fix(parser): change unterminated regex error to be non-recoverable (#…

…5285)

closes #5257

crates/oxc_parser/src/cursor.rs

@@ -223,10 +223,10 @@ impl<'a> ParserImpl<'a> {
     }
 
     /// Tell lexer to read a regex
-    pub(crate) fn read_regex(&mut self) -> (u32, RegExpFlags) {
-        let (token, pattern_end, flags) = self.lexer.next_regex(self.cur_kind());
+    pub(crate) fn read_regex(&mut self) -> Result<(u32, RegExpFlags)> {
+        let (token, pattern_end, flags) = self.lexer.next_regex(self.cur_kind())?;
         self.token = token;
-        (pattern_end, flags)
+        Ok((pattern_end, flags))
     }
 
     /// Tell lexer to read a template substitution tail

crates/oxc_parser/src/js/expression.rs

@@ -189,10 +189,9 @@ impl<'a> ParserImpl<'a> {
                 }
             }
             Kind::LParen => self.parse_parenthesized_expression(span),
-            Kind::Slash | Kind::SlashEq => {
-                let literal = self.parse_literal_regexp();
-                Ok(self.ast.expression_from_reg_exp_literal(literal))
-            }
+            Kind::Slash | Kind::SlashEq => self
+                .parse_literal_regexp()
+                .map(|literal| self.ast.expression_from_reg_exp_literal(literal)),
             // JSXElement, JSXFragment
             Kind::LAngle if self.source_type.is_jsx() => self.parse_jsx_expression(),
             _ => self.parse_identifier_expression(),
@@ -336,11 +335,11 @@ impl<'a> ParserImpl<'a> {
         Ok(self.ast.big_int_literal(self.end_span(span), raw, base))
     }
 
-    pub(crate) fn parse_literal_regexp(&mut self) -> RegExpLiteral<'a> {
+    pub(crate) fn parse_literal_regexp(&mut self) -> Result<RegExpLiteral<'a>> {
         let span = self.start_span();
 
         // split out pattern
-        let (pattern_end, flags) = self.read_regex();
+        let (pattern_end, flags) = self.read_regex()?;
         let pattern_start = self.cur_token().start + 1; // +1 to exclude `/`
         let pattern = &self.source_text[pattern_start as usize..pattern_end as usize];
         self.bump_any();
@@ -350,11 +349,11 @@ impl<'a> ParserImpl<'a> {
             .parse_regular_expression
             .then(|| self.parse_regex_pattern(pattern_start, pattern, flags));
 
-        self.ast.reg_exp_literal(
+        Ok(self.ast.reg_exp_literal(
             self.end_span(span),
             EmptyObject,
             RegExp { pattern: self.ast.atom(pattern), flags },
-        )
+        ))
     }
 
     fn parse_regex_pattern(

crates/oxc_parser/src/lexer/regex.rs

@@ -2,6 +2,7 @@ use oxc_syntax::identifier::is_line_terminator;
 
 use super::{Kind, Lexer, RegExpFlags, Token};
 use crate::diagnostics;
+use oxc_diagnostics::Result;
 
 impl<'a> Lexer<'a> {
     /// Re-tokenize the current `/` or `/=` and return `RegExp`
@@ -10,34 +11,31 @@ impl<'a> Lexer<'a> {
     ///   where a `RegularExpressionLiteral` is permitted
     /// Which means the parser needs to re-tokenize on `PrimaryExpression`,
     /// `RegularExpressionLiteral` only appear on the right hand side of `PrimaryExpression`
-    pub(crate) fn next_regex(&mut self, kind: Kind) -> (Token, u32, RegExpFlags) {
+    pub(crate) fn next_regex(&mut self, kind: Kind) -> Result<(Token, u32, RegExpFlags)> {
         self.token.start = self.offset()
             - match kind {
                 Kind::Slash => 1,
                 Kind::SlashEq => 2,
                 _ => unreachable!(),
             };
-        let (pattern_end, flags) = self.read_regex();
+        let (pattern_end, flags) = self.read_regex()?;
         self.lookahead.clear();
         let token = self.finish_next(Kind::RegExp);
-        (token, pattern_end, flags)
+        Ok((token, pattern_end, flags))
     }
 
     /// 12.9.5 Regular Expression Literals
-    fn read_regex(&mut self) -> (u32, RegExpFlags) {
+    fn read_regex(&mut self) -> Result<(u32, RegExpFlags)> {
         let mut in_escape = false;
         let mut in_character_class = false;
         loop {
             match self.next_char() {
                 None => {
-                    self.error(diagnostics::unterminated_reg_exp(self.unterminated_range()));
-                    return (self.offset(), RegExpFlags::empty());
+                    return Err(diagnostics::unterminated_reg_exp(self.unterminated_range()));
+                    // return (self.offset(), RegExpFlags::empty());
                 }
                 Some(c) if is_line_terminator(c) => {
-                    self.error(diagnostics::unterminated_reg_exp(self.unterminated_range()));
-                    #[allow(clippy::cast_possible_truncation)]
-                    let pattern_end = self.offset() - c.len_utf8() as u32;
-                    return (pattern_end, RegExpFlags::empty());
+                    return Err(diagnostics::unterminated_reg_exp(self.unterminated_range()));
                 }
                 Some(c) => {
                     if in_escape {
@@ -79,6 +77,6 @@ impl<'a> Lexer<'a> {
             flags |= flag;
         }
 
-        (pattern_end, flags)
+        Ok((pattern_end, flags))
     }
 }

tasks/coverage/parser_babel.snap

@@ -1308,12 +1308,6 @@ Expect to Parse: tasks/coverage/babel/packages/babel-parser/test/fixtures/typesc
    ╭─[babel/packages/babel-parser/test/fixtures/core/uncategorised/380/input.js:1:9]
  1 │ var x = /
    ·         ──
- 2 │ /
-   ╰────
-
-  × Unexpected token
-   ╭─[babel/packages/babel-parser/test/fixtures/core/uncategorised/380/input.js:2:2]
- 1 │ var x = /
  2 │ /
    ╰────
 
@@ -1696,19 +1690,6 @@ Expect to Parse: tasks/coverage/babel/packages/babel-parser/test/fixtures/typesc
    ╭─[babel/packages/babel-parser/test/fixtures/core/uncategorised/441/input.js:1:1]
  1 │ /a\
    · ────
- 2 │ /
-   ╰────
-
-  × Invalid regular expression: Invalid escape
-   ╭─[babel/packages/babel-parser/test/fixtures/core/uncategorised/441/input.js:1:3]
- 1 │ /a\
-   ·   ─
- 2 │ /
-   ╰────
-
-  × Unexpected token
-   ╭─[babel/packages/babel-parser/test/fixtures/core/uncategorised/441/input.js:2:2]
- 1 │ /a\
  2 │ /
    ╰────
 
@@ -8218,11 +8199,6 @@ Expect to Parse: tasks/coverage/babel/packages/babel-parser/test/fixtures/typesc
  2 │ /
    ╰────
 
-  × Unexpected token
-   ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0040/input.js:3:1]
- 2 │ /
-   ╰────
-
   × Invalid Unicode escape sequence
    ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0041/input.js:1:17]
  1 │ var x = /[a-z]/\ux
@@ -8397,11 +8373,6 @@ Expect to Parse: tasks/coverage/babel/packages/babel-parser/test/fixtures/typesc
  2 │ /
    ╰────
 
-  × Unexpected token
-   ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0062/input.js:3:1]
- 2 │ /
-   ╰────
-
   × Unterminated string
    ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0063/input.js:1:9]
  1 │ var x = "
@@ -8942,18 +8913,6 @@ Expect to Parse: tasks/coverage/babel/packages/babel-parser/test/fixtures/typesc
  2 │ /
    ╰────
 
-  × Invalid regular expression: Invalid escape
-   ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0157/input.js:1:3]
- 1 │ /a\
-   ·   ─
- 2 │ /
-   ╰────
-
-  × Unexpected token
-   ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0157/input.js:3:1]
- 2 │ /
-   ╰────
-
   × Unexpected token
    ╭─[babel/packages/babel-parser/test/fixtures/esprima/invalid-syntax/migrated_0158/input.js:3:1]
  2 │