");
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "div".to_string(),
+ attributes: vec![("id".to_string(), "main".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn tag_with_unquoted_attribute() {
+ let tokens = tokenize("
");
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "input".to_string(),
+ attributes: vec![("type".to_string(), "text".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn comment() {
+ let tokens = tokenize("");
+ assert_eq!(tokens, vec![Token::Comment(" comment ".to_string())]);
+ }
+
+ #[test]
+ fn empty_comment() {
+ let tokens = tokenize("");
+ assert_eq!(tokens, vec![Token::Comment("".to_string())]);
+ }
+
+ #[test]
+ fn doctype_html() {
+ let tokens = tokenize("");
+ assert_eq!(
+ tokens,
+ vec![Token::Doctype {
+ name: Some("html".to_string()),
+ public_id: None,
+ system_id: None,
+ force_quirks: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn doctype_case_insensitive() {
+ let tokens = tokenize("");
+ assert_eq!(
+ tokens,
+ vec![Token::Doctype {
+ name: Some("html".to_string()),
+ public_id: None,
+ system_id: None,
+ force_quirks: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn char_ref_named() {
+ let tokens = tokenize("&<>"");
+ assert_eq!(tokens, vec![Token::Character("&<>\"".to_string())]);
+ }
+
+ #[test]
+ fn char_ref_numeric_decimal() {
+ let tokens = tokenize("A");
+ assert_eq!(tokens, vec![Token::Character("A".to_string())]);
+ }
+
+ #[test]
+ fn char_ref_numeric_hex() {
+ let tokens = tokenize("A");
+ assert_eq!(tokens, vec![Token::Character("A".to_string())]);
+ }
+
+ #[test]
+ fn char_ref_numeric_hex_uppercase() {
+ let tokens = tokenize("A");
+ assert_eq!(tokens, vec![Token::Character("A".to_string())]);
+ }
+
+ #[test]
+ fn full_html_document() {
+ let tokens =
+ tokenize("
TestHello
");
+ assert_eq!(
+ tokens,
+ vec![
+ Token::StartTag {
+ name: "html".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::StartTag {
+ name: "head".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::StartTag {
+ name: "title".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::Character("Test".to_string()),
+ Token::EndTag {
+ name: "title".to_string(),
+ },
+ Token::EndTag {
+ name: "head".to_string(),
+ },
+ Token::StartTag {
+ name: "body".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::StartTag {
+ name: "p".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::Character("Hello".to_string()),
+ Token::EndTag {
+ name: "p".to_string(),
+ },
+ Token::EndTag {
+ name: "body".to_string(),
+ },
+ Token::EndTag {
+ name: "html".to_string(),
+ },
+ ]
+ );
+ }
+
+ #[test]
+ fn uppercase_tag_names_lowercased() {
+ let tokens = tokenize("
");
+ assert_eq!(
+ tokens,
+ vec![
+ Token::StartTag {
+ name: "div".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::EndTag {
+ name: "div".to_string(),
+ },
+ ]
+ );
+ }
+
+ #[test]
+ fn uppercase_attribute_names_lowercased() {
+ let tokens = tokenize(r#"
"#);
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "div".to_string(),
+ attributes: vec![("class".to_string(), "x".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn duplicate_attributes_first_wins() {
+ let tokens = tokenize(r#"
"#);
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "div".to_string(),
+ attributes: vec![("class".to_string(), "a".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn char_ref_in_attribute() {
+ let tokens = tokenize(r#"
"#);
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "a".to_string(),
+ attributes: vec![("href".to_string(), "?a=1&b=2".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn multiple_attributes() {
+ let tokens = tokenize(r#""#);
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "input".to_string(),
+ attributes: vec![
+ ("type".to_string(), "text".to_string()),
+ ("name".to_string(), "foo".to_string()),
+ ("value".to_string(), "bar".to_string()),
+ ],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn boolean_attribute() {
+ let tokens = tokenize("");
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "input".to_string(),
+ attributes: vec![("disabled".to_string(), "".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn mixed_content() {
+ let tokens = tokenize("Hello World");
+ assert_eq!(
+ tokens,
+ vec![
+ Token::Character("Hello ".to_string()),
+ Token::Comment(" comment ".to_string()),
+ Token::Character(" World".to_string()),
+ ]
+ );
+ }
+
+ #[test]
+ fn doctype_with_public_id() {
+ let tokens = tokenize(
+ r#""#,
+ );
+ assert_eq!(
+ tokens,
+ vec![Token::Doctype {
+ name: Some("html".to_string()),
+ public_id: Some("-//W3C//DTD XHTML 1.0 Strict//EN".to_string()),
+ system_id: Some("http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd".to_string()),
+ force_quirks: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn null_in_text() {
+ let tokens = tokenize("a\0b");
+ assert_eq!(tokens, vec![Token::Character("a\u{FFFD}b".to_string())]);
+ }
+
+ #[test]
+ fn windows_1252_numeric_refs() {
+ // € should map to Euro sign.
+ let tokens = tokenize("€");
+ assert_eq!(tokens, vec![Token::Character("\u{20AC}".to_string())]);
+ }
+
+ #[test]
+ fn attribute_with_empty_value() {
+ let tokens = tokenize(r#""#);
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "div".to_string(),
+ attributes: vec![("class".to_string(), "".to_string())],
+ self_closing: false,
+ }]
+ );
+ }
+
+ #[test]
+ fn adjacent_tags() {
+ let tokens = tokenize("
");
+ assert_eq!(
+ tokens,
+ vec![
+ Token::StartTag {
+ name: "b".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::EndTag {
+ name: "b".to_string(),
+ },
+ Token::StartTag {
+ name: "i".to_string(),
+ attributes: vec![],
+ self_closing: false,
+ },
+ Token::EndTag {
+ name: "i".to_string(),
+ },
+ ]
+ );
+ }
+
+ #[test]
+ fn newlines_in_text() {
+ let tokens = tokenize("line1\nline2\nline3");
+ assert_eq!(
+ tokens,
+ vec![Token::Character("line1\nline2\nline3".to_string())]
+ );
+ }
+
+ #[test]
+ fn self_closing_with_attribute() {
+ let tokens = tokenize(r#"
"#);
+ assert_eq!(
+ tokens,
+ vec![Token::StartTag {
+ name: "img".to_string(),
+ attributes: vec![("src".to_string(), "test.png".to_string())],
+ self_closing: true,
+ }]
+ );
+ }
+
+ #[test]
+ fn less_than_in_text_not_tag() {
+ // A bare '<' not followed by a letter should be emitted as text.
+ let tokens = tokenize("1 < 2");
+ assert_eq!(tokens, vec![Token::Character("1 < 2".to_string())]);
+ }
+
+ #[test]
+ fn ampersand_not_entity() {
+ let tokens = tokenize("a & b");
+ assert_eq!(tokens, vec![Token::Character("a & b".to_string())]);
+ }
+
+ #[test]
+ fn cdata_in_html_becomes_comment() {
+ let tokens = tokenize("");
+ // In HTML (non-foreign) context, CDATA is a parse error → bogus comment.
+ assert_eq!(tokens, vec![Token::Comment("[CDATA[hello]]".to_string())]);
+ }
+}