fix(html): correctly handle BOM in HTML-ish languages

Author: alissonlauffer

crates/biome_html_parser/src/lexer/mod.rs

@@ -82,15 +82,7 @@ impl<'src> HtmlLexer<'src> {
             _ if self.current_kind != T![<] && is_attribute_name_byte(current) => {
                 self.consume_identifier(current, false)
             }
-            _ => {
-                if self.position == 0
-                    && let Some((bom, bom_size)) = self.consume_potential_bom(UNICODE_BOM)
-                {
-                    self.unicode_bom_length = bom_size;
-                    return bom;
-                }
-                self.consume_unexpected_character()
-            }
+            _ => self.consume_unexpected_character(),
         }
     }
 
@@ -134,7 +126,15 @@ impl<'src> HtmlLexer<'src> {
                     self.consume_byte(HTML_LITERAL)
                 }
             }
-            _ => self.consume_html_text(current),
+            _ => {
+                if self.position == 0
+                    && let Some((bom, bom_size)) = self.consume_potential_bom(UNICODE_BOM)
+                {
+                    self.unicode_bom_length = bom_size;
+                    return bom;
+                }
+                self.consume_html_text(current)
+            }
         }
     }
 

crates/biome_html_parser/tests/html_specs/ok/bom.html

@@ -0,0 +1 @@
+<!doctype>

crates/biome_html_parser/tests/html_specs/ok/bom.html.snap

@@ -0,0 +1,53 @@
+---
+source: crates/biome_html_parser/tests/spec_test.rs
+assertion_line: 138
+expression: snapshot
+---
+## Input
+
+```html
+<!doctype>
+
+```
+
+
+## AST
+
+```
+HtmlRoot {
+    bom_token: UNICODE_BOM@0..3 "\u{feff}" [] [],
+    frontmatter: missing (optional),
+    directive: HtmlDirective {
+        l_angle_token: L_ANGLE@3..4 "<" [] [],
+        excl_token: BANG@4..5 "!" [] [],
+        doctype_token: DOCTYPE_KW@5..12 "doctype" [] [],
+        html_token: missing (optional),
+        quirk_token: missing (optional),
+        public_id_token: missing (optional),
+        system_id_token: missing (optional),
+        r_angle_token: R_ANGLE@12..13 ">" [] [],
+    },
+    html: HtmlElementList [],
+    eof_token: EOF@13..14 "" [Newline("\n")] [],
+}
+```
+
+## CST
+
+```
+0: [email protected]
+  0: [email protected] "\u{feff}" [] []
+  1: (empty)
+  2: [email protected]
+    0: [email protected] "<" [] []
+    1: [email protected] "!" [] []
+    2: [email protected] "doctype" [] []
+    3: (empty)
+    4: (empty)
+    5: (empty)
+    6: (empty)
+    7: [email protected] ">" [] []
+  3: [email protected]
+  4: [email protected] "" [Newline("\n")] []
+
+```