URL special case for line comments

src/parse/scanner.rs

@@ -96,11 +96,6 @@ impl<'s> Scanner<'s> {
         self.peek().map(f).unwrap_or(false)
     }
 
-    /// Checks whether the remaining source starts with the given string.
-    pub fn starts_with(&self, string: &str) -> bool {
-        self.rest().starts_with(string)
-    }
-
     /// The previous index in the source string.
     pub fn last_index(&self) -> usize {
         self.eaten()

src/parse/tokens.rs

@@ -72,9 +72,9 @@ impl<'s> Iterator for Tokens<'s> {
             // Whitespace.
             c if c.is_whitespace() => self.whitespace(c),
 
-            // Comments.
+            // Comments with special case for URLs.
-            '/' if self.s.eat_if('/') => self.line_comment(),
             '/' if self.s.eat_if('*') => self.block_comment(),
+            '/' if !self.maybe_in_url() && self.s.eat_if('/') => self.line_comment(),
             '*' if self.s.eat_if('/') => Token::Invalid(self.s.eaten_from(start)),
 
             // Other things.
@@ -216,7 +216,7 @@ impl<'s> Tokens<'s> {
                     self.s.eat_assert(c);
                     Token::Text(&self.s.eaten_from(start))
                 }
-                'u' if self.s.starts_with("u{") => {
+                'u' if self.s.rest().starts_with("u{") => {
                     self.s.eat_assert('u');
                     self.s.eat_assert('{');
                     Token::UnicodeEscape(UnicodeEscapeToken {
@@ -367,7 +367,7 @@ impl<'s> Tokens<'s> {
 
         // Read the fractional part if not already done.
         // Make sure not to confuse a range for the decimal separator.
-        if c != '.' && !self.s.starts_with("..") && self.s.eat_if('.') {
+        if c != '.' && !self.s.rest().starts_with("..") && self.s.eat_if('.') {
             self.s.eat_while(|c| c.is_ascii_digit());
         }
 
@@ -464,6 +464,10 @@ impl<'s> Tokens<'s> {
 
         Token::BlockComment(self.s.get(start .. end))
     }
+
+    fn maybe_in_url(&self) -> bool {
+        self.mode == TokenMode::Markup && self.s.eaten().ends_with(":/")
+    }
 }
 
 impl Debug for Tokens<'_> {
@@ -688,7 +692,6 @@ mod tests {
         t!(Markup[" /"]: r"\a"   => Text(r"\"), Text("a"));
         t!(Markup[" /"]: r"\u"   => Text(r"\"), Text("u"));
         t!(Markup[" /"]: r"\1"   => Text(r"\"), Text("1"));
-        t!(Markup[" /"]: r"\:"   => Text(r"\"), Text(":"));
         t!(Markup[" /"]: r#"\""# => Text(r"\"), Text("\""));
 
         // Test basic unicode escapes.

tests/typ/code/comment.typ

@@ -14,6 +14,11 @@ C/*
 #test(type(/*1*/ 1) //
 , "integer")
 
+---
+// Line comments have a special case for URLs.
+https://example.com \
+https:/* block comments don't ... */
+
 ---
 // End should not appear without start.
 // Error: 1:7-1:9 unexpected end of block comment

tools/support/typst.tmLanguage.json

@@ -6,12 +6,6 @@
   "repository": {
     "common": {
       "patterns": [
-        {
-          "name": "comment.line.double-slash.typst",
-          "begin": "//",
-          "end": "\n",
-          "beginCaptures": { "0": { "name": "punctuation.definition.comment.typst" } }
-        },
         {
           "name": "comment.block.typst",
           "begin": "/\\*",
@@ -38,6 +32,12 @@
     "markup": {
       "patterns": [
         { "include": "#common" },
+        {
+          "name": "comment.line.double-slash.typst",
+          "begin": "(?<!:)//",
+          "end": "\n",
+          "beginCaptures": { "0": { "name": "punctuation.definition.comment.typst" } }
+        },
         {
           "name": "constant.character.escape.content.typst",
           "match": "\\\\([\\\\/\\[\\]{}#*_=~`$-.]|u\\{[0-9a-zA-Z]*\\}?)"
@@ -191,6 +191,12 @@
     "code": {
       "patterns": [
         { "include": "#common" },
+        {
+          "name": "comment.line.double-slash.typst",
+          "begin": "//",
+          "end": "\n",
+          "beginCaptures": { "0": { "name": "punctuation.definition.comment.typst" } }
+        },
         {
           "name": "punctuation.separator.colon.typst",
           "match": ":"