LibWeb: Record position information in CSS Tokens

This is a requirement to be able to use the Tokens for syntax
highlighting.

3 files changed, 30 insertions, 2 deletions

diff --git a/Userland/Libraries/LibWeb/CSS/Parser/Token.h b/Userland/Libraries/LibWeb/CSS/Parser/Token.h
index 9c1d945b68..3c906aa166 100644
--- a/Userland/Libraries/LibWeb/CSS/Parser/Token.h
+++ b/Userland/Libraries/LibWeb/CSS/Parser/Token.h
@@ -56,6 +56,12 @@ public:
         Number,
     };
 
+    struct Position {
+        size_t line { 0 };
+        size_t column { 0 };
+    };
+
+    Type type() const { return m_type; }
     bool is(Type type) const { return m_type == type; }
 
     StringView ident() const
@@ -136,6 +142,9 @@ public:
 
     String to_debug_string() const;
 
+    Position const& start_position() const { return m_start_position; }
+    Position const& end_position() const { return m_end_position; }
+
 private:
     Type m_type { Type::Invalid };
 
@@ -143,6 +152,9 @@ private:
     StringBuilder m_unit;
     HashType m_hash_type { HashType::Unrestricted };
     NumberType m_number_type { NumberType::Integer };
+
+    Position m_start_position;
+    Position m_end_position;
 };
 
 }
diff --git a/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp b/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp
index ad6a52bc29..9c4602606a 100644
--- a/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp
+++ b/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp
@@ -241,7 +241,10 @@ Vector<Token> Tokenizer::parse()
 {
     Vector<Token> tokens;
     for (;;) {
+        auto token_start = m_position;
         auto token = consume_a_token();
+        token.m_start_position = token_start;
+        token.m_end_position = m_position;
         tokens.append(token);
 
         if (token.is(Token::Type::EndOfFile)) {
@@ -256,8 +259,18 @@ u32 Tokenizer::next_code_point()
         return TOKENIZER_EOF;
     m_prev_utf8_iterator = m_utf8_iterator;
     ++m_utf8_iterator;
-    dbgln_if(CSS_TOKENIZER_DEBUG, "(Tokenizer) Next code_point: {:d}", *m_prev_utf8_iterator);
-    return *m_prev_utf8_iterator;
+    auto code_point = *m_prev_utf8_iterator;
+
+    m_prev_position = m_position;
+    if (is_newline(code_point)) {
+        m_position.line++;
+        m_position.column = 0;
+    } else {
+        m_position.column++;
+    }
+
+    dbgln_if(CSS_TOKENIZER_DEBUG, "(Tokenizer) Next code_point: {:d}", code_point);
+    return code_point;
 }
 
 u32 Tokenizer::peek_code_point(size_t offset) const
@@ -579,6 +592,7 @@ void Tokenizer::consume_as_much_whitespace_as_possible()
 void Tokenizer::reconsume_current_input_code_point()
 {
     m_utf8_iterator = m_prev_utf8_iterator;
+    m_position = m_prev_position;
 }
 
 // https://www.w3.org/TR/css-syntax-3/#consume-numeric-token
diff --git a/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.h b/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.h
index a5198bbd4f..d557bee3db 100644
--- a/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.h
+++ b/Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.h
@@ -103,5 +103,7 @@ private:
     Utf8View m_utf8_view;
     AK::Utf8CodePointIterator m_utf8_iterator;
     AK::Utf8CodePointIterator m_prev_utf8_iterator;
+    Token::Position m_position;
+    Token::Position m_prev_position;
 };
 }