From a5f0dd8ee90a8abff3d1a79e7acfa3f8bf3c89c7 Mon Sep 17 00:00:00 2001
From: Ed Page <eopage@gmail.com>
Date: Mon, 2 Aug 2021 09:28:57 -0500
Subject: [PATCH] fix(token): Continue parsing on c-escape

---
 crates/typos/src/tokens.rs | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/crates/typos/src/tokens.rs b/crates/typos/src/tokens.rs
index 3c4afdf..c36db17 100644
--- a/crates/typos/src/tokens.rs
+++ b/crates/typos/src/tokens.rs
@@ -374,7 +374,7 @@ mod parser {
         <T as nom::InputTakeAtPosition>::Item: AsChar + Copy,
         <T as nom::InputIter>::Item: AsChar + Copy,
     {
-        preceded(char('\\'), take_while(is_xid_continue))(input)
+        preceded(take_while1(is_escape), take_while(is_xid_continue))(input)
     }
 
     fn printf<T>(input: T) -> IResult<T, T>
@@ -496,6 +496,12 @@ mod parser {
         unicode_xid::UnicodeXID::is_xid_continue(c)
     }
 
+    #[inline]
+    fn is_escape(i: impl AsChar + Copy) -> bool {
+        let c = i.as_char();
+        c == '\\'
+    }
+
     #[inline]
     fn is_digit_sep(chr: char) -> bool {
         // `_`: number literal separator in Rust and other languages
@@ -990,10 +996,10 @@ mod test {
     fn tokenize_c_escape() {
         let parser = TokenizerBuilder::new().build();
 
-        let input = "Hello \\Hello \\ World";
+        let input = "Hello \\Hello \\ \\\\ World";
         let expected: Vec<Identifier> = vec![
             Identifier::new_unchecked("Hello", Case::None, 0),
-            Identifier::new_unchecked("World", Case::None, 15),
+            Identifier::new_unchecked("World", Case::None, 18),
         ];
         let actual: Vec<_> = parser.parse_bytes(input.as_bytes()).collect();
         assert_eq!(expected, actual);