Fix parsing links with URL encoding (#1291)

Adds a HREF token to represent urls (and unit tests for the parser).

crates/parser/src/latex.rs

@@ -113,6 +113,7 @@ impl<'a> Parser<'a> {
             Token::Pipe | Token::Word | Token::Comma => self.text(context),
             Token::Eq => self.eat(),
             Token::Dollar => self.formula(),
+            Token::Href => self.eat(),
             Token::CommandName(name) => match name {
                 CommandName::Generic => self.generic_command(),
                 CommandName::BeginEnvironment if context.allow_environment => self.environment(),
@@ -1295,6 +1296,7 @@ impl<'a> Parser<'a> {
 
         self.builder.finish_node();
     }
+
 }
 
 pub fn parse_latex(text: &str, config: &SyntaxConfig) -> GreenNode {

crates/parser/src/latex/lexer.rs

@@ -40,6 +40,7 @@ impl<'a> Lexer<'a> {
             Token::Eq => SyntaxKind::EQUALITY_SIGN,
             Token::Pipe => SyntaxKind::WORD,
             Token::Word => SyntaxKind::WORD,
+            Token::Href => SyntaxKind::HREF,
             Token::Dollar => SyntaxKind::DOLLAR,
             Token::CommandName(_) => SyntaxKind::COMMAND_NAME,
         };

crates/parser/src/latex/lexer/types.rs

@@ -43,6 +43,9 @@ pub enum Token {
     #[regex(r"[^\s\\%\{\},\$\[\]\(\)=\|]+")]
     Word,
 
+    #[regex(r"[a-zA-Z]+:\/\/[^{}]+")]
+    Href,
+
     #[regex(r"\$\$?")]
     Dollar,