 │ var x = "
@@ -8681,6 +8703,11 @@ Expect to Parse: "typescript/types/const-type-parameters-babel-7/input.ts"
  2 │ /
    ╰────
 
+  × Unexpected token
+   ╭─[esprima/invalid-syntax/migrated_0157/input.js:2:1]
+ 2 │ /
+   ╰────
+
   × Unexpected token
    ╭─[esprima/invalid-syntax/migrated_0158/input.js:2:1]
  2 │  

tasks/coverage/parser_test262.snap

@@ -18251,6 +18251,11 @@ Expect Syntax Error: "language/import/import-attributes/json-named-bindings.js"
     · ─
     ╰────
 
+  × Unexpected token
+    ╭─[language/line-terminators/invalid-regexp-cr.js:18:1]
+ 18 │ /
+    ╰────
+
   × Unterminated regular expression
     ╭─[language/line-terminators/invalid-regexp-lf.js:16:1]
  16 │ 
@@ -18259,20 +18264,35 @@ Expect Syntax Error: "language/import/import-attributes/json-named-bindings.js"
  18 │ /
     ╰────
 
+  × Unexpected token
+    ╭─[language/line-terminators/invalid-regexp-lf.js:18:1]
+ 18 │ /
+    ╰────
+
   × Unterminated regular expression
     ╭─[language/line-terminators/invalid-regexp-ls.js:16:1]
  16 │ 
  17 │ /
/
     · ──
     ╰────
 
+  × Unexpected token
+    ╭─[language/line-terminators/invalid-regexp-ls.js:17:1]
+ 17 │ /
/
+    ╰────
+
   × Unterminated regular expression
     ╭─[language/line-terminators/invalid-regexp-ps.js:16:1]
  16 │ 
  17 │ /
/
     · ──
     ╰────
 
+  × Unexpected token
+    ╭─[language/line-terminators/invalid-regexp-ps.js:17:1]
+ 17 │ /
/
+    ╰────
+
   × Unterminated string
     ╭─[language/line-terminators/invalid-string-cr.js:15:1]
  15 │ 
@@ -31537,6 +31557,11 @@ Expect Syntax Error: "language/import/import-attributes/json-named-bindings.js"
     ·                   ───────
     ╰────
 
+  × Expected `}` but found `EOF`
+    ╭─[language/statements/function/invalid-function-body-1.js:17:1]
+ 17 │ function __func(){/ ABC}
+    ╰────
+
   × Unexpected token
     ╭─[language/statements/function/invalid-function-body-2.js:16:1]
  16 │ 

tasks/coverage/parser_typescript.snap

@@ -17136,6 +17136,11 @@ Expect to Parse: "conformance/salsa/plainJSRedeclare3.ts"
    ·     ────────────
    ╰────
 
+  × Expected `)` but found `EOF`
+   ╭─[conformance/parser/ecmascript5/RegularExpressions/parserRegularExpressionDivideAmbiguity4.ts:1:1]
+ 1 │ foo(/notregexp);
+   ╰────
+
   × Expected a semicolon or an implicit semicolon after a statement, but found none
    ╭─[conformance/parser/ecmascript5/RegularExpressions/parserRegularExpressionDivideAmbiguity7.ts:1:1]
  1 │ (a/8