Attachment 373517 Details for Bug 199520 – Patch

[patch] Patch

bug-199520-20190705111957.patch (text/plain), 66.54 KB, created by Robin Morisset on 2019-07-05 11:19:58 PDT

(hide)

Description:

Filename:

MIME Type:

Creator: Robin Morisset

Created: 2019-07-05 11:19:58 PDT

Size: 66.54 KB

patch

obsolete

>Subversion Revision: 247130
>diff --git a/Source/WebCore/ChangeLog b/Source/WebCore/ChangeLog
>index d062aa9b155dc63d09a7e74fd871101e518f5930..b660670783e6e6dc97ed49185caa2e371c50c4f4 100644
>--- a/Source/WebCore/ChangeLog
>+++ b/Source/WebCore/ChangeLog
>@@ -1,3 +1,98 @@
>+2019-07-05  Robin Morisset  <rmorisset@apple.com>
>+
>+        [WHLSL] Optimize the lexer
>+        https://bugs.webkit.org/show_bug.cgi?id=199520
>+
>+        Reviewed by NOBODY (OOPS!).
>+
>+        This patch includes a number of small optimizations to the lexer:
>+        - By creating two new types of Token (Invalid and EOF), we can avoid having to manipulate Optional<Token> everywhere
>+        - By combining the lexing of identifiers, keywords and operator names, we can avoid re-reading them several times
>+        - By storing mere offsets in the tokens instead of string views, we save space in the common case where the string view is unused (e.g. for keywords)
>+        - By storing mere offsets in the tokens instead of line numbers, we can simplify the whitespace and comment lexing; and we can easily recompute the line number on demand
>+        - By using templates we can avoid constructing and destructing vectors at runtime every time we call tryTypes, consumeTypes or peekTypes; as well as avoiding a call to find().
>+        - I also hoisted the length check out of the loop in Lexer::string()
>+        All of this combined brought the time spent in the parser when running LayoutTests/webgpu/whlsl-test-harness-test.html 5 times on my machine from 612ms to 462ms.
>+
>+        I also removed the 'anyCharacter' method since it was dead code.
>+        I also removed Lexer::position(), but it is directly equivalent now to peek().startOffset.
>+
>+        No new tests as there is no intended functional change.
>+
>+        * Modules/webgpu/WHLSL/WHLSLLexer.cpp:
>+        (WebCore::WHLSL::Lexer::Token::typeName):
>+        (WebCore::WHLSL::Lexer::consumeTokenFromStream):
>+        (WebCore::WHLSL::Lexer::lineNumberFromOffset):
>+        (WebCore::WHLSL::isWhitespace):
>+        (WebCore::WHLSL::isNewline):
>+        (WebCore::WHLSL::Lexer::skipWhitespaceAndComments):
>+        (WebCore::WHLSL::Lexer::digitStar const):
>+        (WebCore::WHLSL::Lexer::completeOperatorName const):
>+        * Modules/webgpu/WHLSL/WHLSLLexer.h:
>+        (WebCore::WHLSL::Lexer::Lexer):
>+        (WebCore::WHLSL::Lexer::Token::stringView):
>+        (WebCore::WHLSL::Lexer::consumeToken):
>+        (WebCore::WHLSL::Lexer::peek const):
>+        (WebCore::WHLSL::Lexer::peekFurther const):
>+        (WebCore::WHLSL::Lexer::state const):
>+        (WebCore::WHLSL::Lexer::setState):
>+        (WebCore::WHLSL::Lexer::isFullyConsumed const):
>+        (WebCore::WHLSL::Lexer::errorString):
>+        (WebCore::WHLSL::Lexer::string const):
>+        * Modules/webgpu/WHLSL/WHLSLNameResolver.h:
>+        * Modules/webgpu/WHLSL/WHLSLParser.cpp:
>+        (WebCore::WHLSL::Parser::parse):
>+        (WebCore::WHLSL::Parser::peek):
>+        (WebCore::WHLSL::Parser::peekFurther):
>+        (WebCore::WHLSL::Types::includes):
>+        (WebCore::WHLSL::Types::appendNameTo):
>+        (WebCore::WHLSL::Types<t>::includes):
>+        (WebCore::WHLSL::Types<t>::appendNameTo):
>+        (WebCore::WHLSL::Parser::peekTypes):
>+        (WebCore::WHLSL::Parser::tryType):
>+        (WebCore::WHLSL::Parser::tryTypes):
>+        (WebCore::WHLSL::Parser::consumeType):
>+        (WebCore::WHLSL::Parser::consumeTypes):
>+        (WebCore::WHLSL::Parser::consumeIntegralLiteral):
>+        (WebCore::WHLSL::Parser::parseConstantExpression):
>+        (WebCore::WHLSL::Parser::parseTypeArgument):
>+        (WebCore::WHLSL::Parser::parseTypeSuffixAbbreviated):
>+        (WebCore::WHLSL::Parser::parseTypeSuffixNonAbbreviated):
>+        (WebCore::WHLSL::Parser::parseType):
>+        (WebCore::WHLSL::Parser::parseTypeDefinition):
>+        (WebCore::WHLSL::Parser::parseBuiltInSemantic):
>+        (WebCore::WHLSL::Parser::parseResourceSemantic):
>+        (WebCore::WHLSL::Parser::parseQualifiers):
>+        (WebCore::WHLSL::Parser::parseStructureElement):
>+        (WebCore::WHLSL::Parser::parseStructureDefinition):
>+        (WebCore::WHLSL::Parser::parseEnumerationDefinition):
>+        (WebCore::WHLSL::Parser::parseEnumerationMember):
>+        (WebCore::WHLSL::Parser::parseNativeTypeDeclaration):
>+        (WebCore::WHLSL::Parser::parseParameter):
>+        (WebCore::WHLSL::Parser::parseComputeFunctionDeclaration):
>+        (WebCore::WHLSL::Parser::parseVertexOrFragmentFunctionDeclaration):
>+        (WebCore::WHLSL::Parser::parseRegularFunctionDeclaration):
>+        (WebCore::WHLSL::Parser::parseBlockBody):
>+        (WebCore::WHLSL::Parser::parseSwitchCase):
>+        (WebCore::WHLSL::Parser::parseVariableDeclaration):
>+        (WebCore::WHLSL::Parser::parseStatement):
>+        (WebCore::WHLSL::Parser::parseEffectfulAssignment):
>+        (WebCore::WHLSL::Parser::parseLimitedSuffixOperator):
>+        (WebCore::WHLSL::Parser::parseSuffixOperator):
>+        (WebCore::WHLSL::Parser::completeAssignment):
>+        (WebCore::WHLSL::Parser::parsePossibleTernaryConditional):
>+        (WebCore::WHLSL::Parser::completePossibleLogicalBinaryOperation):
>+        (WebCore::WHLSL::Parser::completePossibleRelationalBinaryOperation):
>+        (WebCore::WHLSL::Parser::completePossibleShift):
>+        (WebCore::WHLSL::Parser::completePossibleAdd):
>+        (WebCore::WHLSL::Parser::completePossibleMultiply):
>+        (WebCore::WHLSL::Parser::parsePossiblePrefix):
>+        (WebCore::WHLSL::Parser::parseCallExpression):
>+        (WebCore::WHLSL::Parser::parseTerm):
>+        * Modules/webgpu/WHLSL/WHLSLParser.h:
>+        * Modules/webgpu/WHLSL/WHLSLPreserveVariableLifetimes.cpp:
>+        (WebCore::WHLSL::anonymousToken):
>+
> 2019-07-03  Simon Fraser  <simon.fraser@apple.com>
> 
>         RELEASE_ASSERT in WebCore: WebCore::ScrollingStateTree::insertNode()
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.cpp
>index b836c5d6ebfa0b7fe75cf0a7d40caf2c822d0c53..b9ed5c0176349eaebaa7f90b46c8aaf021da7676 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.cpp
>@@ -247,6 +247,10 @@ const char* Lexer::Token::typeName(Type type)
>         return "!";
>     case Type::At:
>         return "@";
>+    case Type::EndOfFile:
>+        return "EOF";
>+    case Type::Invalid:
>+        return "LEXING_ERROR";
>     }
> }
> 
>@@ -372,28 +376,31 @@ auto Lexer::recognizeKeyword(unsigned end) -> Optional<Token::Type>
>     return WTF::nullopt;
> }
> 
>-auto Lexer::consumeTokenFromStream() -> Optional<Token>
>+auto Lexer::consumeTokenFromStream() -> Token
> {
>-    auto prepare = [&](unsigned newOffset, Token::Type type) -> Optional<Token> {
>+    auto prepare = [&](unsigned newOffset, Token::Type type) -> Token {
>         auto oldOffset = m_offset;
>         m_offset = newOffset;
>         skipWhitespaceAndComments();
>-        return {{ m_stringView.substring(oldOffset, newOffset - oldOffset), m_lineNumber, type }};
>+        return { oldOffset, newOffset, type };
>     };
> 
>+    if (auto newOffset = identifier(m_offset)) {
>+        if (auto result = recognizeKeyword(*newOffset)) {
>+            if (*result == Token::Type::Operator) {
>+                if (auto newerOffset = completeOperatorName(*newOffset))
>+                    return prepare(*newerOffset, Token::Type::OperatorName);
>+            }
>+            return prepare(*newOffset, *result);
>+        }
>+        return prepare(*newOffset, Token::Type::Identifier);
>+    }
>     if (auto newOffset = floatLiteral(m_offset))
>         return prepare(*newOffset, Token::Type::FloatLiteral);
>     if (auto newOffset = uintLiteral(m_offset))
>         return prepare(*newOffset, Token::Type::UintLiteral);
>     if (auto newOffset = intLiteral(m_offset))
>         return prepare(*newOffset, Token::Type::IntLiteral);
>-    if (auto newOffset = operatorName(m_offset))
>-        return prepare(*newOffset, Token::Type::OperatorName);
>-    if (auto newOffset = identifier(m_offset)) {
>-        if (auto result = recognizeKeyword(*newOffset))
>-            return prepare(*newOffset, *result);
>-        return prepare(*newOffset, Token::Type::Identifier);
>-    }
>     // Sorted by length, so longer matches are preferable to shorter matches.
>     if (auto newOffset = string(">>=", m_offset))
>         return prepare(*newOffset, Token::Type::RightShiftEquals);
>@@ -490,20 +497,28 @@ auto Lexer::consumeTokenFromStream() -> Optional<Token>
>     if (auto newOffset = character('@', m_offset))
>         return prepare(*newOffset, Token::Type::At);
> 
>-    return WTF::nullopt;
>+    if (isFullyConsumed())
>+        return prepare(m_offset, Token::Type::EndOfFile);
>+    return prepare(m_offset, Token::Type::Invalid);
> }
> 
>-void Lexer::skipWhitespaceAndComments()
>+unsigned Lexer::lineNumberFromOffset(unsigned targetOffset)
> {
>-    unsigned savedOffset;
>-    do {
>-        savedOffset = m_offset;
>-        skipWhitespace();
>-        skipLineComment();
>-        skipLongComment();
>-    } while (savedOffset != m_offset);
>+    // Counting from 1 to match most text editors.
>+    unsigned lineNumber = 1;
>+    for (unsigned offset = 0; offset < targetOffset; ++offset) {
>+        if (m_stringView[offset] == '\n')
>+            ++lineNumber;
>+    }
>+    return lineNumber;
> }
> 
>+// We can take advantage of two properties of Unicode:
>+// 1. The consitutent UTF-16 code units for all non-BMP code points are surrogates,
>+//        which means we'll never see a false match. If we see a BMP code unit, we
>+//        really have a BMP code point.
>+// 2. Everything we're looking for is in BMP
>+
> static inline bool isWhitespace(UChar codeUnit)
> {
>     switch (codeUnit) {
>@@ -528,53 +543,29 @@ static inline bool isNewline(UChar codeUnit)
>     }
> }
> 
>-// We can take advantage of two properties of Unicode:
>-// 1. The consitutent UTF-16 code units for all non-BMP code points are surrogates,
>-//        which means we'll never see a false match. If we see a BMP code unit, we
>-//        really have a BMP code point.
>-// 2. Everything we're looking for is in BMP
>-
>-void Lexer::skipWhitespace()
>-{
>-    for ( ; m_offset < m_stringView.length() && isWhitespace(m_stringView[m_offset]); ++m_offset) {
>-        if (m_stringView[m_offset] == '\r' && m_offset + 1 < m_stringView.length() && m_stringView[m_offset + 1] == '\n') {
>-            ++m_offset;
>-            ++m_lineNumber;
>-        } else if (isNewline(m_stringView[m_offset]))
>-            ++m_lineNumber;
>-    }
>-}
>-
>-void Lexer::skipLineComment()
>-{
>-    if (m_offset + 1 >= m_stringView.length() || m_stringView[m_offset] != '/' || m_stringView[m_offset + 1] != '/')
>-        return;
>-
>-    m_offset += 2;
>-    for ( ; m_offset < m_stringView.length() && !isNewline(m_stringView[m_offset]); ++m_offset) { }
>-}
>-
>-void Lexer::skipLongComment()
>+void Lexer::skipWhitespaceAndComments()
> {
>-    if (m_offset + 1 >= m_stringView.length() || m_stringView[m_offset] != '/' || m_stringView[m_offset + 1] != '*')
>-        return;
>-
>-    m_offset += 2;
>-    do {
>-        for ( ; m_offset < m_stringView.length() && m_stringView[m_offset] != '*'; ++m_offset) {
>-            if (m_stringView[m_offset] == '\r' && m_offset + 1 < m_stringView.length() && m_stringView[m_offset + 1] == '\n') {
>-                ++m_offset;
>-                ++m_lineNumber;
>-            } else if (isNewline(m_stringView[m_offset]))
>-                ++m_lineNumber;
>-        }
>-        if (m_offset < m_stringView.length())
>-            ++m_offset;
>-        if (m_offset < m_stringView.length() && m_stringView[m_offset] == '/') {
>+    while (m_offset < m_stringView.length()) {
>+        if (isWhitespace(m_stringView[m_offset]))
>             ++m_offset;
>+        else if (m_stringView[m_offset] == '/' && m_offset + 1 < m_stringView.length()) {
>+            if (m_stringView[m_offset + 1] == '/') {
>+                // Line comment
>+                m_offset += 2;
>+                // Note that in the case of \r\n this makes the comment end on the \r. It is fine, as the \n after that is simple whitespace.
>+                for ( ; m_offset < m_stringView.length() && !isNewline(m_stringView[m_offset]); ++m_offset) { }
>+            } else if (m_stringView[m_offset + 1] == '*') {
>+                // Long comment
>+                for ( ; m_offset < m_stringView.length() ; ++m_offset) {
>+                    if (m_stringView[m_offset] == '*' && m_offset + 1 < m_stringView.length() && m_stringView[m_offset + 1] == '/') {
>+                        m_offset += 2;
>+                        break;
>+                    }
>+                }
>+            }
>+        } else
>             break;
>-        }
>-    } while (m_offset < m_stringView.length());
>+    }
> }
> 
> // Regular expression are unnecessary; we shouldn't need to compile them.
>@@ -676,12 +667,9 @@ Optional<unsigned> Lexer::digit(unsigned offset) const
> 
> unsigned Lexer::digitStar(unsigned offset) const
> {
>-    while (true) {
>-        auto result = digit(offset);
>-        if (!result)
>-            return offset;
>+    while (auto result = digit(offset))
>         offset = *result;
>-    }
>+    return offset;
> }
> 
> Optional<unsigned> Lexer::character(char character, unsigned offset) const
>@@ -754,68 +742,66 @@ Optional<unsigned> Lexer::identifier(unsigned offset) const
>     return validIdentifier(offset);
> }
> 
>-Optional<unsigned> Lexer::operatorName(unsigned offset) const
>+Optional<unsigned> Lexer::completeOperatorName(unsigned offset) const
> {
>-    if (auto result = string("operator&.", offset))
>+    // Sorted by length, so longer matches are preferable to shorter matches.
>+    if (auto result = string("&[]", offset))
>+        return result;
>+    if (auto result = string("[]=", offset))
>+        return result;
>+    if (auto result = string(">>", offset))
>+        return result;
>+    if (auto result = string("<<", offset))
>+        return result;
>+    if (auto result = string("++", offset))
>+        return result;
>+    if (auto result = string("--", offset))
>+        return result;
>+    if (auto result = string("&&", offset))
>+        return result;
>+    if (auto result = string("||", offset))
>+        return result;
>+    if (auto result = string(">=", offset))
>+        return result;
>+    if (auto result = string("<=", offset))
>+        return result;
>+    if (auto result = string("==", offset))
>+        return result;
>+    if (auto result = string("[]", offset))
>+        return result;
>+    if (auto result = string("&.", offset))
>         return validIdentifier(*result);
>-    if (auto result = string("operator.", offset)) {
>-        if ((result = validIdentifier(*result))) {
>-            if (auto result2 = character('=', *result))
>-                return result2;
>-            return *result;
>+    if (auto result = character('+', offset))
>+        return result;
>+    if (auto result = character('-', offset))
>+        return result;
>+    if (auto result = character('*', offset))
>+        return result;
>+    if (auto result = character('/', offset))
>+        return result;
>+    if (auto result = character('%', offset))
>+        return result;
>+    if (auto result = character('<', offset))
>+        return result;
>+    if (auto result = character('>', offset))
>+        return result;
>+    if (auto result = character('!', offset))
>+        return result;
>+    if (auto result = character('~', offset))
>+        return result;
>+    if (auto result = character('&', offset))
>+        return result;
>+    if (auto result = character('^', offset))
>+        return result;
>+    if (auto result = character('|', offset))
>+        return result;
>+    if (auto result = character('.', offset)) {
>+        if (auto result2 = validIdentifier(*result)) {
>+            if (auto result3 = character('=', *result2))
>+                return result3;
>+            return *result2;
>         }
>     }
>-    if (auto result = string("operator", offset)) {
>-        // Sorted by length, so longer matches are preferable to shorter matches.
>-        if (auto result2 = string("&[]", *result))
>-            return result2;
>-        if (auto result2 = string("[]=", *result))
>-            return result2;
>-        if (auto result2 = string(">>", *result))
>-            return result2;
>-        if (auto result2 = string("<<", *result))
>-            return result2;
>-        if (auto result2 = string("++", *result))
>-            return result2;
>-        if (auto result2 = string("--", *result))
>-            return result2;
>-        if (auto result2 = string("&&", *result))
>-            return result2;
>-        if (auto result2 = string("||", *result))
>-            return result2;
>-        if (auto result2 = string(">=", *result))
>-            return result2;
>-        if (auto result2 = string("<=", *result))
>-            return result2;
>-        if (auto result2 = string("==", *result))
>-            return result2;
>-        if (auto result2 = string("[]", *result))
>-            return result2;
>-        if (auto result2 = character('+', *result))
>-            return result2;
>-        if (auto result2 = character('-', *result))
>-            return result2;
>-        if (auto result2 = character('*', *result))
>-            return result2;
>-        if (auto result2 = character('/', *result))
>-            return result2;
>-        if (auto result2 = character('%', *result))
>-            return result2;
>-        if (auto result2 = character('<', *result))
>-            return result2;
>-        if (auto result2 = character('>', *result))
>-            return result2;
>-        if (auto result2 = character('!', *result))
>-            return result2;
>-        if (auto result2 = character('~', *result))
>-            return result2;
>-        if (auto result2 = character('&', *result))
>-            return result2;
>-        if (auto result2 = character('^', *result))
>-            return result2;
>-        if (auto result2 = character('|', *result))
>-            return result2;
>-    }
>     return WTF::nullopt;
> }
> 
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.h b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.h
>index 4f213d19b5065f7478a4b97caeb6df8f47404e5f..e2e22aa8f12bd6157b6aaec5008e2da9e1e1e20a 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLLexer.h
>@@ -46,9 +46,7 @@ public:
>         : m_stringView(stringView)
>     {
>         skipWhitespaceAndComments();
>-        m_offsetRingBuffer[0] = m_offset;
>         m_ringBuffer[0] = consumeTokenFromStream();
>-        m_offsetRingBuffer[1] = m_offset;
>         m_ringBuffer[1] = consumeTokenFromStream();
>     }
> 
>@@ -59,15 +57,9 @@ public:
>     Lexer& operator=(Lexer&&) = default;
> 
>     struct Token {
>-        Token() = delete;
>-        Token(const Token&) = default;
>-        Token(Token&&) = default;
>-        Token& operator=(const Token&) = default;
>-        Token& operator=(Token&&) = default;
>-
>-        StringView stringView;
>-        unsigned lineNumber;
>-        enum class Type {
>+        unsigned startOffset;
>+        unsigned endOffset;
>+        enum class Type : uint8_t {
>             IntLiteral,
>             UintLiteral,
>             FloatLiteral,
>@@ -174,40 +166,42 @@ public:
>             Tilde,
>             ExclamationPoint,
>             At,
>-        } type;
>+            EndOfFile,
>+            Invalid
>+        } type {Type::Invalid};
> 
>         static const char* typeName(Type);
>+
>+        const StringView stringView(const Lexer& lexer)
>+        {
>+            return lexer.m_stringView.substring(startOffset, endOffset - startOffset);
>+        }
>     };
> 
>-    Optional<Token> consumeToken()
>+    Token consumeToken()
>     {
>         auto result = m_ringBuffer[m_ringBufferIndex];
>-        m_offsetRingBuffer[m_ringBufferIndex] = m_offset;
>         m_ringBuffer[m_ringBufferIndex] = consumeTokenFromStream();
>         m_ringBufferIndex = (m_ringBufferIndex + 1) % 2;
>         return result;
>     }
> 
>-    Optional<Token> peek() const
>+    Token peek() const
>     {
>         return m_ringBuffer[m_ringBufferIndex];
>     }
> 
>-    Optional<Token> peekFurther() const
>+    Token peekFurther() const
>     {
>         return m_ringBuffer[(m_ringBufferIndex + 1) % 2];
>     }
> 
>-    Optional<unsigned> position() const { return m_offsetRingBuffer[m_ringBufferIndex]; }
>-
>     // FIXME: We should not need this
>     // https://bugs.webkit.org/show_bug.cgi?id=198357
>     struct State {
>-        Optional<Token> ringBuffer[2];
>-        Optional<unsigned> m_offsetRingBuffer[2];
>+        Token ringBuffer[2];
>         unsigned ringBufferIndex;
>         unsigned offset;
>-        unsigned lineNumber;
>     };
> 
>     State state() const
>@@ -215,11 +209,8 @@ public:
>         State state;
>         state.ringBuffer[0] = m_ringBuffer[0];
>         state.ringBuffer[1] = m_ringBuffer[1];
>-        state.m_offsetRingBuffer[0] = m_offsetRingBuffer[0];
>-        state.m_offsetRingBuffer[1] = m_offsetRingBuffer[1];
>         state.ringBufferIndex = m_ringBufferIndex;
>         state.offset = m_offset;
>-        state.lineNumber = m_lineNumber;
>         return state;
>     }
> 
>@@ -227,11 +218,8 @@ public:
>     {
>         m_ringBuffer[0] = state.ringBuffer[0];
>         m_ringBuffer[1] = state.ringBuffer[1];
>-        m_offsetRingBuffer[0] = state.m_offsetRingBuffer[0];
>-        m_offsetRingBuffer[1] = state.m_offsetRingBuffer[1];
>         m_ringBufferIndex = state.ringBufferIndex;
>         m_offset = state.offset;
>-        m_lineNumber = state.lineNumber;
> 
>     }
> 
>@@ -239,30 +227,26 @@ public:
>     {
>         m_ringBuffer[0] = WTFMove(state.ringBuffer[0]);
>         m_ringBuffer[1] = WTFMove(state.ringBuffer[1]);
>-        m_offsetRingBuffer[0] = WTFMove(state.m_offsetRingBuffer[0]);
>-        m_offsetRingBuffer[1] = WTFMove(state.m_offsetRingBuffer[1]);
>         m_ringBufferIndex = WTFMove(state.ringBufferIndex);
>         m_offset = WTFMove(state.offset);
>-        m_lineNumber = WTFMove(state.lineNumber);
>     }
> 
>     bool isFullyConsumed() const
>     {
>-        return !peek();
>+        return peek().type == Token::Type::EndOfFile;
>     }
> 
>     String errorString(const Token& token, const String& message)
>     {
>-        return makeString("Parse error at line ", token.lineNumber, ": ", message);
>+        return makeString("Parse error at line ", lineNumberFromOffset(token.startOffset), ": ", message);
>     }
> 
> private:
>-    Optional<Token> consumeTokenFromStream();
>+    Token consumeTokenFromStream();
> 
>     void skipWhitespaceAndComments();
>-    void skipWhitespace();
>-    void skipLineComment();
>-    void skipLongComment();
>+
>+    unsigned lineNumberFromOffset(unsigned offset);
> 
>     Optional<Token::Type> recognizeKeyword(unsigned end);
> 
>@@ -277,43 +261,31 @@ private:
>     Optional<unsigned> digit(unsigned) const;
>     unsigned digitStar(unsigned) const;
>     Optional<unsigned> character(char, unsigned) const;
>-    template<unsigned length> Optional<unsigned> anyCharacter(const char (&string)[length], unsigned) const;
>     Optional<unsigned> coreFloatLiteralType1(unsigned) const;
>     Optional<unsigned> coreFloatLiteral(unsigned) const;
>     Optional<unsigned> floatLiteral(unsigned) const;
>     template<unsigned length> Optional<unsigned> string(const char (&string)[length], unsigned) const;
>     Optional<unsigned> validIdentifier(unsigned) const;
>     Optional<unsigned> identifier(unsigned) const;
>-    Optional<unsigned> operatorName(unsigned) const;
>+    Optional<unsigned> completeOperatorName(unsigned) const;
> 
>     StringView m_stringView;
>-    Optional<Token> m_ringBuffer[2];
>-    Optional<unsigned> m_offsetRingBuffer[2];
>+    Token m_ringBuffer[2];
>     unsigned m_ringBufferIndex { 0 };
>     unsigned m_offset { 0 };
>-    unsigned m_lineNumber { 0 };
> };
> 
> template<unsigned length> Optional<unsigned> Lexer::string(const char (&string)[length], unsigned offset) const
> {
>+    if (offset + length > m_stringView.length())
>+        return WTF::nullopt;
>     for (unsigned i = 0; i < length - 1; ++i) {
>-        if (offset + i >= m_stringView.length() || m_stringView[offset + i] != string[i])
>+        if (m_stringView[offset + i] != string[i])
>             return WTF::nullopt;
>     }
>     return offset + length - 1;
> }
> 
>-template<unsigned length> Optional<unsigned> Lexer::anyCharacter(const char (&string)[length], unsigned offset) const
>-{
>-    if (offset >= m_stringView.length())
>-        return WTF::nullopt;
>-    for (unsigned i = 0; i < length - 1; ++i) {
>-        if (m_stringView[offset] == string[i])
>-            return offset + 1;
>-    }
>-    return WTF::nullopt;
>-}
>-
> } // namespace WHLSL
> 
> } // namespace WebCore
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLNameResolver.h b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLNameResolver.h
>index 13f210bfc5bbd4a980a769fbd3efc817b97231e1..9237b37feeb6a18ec2d0b212ed0e984d09169c7b 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLNameResolver.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLNameResolver.h
>@@ -44,14 +44,13 @@ public:
> 
>     virtual ~NameResolver();
> 
>-    void visit(AST::FunctionDefinition&) override;
>-
>     void setCurrentFunctionDefinition(AST::FunctionDefinition* functionDefinition)
>     {
>         m_currentFunction = functionDefinition;
>     }
> 
> private:
>+    void visit(AST::FunctionDefinition&) override;
>     void visit(AST::NativeFunctionDeclaration&) override;
>     void visit(AST::TypeReference&) override;
>     void visit(AST::Block&) override;
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.cpp
>index 40f269df1e308505b59720b23e72b6752a64f08b..22b1a9b89306e009a0f4f8412564e4ead7fe15a5 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.cpp
>@@ -66,9 +66,9 @@ auto Parser::parse(Program& program, StringView stringView, Mode mode) -> Option
> 
>     while (!m_lexer.isFullyConsumed()) {
>         auto token = m_lexer.peek();
>-        if (!token)
>-            break;
>-        switch (token->type) {
>+        switch (token.type) {
>+        case Lexer::Token::Type::Invalid:
>+            return WTF::nullopt;
>         case Lexer::Token::Type::Semicolon:
>             m_lexer.consumeToken();
>             continue;
>@@ -135,75 +135,94 @@ auto Parser::fail(const String& message, TryToPeek tryToPeek) -> Unexpected<Erro
> 
> auto Parser::peek() -> Expected<Lexer::Token, Error>
> {
>-    if (auto token = m_lexer.peek()) {
>-        return *token;
>-    }
>+    auto token = m_lexer.peek();
>+    if (token.type != Lexer::Token::Type::Invalid && token.type != Lexer::Token::Type::EndOfFile)
>+        return { token };
>     return fail("Cannot consume token"_str, TryToPeek::No);
> }
> 
> auto Parser::peekFurther() -> Expected<Lexer::Token, Error>
> {
>-    if (auto token = m_lexer.peekFurther()) {
>-        return *token;
>-    }
>+    auto token = m_lexer.peekFurther();
>+    if (token.type != Lexer::Token::Type::Invalid && token.type != Lexer::Token::Type::EndOfFile)
>+        return { token };
>     return fail("Cannot consume two tokens"_str, TryToPeek::No);
> }
> 
>-bool Parser::peekTypes(const Vector<Lexer::Token::Type>& types)
>+template <Lexer::Token::Type t, Lexer::Token::Type... ts>
>+struct Types {
>+    static bool includes(Lexer::Token::Type type)
>+    {
>+        return t == type || Types<ts...>::includes(type);
>+    }
>+
>+    static void appendNameTo(StringBuilder& builder)
>+    {
>+        builder.append(Lexer::Token::typeName(t));
>+        builder.append(", ");
>+        Types<ts...>::appendNameTo(builder);
>+    }
>+};
>+template <Lexer::Token::Type t>
>+struct Types<t> {
>+    static bool includes(Lexer::Token::Type type)
>+    {
>+        return t == type;
>+    }
>+
>+    static void appendNameTo(StringBuilder& builder)
>+    {
>+        builder.append(Lexer::Token::typeName(t));
>+    }
>+};
>+
>+template <Lexer::Token::Type... types>
>+bool Parser::peekTypes()
> {
>-    if (auto token = m_lexer.peek())
>-        return std::find(types.begin(), types.end(), token->type) != types.end();
>-    return false;
>+    auto token = m_lexer.peek();
>+    return Types<types...>::includes(token.type);
> }
> 
> Optional<Lexer::Token> Parser::tryType(Lexer::Token::Type type)
> {
>-    if (auto token = m_lexer.peek()) {
>-        if (token->type == type)
>-            return m_lexer.consumeToken();
>-    }
>+    auto token = m_lexer.peek();
>+    if (token.type == type)
>+        return { m_lexer.consumeToken() };
>     return WTF::nullopt;
> }
> 
>-Optional<Lexer::Token> Parser::tryTypes(const Vector<Lexer::Token::Type>& types)
>+template <Lexer::Token::Type... types>
>+Optional<Lexer::Token> Parser::tryTypes()
> {
>-    if (auto token = m_lexer.peek()) {
>-        if (std::find(types.begin(), types.end(), token->type) != types.end())
>-            return m_lexer.consumeToken();
>-    }
>+    auto token = m_lexer.peek();
>+    if (Types<types...>::includes(token.type))
>+        return { m_lexer.consumeToken() };
>     return WTF::nullopt;
> }
> 
> auto Parser::consumeType(Lexer::Token::Type type) -> Expected<Lexer::Token, Error>
> {
>-    if (auto token = m_lexer.consumeToken()) {
>-        if (token->type == type)
>-            return *token;
>-        return fail(makeString("Unexpected token (expected ", Lexer::Token::typeName(type), " got ", Lexer::Token::typeName(token->type), ")"));
>-    }
>-    return fail(makeString("Cannot consume token (expected ", Lexer::Token::typeName(type), ")"));
>+    auto token = m_lexer.consumeToken();
>+    if (token.type == type)
>+        return { token };
>+    return fail(makeString("Unexpected token (expected ", Lexer::Token::typeName(type), " got ", Lexer::Token::typeName(token.type), ")"));
> }
> 
>-auto Parser::consumeTypes(const Vector<Lexer::Token::Type>& types) -> Expected<Lexer::Token, Error>
>+template <Lexer::Token::Type... types>
>+auto Parser::consumeTypes() -> Expected<Lexer::Token, Error>
> {
>     auto buildExpectedString = [&]() -> String {
>         StringBuilder builder;
>         builder.append("[");
>-        for (unsigned i = 0; i < types.size(); ++i) {
>-            if (i > 0)
>-                builder.append(", ");
>-            builder.append(Lexer::Token::typeName(types[i]));
>-        }
>+        Types<types...>::appendNameTo(builder);
>         builder.append("]");
>         return builder.toString();
>     };
> 
>-    if (auto token = m_lexer.consumeToken()) {
>-        if (std::find(types.begin(), types.end(), token->type) != types.end())
>-            return *token;
>-        return fail(makeString("Unexpected token (expected one of ", buildExpectedString(), " got ", Lexer::Token::typeName(token->type), ")"));
>-    }
>-    return fail(makeString("Cannot consume token (expected ", buildExpectedString(), ")"));
>+    auto token = m_lexer.consumeToken();
>+    if (Types<types...>::includes(token.type))
>+        return { token };
>+    return fail(makeString("Unexpected token (expected one of ", buildExpectedString(), " got ", Lexer::Token::typeName(token.type), ")"));
> }
> 
> static int digitValue(UChar character)
>@@ -279,20 +298,20 @@ static Expected<float, Parser::Error> floatLiteralToFloat(StringView text)
> 
> auto Parser::consumeIntegralLiteral() -> Expected<Variant<int, unsigned>, Error>
> {
>-    auto integralLiteralToken = consumeTypes({ Lexer::Token::Type::IntLiteral, Lexer::Token::Type::UintLiteral });
>+    auto integralLiteralToken = consumeTypes<Lexer::Token::Type::IntLiteral, Lexer::Token::Type::UintLiteral>();
>     if (!integralLiteralToken)
>         return Unexpected<Error>(integralLiteralToken.error());
> 
>     switch (integralLiteralToken->type) {
>     case Lexer::Token::Type::IntLiteral: {
>-        auto result = intLiteralToInt(integralLiteralToken->stringView);
>+        auto result = intLiteralToInt(integralLiteralToken->stringView(m_lexer));
>         if (result)
>             return {{ *result }};
>         return Unexpected<Error>(result.error());
>     }
>     default: {
>         ASSERT(integralLiteralToken->type == Lexer::Token::Type::UintLiteral);
>-        auto result = uintLiteralToUint(integralLiteralToken->stringView);
>+        auto result = uintLiteralToUint(integralLiteralToken->stringView(m_lexer));
>         if (result)
>             return {{ *result }};
>         return Unexpected<Error>(result.error());
>@@ -335,33 +354,32 @@ static Expected<unsigned, Parser::Error> recognizeSimpleUnsignedInteger(StringVi
> 
> auto Parser::parseConstantExpression() -> Expected<AST::ConstantExpression, Error>
> {
>-    auto type = consumeTypes({
>+    auto type = consumeTypes<
>         Lexer::Token::Type::IntLiteral,
>         Lexer::Token::Type::UintLiteral,
>         Lexer::Token::Type::FloatLiteral,
>         Lexer::Token::Type::Null,
>         Lexer::Token::Type::True,
>         Lexer::Token::Type::False,
>-        Lexer::Token::Type::Identifier,
>-    });
>+        Lexer::Token::Type::Identifier>();
>     if (!type)
>         return Unexpected<Error>(type.error());
> 
>     switch (type->type) {
>     case Lexer::Token::Type::IntLiteral: {
>-        auto value = intLiteralToInt(type->stringView);
>+        auto value = intLiteralToInt(type->stringView(m_lexer));
>         if (!value)
>             return Unexpected<Error>(value.error());
>         return {{ AST::IntegerLiteral(WTFMove(*type), *value) }};
>     }
>     case Lexer::Token::Type::UintLiteral: {
>-        auto value = uintLiteralToUint(type->stringView);
>+        auto value = uintLiteralToUint(type->stringView(m_lexer));
>         if (!value)
>             return Unexpected<Error>(value.error());
>         return {{ AST::UnsignedIntegerLiteral(WTFMove(*type), *value) }};
>     }
>     case Lexer::Token::Type::FloatLiteral: {
>-        auto value = floatLiteralToFloat(type->stringView);
>+        auto value = floatLiteralToFloat(type->stringView(m_lexer));
>         if (!value)
>             return Unexpected<Error>(value.error());
>         return {{ AST::FloatLiteral(WTFMove(*type), *value) }};
>@@ -376,7 +394,7 @@ auto Parser::parseConstantExpression() -> Expected<AST::ConstantExpression, Erro
>         ASSERT(type->type == Lexer::Token::Type::Identifier);
>         CONSUME_TYPE(origin, FullStop);
>         CONSUME_TYPE(next, Identifier);
>-        return { AST::EnumerationMemberLiteral(WTFMove(*origin), type->stringView.toString(), next->stringView.toString()) };
>+        return { AST::EnumerationMemberLiteral(WTFMove(*origin), type->stringView(m_lexer).toString(), next->stringView(m_lexer).toString()) };
>     }
>     }
> }
>@@ -390,7 +408,7 @@ auto Parser::parseTypeArgument() -> Expected<AST::TypeArgument, Error>
>         return AST::TypeArgument(WTFMove(*constantExpression));
>     }
>     CONSUME_TYPE(result, Identifier);
>-    return AST::TypeArgument(makeUniqueRef<AST::TypeReference>(Lexer::Token(*result), result->stringView.toString(), AST::TypeArguments()));
>+    return AST::TypeArgument(makeUniqueRef<AST::TypeReference>(Lexer::Token(*result), result->stringView(m_lexer).toString(), AST::TypeArguments()));
> }
> 
> auto Parser::parseTypeArguments() -> Expected<AST::TypeArguments, Error>
>@@ -422,7 +440,10 @@ auto Parser::parseTypeArguments() -> Expected<AST::TypeArguments, Error>
> 
> auto Parser::parseTypeSuffixAbbreviated() -> Expected<TypeSuffixAbbreviated, Error>
> {
>-    auto token = consumeTypes({ Lexer::Token::Type::Star, Lexer::Token::Type::SquareBracketPair, Lexer::Token::Type::LeftSquareBracket });
>+    auto token = consumeTypes<
>+        Lexer::Token::Type::Star,
>+        Lexer::Token::Type::SquareBracketPair,
>+        Lexer::Token::Type::LeftSquareBracket>();
>     if (!token)
>         return Unexpected<Error>(token.error());
>     if (token->type == Lexer::Token::Type::LeftSquareBracket) {
>@@ -437,7 +458,10 @@ auto Parser::parseTypeSuffixAbbreviated() -> Expected<TypeSuffixAbbreviated, Err
> 
> auto Parser::parseTypeSuffixNonAbbreviated() -> Expected<TypeSuffixNonAbbreviated, Error>
> {
>-    auto token = consumeTypes({ Lexer::Token::Type::Star, Lexer::Token::Type::SquareBracketPair, Lexer::Token::Type::LeftSquareBracket });
>+    auto token = consumeTypes<
>+        Lexer::Token::Type::Star,
>+        Lexer::Token::Type::SquareBracketPair,
>+        Lexer::Token::Type::LeftSquareBracket>();
>     if (!token)
>         return Unexpected<Error>(token.error());
>     if (token->type == Lexer::Token::Type::LeftSquareBracket) {
>@@ -447,7 +471,11 @@ auto Parser::parseTypeSuffixNonAbbreviated() -> Expected<TypeSuffixNonAbbreviate
>         CONSUME_TYPE(rightSquareBracket, RightSquareBracket);
>         return {{ *token, WTF::nullopt, *numElements }};
>     }
>-    auto addressSpaceToken = consumeTypes({ Lexer::Token::Type::Constant, Lexer::Token::Type::Device, Lexer::Token::Type::Threadgroup, Lexer::Token::Type::Thread});
>+    auto addressSpaceToken = consumeTypes<
>+        Lexer::Token::Type::Constant,
>+        Lexer::Token::Type::Device,
>+        Lexer::Token::Type::Threadgroup,
>+        Lexer::Token::Type::Thread>();
>     if (!addressSpaceToken)
>         return Unexpected<Error>(addressSpaceToken.error());
>     AST::AddressSpace addressSpace;
>@@ -471,7 +499,11 @@ auto Parser::parseTypeSuffixNonAbbreviated() -> Expected<TypeSuffixNonAbbreviate
> 
> auto Parser::parseType() -> Expected<UniqueRef<AST::UnnamedType>, Error>
> {
>-    auto addressSpaceToken = tryTypes({ Lexer::Token::Type::Constant, Lexer::Token::Type::Device, Lexer::Token::Type::Threadgroup, Lexer::Token::Type::Thread});
>+    auto addressSpaceToken = tryTypes<
>+        Lexer::Token::Type::Constant,
>+        Lexer::Token::Type::Device,
>+        Lexer::Token::Type::Threadgroup,
>+        Lexer::Token::Type::Thread>();
> 
>     CONSUME_TYPE(name, Identifier);
>     PARSE(typeArguments, TypeArguments);
>@@ -505,7 +537,7 @@ auto Parser::parseType() -> Expected<UniqueRef<AST::UnnamedType>, Error>
>             }
>         };
>         PARSE(firstTypeSuffixAbbreviated, TypeSuffixAbbreviated);
>-        UniqueRef<AST::UnnamedType> result = makeUniqueRef<AST::TypeReference>(WTFMove(*addressSpaceToken), name->stringView.toString(), WTFMove(*typeArguments));
>+        UniqueRef<AST::UnnamedType> result = makeUniqueRef<AST::TypeReference>(WTFMove(*addressSpaceToken), name->stringView(m_lexer).toString(), WTFMove(*typeArguments));
>         auto next = constructTypeFromSuffixAbbreviated(*firstTypeSuffixAbbreviated, WTFMove(result));
>         result = WTFMove(next);
>         while (true) {
>@@ -534,7 +566,7 @@ auto Parser::parseType() -> Expected<UniqueRef<AST::UnnamedType>, Error>
>             return { makeUniqueRef<AST::ArrayType>(Lexer::Token(typeSuffixNonAbbreviated.token), WTFMove(previous), *typeSuffixNonAbbreviated.numElements) };
>         }
>     };
>-    UniqueRef<AST::UnnamedType> result = makeUniqueRef<AST::TypeReference>(WTFMove(*name), name->stringView.toString(), WTFMove(*typeArguments));
>+    UniqueRef<AST::UnnamedType> result = makeUniqueRef<AST::TypeReference>(WTFMove(*name), name->stringView(m_lexer).toString(), WTFMove(*typeArguments));
>     while (true) {
>         PEEK(nextToken);
>         if (nextToken->type != Lexer::Token::Type::Star
>@@ -557,12 +589,12 @@ auto Parser::parseTypeDefinition() -> Expected<AST::TypeDefinition, Error>
>     CONSUME_TYPE(equals, EqualsSign);
>     PARSE(type, Type);
>     CONSUME_TYPE(semicolon, Semicolon);
>-    return AST::TypeDefinition(WTFMove(*origin), name->stringView.toString(), WTFMove(*type));
>+    return AST::TypeDefinition(WTFMove(*origin), name->stringView(m_lexer).toString(), WTFMove(*type));
> }
> 
> auto Parser::parseBuiltInSemantic() -> Expected<AST::BuiltInSemantic, Error>
> {
>-    auto origin = consumeTypes({
>+    auto origin = consumeTypes<
>         Lexer::Token::Type::SVInstanceID,
>         Lexer::Token::Type::SVVertexID,
>         Lexer::Token::Type::PSize,
>@@ -576,7 +608,7 @@ auto Parser::parseBuiltInSemantic() -> Expected<AST::BuiltInSemantic, Error>
>         Lexer::Token::Type::SVDispatchThreadID,
>         Lexer::Token::Type::SVGroupID,
>         Lexer::Token::Type::SVGroupIndex,
>-        Lexer::Token::Type::SVGroupThreadID});
>+        Lexer::Token::Type::SVGroupThreadID>();
>     if (!origin)
>         return Unexpected<Error>(origin.error());
> 
>@@ -623,14 +655,15 @@ auto Parser::parseResourceSemantic() -> Expected<AST::ResourceSemantic, Error>
>     CONSUME_TYPE(leftParenthesis, LeftParenthesis);
> 
>     CONSUME_TYPE(info, Identifier);
>-    if (info->stringView.length() < 2 || (info->stringView[0] != 'u'
>-        && info->stringView[0] != 't'
>-        && info->stringView[0] != 'b'
>-        && info->stringView[0] != 's'))
>-        return Unexpected<Error>(Error(makeString(info->stringView.substring(0, 1), " is not a known resource type ('u', 't', 'b', or 's')")));
>+    auto infoStringView = info->stringView(m_lexer);
>+    if (infoStringView.length() < 2 || (infoStringView[0] != 'u'
>+        && infoStringView[0] != 't'
>+        && infoStringView[0] != 'b'
>+        && infoStringView[0] != 's'))
>+        return Unexpected<Error>(Error(makeString(infoStringView.substring(0, 1), " is not a known resource type ('u', 't', 'b', or 's')")));
> 
>     AST::ResourceSemantic::Mode mode;
>-    switch (info->stringView[0]) {
>+    switch (infoStringView[0]) {
>     case 'u':
>         mode = AST::ResourceSemantic::Mode::UnorderedAccessView;
>         break;
>@@ -645,19 +678,20 @@ auto Parser::parseResourceSemantic() -> Expected<AST::ResourceSemantic, Error>
>         break;
>     }
> 
>-    auto index = recognizeSimpleUnsignedInteger(info->stringView.substring(1));
>+    auto index = recognizeSimpleUnsignedInteger(infoStringView.substring(1));
>     if (!index)
>         return Unexpected<Error>(index.error());
> 
>     unsigned space = 0;
>     if (tryType(Lexer::Token::Type::Comma)) {
>         CONSUME_TYPE(spaceToken, Identifier);
>+        auto spaceTokenStringView = spaceToken->stringView(m_lexer);
>         auto prefix = "space"_str;
>-        if (!spaceToken->stringView.startsWith(StringView(prefix)))
>-            return Unexpected<Error>(Error(makeString("Second argument to resource semantic ", spaceToken->stringView, " needs be of the form 'space0'")));
>-        if (spaceToken->stringView.length() <= prefix.length())
>-            return Unexpected<Error>(Error(makeString("Second argument to resource semantic ", spaceToken->stringView, " needs be of the form 'space0'")));
>-        auto spaceValue = recognizeSimpleUnsignedInteger(spaceToken->stringView.substring(prefix.length()));
>+        if (!spaceTokenStringView.startsWith(StringView(prefix)))
>+            return Unexpected<Error>(Error(makeString("Second argument to resource semantic ", spaceTokenStringView, " needs be of the form 'space0'")));
>+        if (spaceTokenStringView.length() <= prefix.length())
>+            return Unexpected<Error>(Error(makeString("Second argument to resource semantic ", spaceTokenStringView, " needs be of the form 'space0'")));
>+        auto spaceValue = recognizeSimpleUnsignedInteger(spaceTokenStringView.substring(prefix.length()));
>         if (!spaceValue)
>             return Unexpected<Error>(spaceValue.error());
>         space = *spaceValue;
>@@ -717,16 +751,17 @@ AST::Qualifiers Parser::parseQualifiers()
> {
>     AST::Qualifiers qualifiers;
>     while (auto next = tryType(Lexer::Token::Type::Qualifier)) {
>-        if ("nointerpolation" == next->stringView)
>+        auto nextStringView = next->stringView(m_lexer);
>+        if ("nointerpolation" == nextStringView)
>             qualifiers.append(AST::Qualifier::Nointerpolation);
>-        else if ("noperspective" == next->stringView)
>+        else if ("noperspective" == nextStringView)
>             qualifiers.append(AST::Qualifier::Noperspective);
>-        else if ("uniform" == next->stringView)
>+        else if ("uniform" == nextStringView)
>             qualifiers.append(AST::Qualifier::Uniform);
>-        else if ("centroid" == next->stringView)
>+        else if ("centroid" == nextStringView)
>             qualifiers.append(AST::Qualifier::Centroid);
>         else {
>-            ASSERT("sample" == next->stringView);
>+            ASSERT("sample" == nextStringView);
>             qualifiers.append(AST::Qualifier::Sample);
>         }
>     }
>@@ -744,7 +779,7 @@ auto Parser::parseStructureElement() -> Expected<AST::StructureElement, Error>
>     PARSE(semantic, Semantic);
>     CONSUME_TYPE(semicolon, Semicolon);
> 
>-    return AST::StructureElement(WTFMove(*origin), WTFMove(qualifiers), WTFMove(*type), name->stringView.toString(), WTFMove(*semantic));
>+    return AST::StructureElement(WTFMove(*origin), WTFMove(qualifiers), WTFMove(*type), name->stringView(m_lexer).toString(), WTFMove(*semantic));
> }
> 
> auto Parser::parseStructureDefinition() -> Expected<AST::StructureDefinition, Error>
>@@ -759,7 +794,7 @@ auto Parser::parseStructureDefinition() -> Expected<AST::StructureDefinition, Er
>         structureElements.append(WTFMove(*structureElement));
>     }
> 
>-    return AST::StructureDefinition(WTFMove(*origin), name->stringView.toString(), WTFMove(structureElements));
>+    return AST::StructureDefinition(WTFMove(*origin), name->stringView(m_lexer).toString(), WTFMove(structureElements));
> }
> 
> auto Parser::parseEnumerationDefinition() -> Expected<AST::EnumerationDefinition, Error>
>@@ -781,7 +816,7 @@ auto Parser::parseEnumerationDefinition() -> Expected<AST::EnumerationDefinition
> 
>     PARSE(firstEnumerationMember, EnumerationMember);
> 
>-    AST::EnumerationDefinition result(WTFMove(*origin), name->stringView.toString(), WTFMove(*type));
>+    AST::EnumerationDefinition result(WTFMove(*origin), name->stringView(m_lexer).toString(), WTFMove(*type));
>     auto success = result.add(WTFMove(*firstEnumerationMember));
>     if (!success)
>         return fail("Cannot add enumeration member"_str);
>@@ -801,7 +836,7 @@ auto Parser::parseEnumerationDefinition() -> Expected<AST::EnumerationDefinition
> auto Parser::parseEnumerationMember() -> Expected<AST::EnumerationMember, Error>
> {
>     CONSUME_TYPE(identifier, Identifier);
>-    auto name = identifier->stringView.toString();
>+    auto name = identifier->stringView(m_lexer).toString();
> 
>     if (tryType(Lexer::Token::Type::EqualsSign)) {
>         PARSE(constantExpression, ConstantExpression);
>@@ -818,7 +853,7 @@ auto Parser::parseNativeTypeDeclaration() -> Expected<AST::NativeTypeDeclaration
>     PARSE(typeArguments, TypeArguments);
>     CONSUME_TYPE(semicolon, Semicolon);
> 
>-    return AST::NativeTypeDeclaration(WTFMove(*origin), name->stringView.toString(), WTFMove(*typeArguments));
>+    return AST::NativeTypeDeclaration(WTFMove(*origin), name->stringView(m_lexer).toString(), WTFMove(*typeArguments));
> }
> 
> auto Parser::parseNumThreadsFunctionAttribute() -> Expected<AST::NumThreadsFunctionAttribute, Error>
>@@ -873,7 +908,7 @@ auto Parser::parseParameter() -> Expected<AST::VariableDeclaration, Error>
> 
>     String name;
>     if (auto token = tryType(Lexer::Token::Type::Identifier))
>-        name = token->stringView.toString();
>+        name = token->stringView(m_lexer).toString();
> 
>     PARSE(semantic, Semantic);
> 
>@@ -920,12 +955,12 @@ auto Parser::parseComputeFunctionDeclaration() -> Expected<AST::FunctionDeclarat
>     PARSE(parameters, Parameters);
>     PARSE(semantic, Semantic);
>     bool isOperator = false;
>-    return AST::FunctionDeclaration(WTFMove(*origin), WTFMove(*attributeBlock), AST::EntryPointType::Compute, WTFMove(*type), name->stringView.toString(), WTFMove(*parameters), WTFMove(*semantic), isOperator);
>+    return AST::FunctionDeclaration(WTFMove(*origin), WTFMove(*attributeBlock), AST::EntryPointType::Compute, WTFMove(*type), name->stringView(m_lexer).toString(), WTFMove(*parameters), WTFMove(*semantic), isOperator);
> }
> 
> auto Parser::parseVertexOrFragmentFunctionDeclaration() -> Expected<AST::FunctionDeclaration, Error>
> {
>-    auto entryPoint = consumeTypes({ Lexer::Token::Type::Vertex, Lexer::Token::Type::Fragment });
>+    auto entryPoint = consumeTypes<Lexer::Token::Type::Vertex, Lexer::Token::Type::Fragment>();
>     if (!entryPoint)
>         return Unexpected<Error>(entryPoint.error());
>     auto entryPointType = (entryPoint->type == Lexer::Token::Type::Vertex) ? AST::EntryPointType::Vertex : AST::EntryPointType::Fragment;
>@@ -936,7 +971,7 @@ auto Parser::parseVertexOrFragmentFunctionDeclaration() -> Expected<AST::Functio
>     PARSE(semantic, Semantic);
> 
>     bool isOperator = false;
>-    return AST::FunctionDeclaration(WTFMove(*entryPoint), { }, entryPointType, WTFMove(*type), name->stringView.toString(), WTFMove(*parameters), WTFMove(*semantic), isOperator);
>+    return AST::FunctionDeclaration(WTFMove(*entryPoint), { }, entryPointType, WTFMove(*type), name->stringView(m_lexer).toString(), WTFMove(*parameters), WTFMove(*semantic), isOperator);
> }
> 
> auto Parser::parseRegularFunctionDeclaration() -> Expected<AST::FunctionDeclaration, Error>
>@@ -945,7 +980,7 @@ auto Parser::parseRegularFunctionDeclaration() -> Expected<AST::FunctionDeclarat
> 
>     PARSE(type, Type);
> 
>-    auto name = consumeTypes({ Lexer::Token::Type::Identifier, Lexer::Token::Type::OperatorName });
>+    auto name = consumeTypes<Lexer::Token::Type::Identifier, Lexer::Token::Type::OperatorName>();
>     if (!name)
>         return Unexpected<Error>(name.error());
>     auto isOperator = name->type == Lexer::Token::Type::OperatorName;
>@@ -953,7 +988,7 @@ auto Parser::parseRegularFunctionDeclaration() -> Expected<AST::FunctionDeclarat
>     PARSE(parameters, Parameters);
>     PARSE(semantic, Semantic);
> 
>-    return AST::FunctionDeclaration(WTFMove(*origin), { }, WTF::nullopt, WTFMove(*type), name->stringView.toString(), WTFMove(*parameters), WTFMove(*semantic), isOperator);
>+    return AST::FunctionDeclaration(WTFMove(*origin), { }, WTF::nullopt, WTFMove(*type), name->stringView(m_lexer).toString(), WTFMove(*parameters), WTFMove(*semantic), isOperator);
> }
> 
> auto Parser::parseOperatorFunctionDeclaration() -> Expected<AST::FunctionDeclaration, Error>
>@@ -1003,7 +1038,7 @@ auto Parser::parseBlock() -> Expected<AST::Block, Error>
> auto Parser::parseBlockBody(Lexer::Token&& origin) -> Expected<AST::Block, Error>
> {
>     AST::Statements statements;
>-    while (!peekTypes({Lexer::Token::Type::RightCurlyBracket, Lexer::Token::Type::Case, Lexer::Token::Type::Default})) {
>+    while (!peekTypes<Lexer::Token::Type::RightCurlyBracket, Lexer::Token::Type::Case, Lexer::Token::Type::Default>()) {
>         PARSE(statement, Statement);
>         statements.append(WTFMove(*statement));
>     }
>@@ -1054,7 +1089,7 @@ auto Parser::parseSwitchStatement() -> Expected<AST::SwitchStatement, Error>
> 
> auto Parser::parseSwitchCase() -> Expected<AST::SwitchCase, Error>
> {
>-    auto origin = consumeTypes({ Lexer::Token::Type::Case, Lexer::Token::Type::Default });
>+    auto origin = consumeTypes<Lexer::Token::Type::Case, Lexer::Token::Type::Default>();
>     if (!origin)
>         return Unexpected<Error>(origin.error());
> 
>@@ -1161,7 +1196,7 @@ auto Parser::parseVariableDeclaration(UniqueRef<AST::UnnamedType>&& type) -> Exp
>         initializer = initializingExpression.value().moveToUniquePtr();
>     }
> 
>-    return AST::VariableDeclaration(WTFMove(*origin), WTFMove(qualifiers), { WTFMove(type) }, name->stringView.toString(), WTFMove(*semantic), WTFMove(initializer));
>+    return AST::VariableDeclaration(WTFMove(*origin), WTFMove(qualifiers), { WTFMove(type) }, name->stringView(m_lexer).toString(), WTFMove(*semantic), WTFMove(initializer));
> }
> 
> auto Parser::parseVariableDeclarations() -> Expected<AST::VariableDeclarationsStatement, Error>
>@@ -1218,36 +1253,36 @@ auto Parser::parseStatement() -> Expected<UniqueRef<AST::Statement>, Error>
>     case Lexer::Token::Type::Break: {
>         auto breakToken = m_lexer.consumeToken();
>         CONSUME_TYPE(semicolon, Semicolon);
>-        auto breakObject = AST::Break(WTFMove(*breakToken));
>+        auto breakObject = AST::Break(WTFMove(breakToken));
>         return { makeUniqueRef<AST::Break>(WTFMove(breakObject)) };
>     }
>     case Lexer::Token::Type::Continue: {
>         auto continueToken = m_lexer.consumeToken();
>         CONSUME_TYPE(semicolon, Semicolon);
>-        auto continueObject = AST::Continue(WTFMove(*continueToken));
>+        auto continueObject = AST::Continue(WTFMove(continueToken));
>         return { makeUniqueRef<AST::Continue>(WTFMove(continueObject)) };
>     }
>     case Lexer::Token::Type::Fallthrough: {
>         auto fallthroughToken = m_lexer.consumeToken();
>         CONSUME_TYPE(semicolon, Semicolon);
>-        auto fallthroughObject = AST::Fallthrough(WTFMove(*fallthroughToken));
>+        auto fallthroughObject = AST::Fallthrough(WTFMove(fallthroughToken));
>         return { makeUniqueRef<AST::Fallthrough>(WTFMove(fallthroughObject)) };
>     }
>     case Lexer::Token::Type::Trap: {
>         auto trapToken = m_lexer.consumeToken();
>         CONSUME_TYPE(semicolon, Semicolon);
>-        auto trapObject = AST::Trap(WTFMove(*trapToken));
>+        auto trapObject = AST::Trap(WTFMove(trapToken));
>         return { makeUniqueRef<AST::Trap>(WTFMove(trapObject)) };
>     }
>     case Lexer::Token::Type::Return: {
>         auto returnToken = m_lexer.consumeToken();
>         if (auto semicolon = tryType(Lexer::Token::Type::Semicolon)) {
>-            auto returnObject = AST::Return(WTFMove(*returnToken), WTF::nullopt);
>+            auto returnObject = AST::Return(WTFMove(returnToken), WTF::nullopt);
>             return { makeUniqueRef<AST::Return>(WTFMove(returnObject)) };
>         }
>         PARSE(expression, Expression);
>         CONSUME_TYPE(finalSemicolon, Semicolon);
>-        auto returnObject = AST::Return(WTFMove(*returnToken), { WTFMove(*expression) });
>+        auto returnObject = AST::Return(WTFMove(returnToken), { WTFMove(*expression) });
>         return { makeUniqueRef<AST::Return>(WTFMove(returnObject)) };
>     }
>     case Lexer::Token::Type::Constant:
>@@ -1321,7 +1356,7 @@ auto Parser::parseEffectfulAssignment() -> Expected<UniqueRef<AST::Expression>,
>     bool isEffectful = false;
>     PARSE(expression, PossiblePrefix, &isEffectful);
> 
>-    if (!isEffectful || peekTypes({
>+    if (!isEffectful || peekTypes<
>         Lexer::Token::Type::EqualsSign,
>         Lexer::Token::Type::PlusEquals,
>         Lexer::Token::Type::MinusEquals,
>@@ -1333,7 +1368,7 @@ auto Parser::parseEffectfulAssignment() -> Expected<UniqueRef<AST::Expression>,
>         Lexer::Token::Type::OrEquals,
>         Lexer::Token::Type::RightShiftEquals,
>         Lexer::Token::Type::LeftShiftEquals
>-    })) {
>+    >()) {
>         return completeAssignment(WTFMove(*origin), WTFMove(*expression));
>     }
> 
>@@ -1342,7 +1377,10 @@ auto Parser::parseEffectfulAssignment() -> Expected<UniqueRef<AST::Expression>,
> 
> auto Parser::parseLimitedSuffixOperator(UniqueRef<AST::Expression>&& previous) -> SuffixExpression
> {
>-    auto type = consumeTypes({ Lexer::Token::Type::FullStop, Lexer::Token::Type::Arrow, Lexer::Token::Type::LeftSquareBracket });
>+    auto type = consumeTypes<
>+        Lexer::Token::Type::FullStop,
>+        Lexer::Token::Type::Arrow,
>+        Lexer::Token::Type::LeftSquareBracket>();
>     if (!type)
>         return SuffixExpression(WTFMove(previous), false);
> 
>@@ -1351,13 +1389,13 @@ auto Parser::parseLimitedSuffixOperator(UniqueRef<AST::Expression>&& previous) -
>         auto identifier = consumeType(Lexer::Token::Type::Identifier);
>         if (!identifier)
>             return SuffixExpression(WTFMove(previous), false);
>-        return SuffixExpression(makeUniqueRef<AST::DotExpression>(WTFMove(*type), WTFMove(previous), identifier->stringView.toString()), true);
>+        return SuffixExpression(makeUniqueRef<AST::DotExpression>(WTFMove(*type), WTFMove(previous), identifier->stringView(m_lexer).toString()), true);
>     }
>     case Lexer::Token::Type::Arrow: {
>         auto identifier = consumeType(Lexer::Token::Type::Identifier);
>         if (!identifier)
>             return SuffixExpression(WTFMove(previous), false);
>-        return SuffixExpression(makeUniqueRef<AST::DotExpression>(Lexer::Token(*type), makeUniqueRef<AST::DereferenceExpression>(WTFMove(*type), WTFMove(previous)), identifier->stringView.toString()), true);
>+        return SuffixExpression(makeUniqueRef<AST::DotExpression>(Lexer::Token(*type), makeUniqueRef<AST::DereferenceExpression>(WTFMove(*type), WTFMove(previous)), identifier->stringView(m_lexer).toString()), true);
>     }
>     default: {
>         ASSERT(type->type == Lexer::Token::Type::LeftSquareBracket);
>@@ -1373,7 +1411,12 @@ auto Parser::parseLimitedSuffixOperator(UniqueRef<AST::Expression>&& previous) -
> 
> auto Parser::parseSuffixOperator(UniqueRef<AST::Expression>&& previous) -> SuffixExpression
> {
>-    auto suffix = consumeTypes({ Lexer::Token::Type::FullStop, Lexer::Token::Type::Arrow, Lexer::Token::Type::LeftSquareBracket, Lexer::Token::Type::PlusPlus, Lexer::Token::Type::MinusMinus });
>+    auto suffix = consumeTypes<
>+        Lexer::Token::Type::FullStop,
>+        Lexer::Token::Type::Arrow,
>+        Lexer::Token::Type::LeftSquareBracket,
>+        Lexer::Token::Type::PlusPlus,
>+        Lexer::Token::Type::MinusMinus>();
>     if (!suffix)
>         return SuffixExpression(WTFMove(previous), false);
> 
>@@ -1382,13 +1425,13 @@ auto Parser::parseSuffixOperator(UniqueRef<AST::Expression>&& previous) -> Suffi
>         auto identifier = consumeType(Lexer::Token::Type::Identifier);
>         if (!identifier)
>             return SuffixExpression(WTFMove(previous), false);
>-        return SuffixExpression(makeUniqueRef<AST::DotExpression>(WTFMove(*suffix), WTFMove(previous), identifier->stringView.toString()), true);
>+        return SuffixExpression(makeUniqueRef<AST::DotExpression>(WTFMove(*suffix), WTFMove(previous), identifier->stringView(m_lexer).toString()), true);
>     }
>     case Lexer::Token::Type::Arrow: {
>         auto identifier = consumeType(Lexer::Token::Type::Identifier);
>         if (!identifier)
>             return SuffixExpression(WTFMove(previous), false);
>-        return SuffixExpression(makeUniqueRef<AST::DotExpression>(Lexer::Token(*suffix), makeUniqueRef<AST::DereferenceExpression>(WTFMove(*suffix), WTFMove(previous)), identifier->stringView.toString()), true);
>+        return SuffixExpression(makeUniqueRef<AST::DotExpression>(Lexer::Token(*suffix), makeUniqueRef<AST::DereferenceExpression>(WTFMove(*suffix), WTFMove(previous)), identifier->stringView(m_lexer).toString()), true);
>     }
>     case Lexer::Token::Type::LeftSquareBracket: {
>         auto expression = parseExpression();
>@@ -1451,7 +1494,7 @@ auto Parser::completeTernaryConditional(Lexer::Token&& origin, UniqueRef<AST::Ex
> 
> auto Parser::completeAssignment(Lexer::Token&& origin, UniqueRef<AST::Expression>&& left) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    auto assignmentOperator = consumeTypes({
>+    auto assignmentOperator = consumeTypes<
>         Lexer::Token::Type::EqualsSign,
>         Lexer::Token::Type::PlusEquals,
>         Lexer::Token::Type::MinusEquals,
>@@ -1462,8 +1505,7 @@ auto Parser::completeAssignment(Lexer::Token&& origin, UniqueRef<AST::Expression
>         Lexer::Token::Type::AndEquals,
>         Lexer::Token::Type::OrEquals,
>         Lexer::Token::Type::RightShiftEquals,
>-        Lexer::Token::Type::LeftShiftEquals
>-    });
>+        Lexer::Token::Type::LeftShiftEquals>();
>     if (!assignmentOperator)
>         return Unexpected<Error>(assignmentOperator.error());
> 
>@@ -1522,7 +1564,7 @@ auto Parser::parsePossibleTernaryConditional() -> Expected<UniqueRef<AST::Expres
> 
>     PARSE(expression, PossiblePrefix);
> 
>-    if (peekTypes({Lexer::Token::Type::EqualsSign,
>+    if (peekTypes<Lexer::Token::Type::EqualsSign,
>         Lexer::Token::Type::PlusEquals,
>         Lexer::Token::Type::MinusEquals,
>         Lexer::Token::Type::TimesEquals,
>@@ -1532,7 +1574,7 @@ auto Parser::parsePossibleTernaryConditional() -> Expected<UniqueRef<AST::Expres
>         Lexer::Token::Type::AndEquals,
>         Lexer::Token::Type::OrEquals,
>         Lexer::Token::Type::RightShiftEquals,
>-        Lexer::Token::Type::LeftShiftEquals})) {
>+        Lexer::Token::Type::LeftShiftEquals>()) {
>         return completeAssignment(WTFMove(*origin), WTFMove(*expression));
>     }
> 
>@@ -1556,13 +1598,13 @@ auto Parser::parsePossibleLogicalBinaryOperation() -> Expected<UniqueRef<AST::Ex
> 
> auto Parser::completePossibleLogicalBinaryOperation(UniqueRef<AST::Expression>&& previous) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    while (auto logicalBinaryOperation = tryTypes({
>+    while (auto logicalBinaryOperation = tryTypes<
>         Lexer::Token::Type::OrOr,
>         Lexer::Token::Type::AndAnd,
>         Lexer::Token::Type::Or,
>         Lexer::Token::Type::Xor,
>         Lexer::Token::Type::And
>-        })) {
>+        >()) {
>         PARSE(next, PossibleRelationalBinaryOperation);
> 
>         switch (logicalBinaryOperation->type) {
>@@ -1608,14 +1650,14 @@ auto Parser::parsePossibleRelationalBinaryOperation() -> Expected<UniqueRef<AST:
> 
> auto Parser::completePossibleRelationalBinaryOperation(UniqueRef<AST::Expression>&& previous) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    while (auto relationalBinaryOperation = tryTypes({
>+    while (auto relationalBinaryOperation = tryTypes<
>         Lexer::Token::Type::LessThanSign,
>         Lexer::Token::Type::GreaterThanSign,
>         Lexer::Token::Type::LessThanOrEqualTo,
>         Lexer::Token::Type::GreaterThanOrEqualTo,
>         Lexer::Token::Type::EqualComparison,
>         Lexer::Token::Type::NotEqual
>-        })) {
>+        >()) {
>         PARSE(next, PossibleShift);
> 
>         Vector<UniqueRef<AST::Expression>> callArguments;
>@@ -1663,10 +1705,10 @@ auto Parser::parsePossibleShift() -> Expected<UniqueRef<AST::Expression>, Error>
> 
> auto Parser::completePossibleShift(UniqueRef<AST::Expression>&& previous) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    while (auto shift = tryTypes({
>+    while (auto shift = tryTypes<
>         Lexer::Token::Type::LeftShift,
>         Lexer::Token::Type::RightShift
>-        })) {
>+        >()) {
>         PARSE(next, PossibleAdd);
> 
>         Vector<UniqueRef<AST::Expression>> callArguments;
>@@ -1697,10 +1739,10 @@ auto Parser::parsePossibleAdd() -> Expected<UniqueRef<AST::Expression>, Error>
> 
> auto Parser::completePossibleAdd(UniqueRef<AST::Expression>&& previous) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    while (auto add = tryTypes({
>+    while (auto add = tryTypes<
>         Lexer::Token::Type::Plus,
>         Lexer::Token::Type::Minus
>-        })) {
>+        >()) {
>         PARSE(next, PossibleMultiply);
> 
>         Vector<UniqueRef<AST::Expression>> callArguments;
>@@ -1731,11 +1773,11 @@ auto Parser::parsePossibleMultiply() -> Expected<UniqueRef<AST::Expression>, Err
> 
> auto Parser::completePossibleMultiply(UniqueRef<AST::Expression>&& previous) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    while (auto multiply = tryTypes({
>+    while (auto multiply = tryTypes<
>         Lexer::Token::Type::Star,
>         Lexer::Token::Type::Divide,
>         Lexer::Token::Type::Mod
>-        })) {
>+        >()) {
>         PARSE(next, PossiblePrefix);
> 
>         Vector<UniqueRef<AST::Expression>> callArguments;
>@@ -1764,7 +1806,7 @@ auto Parser::completePossibleMultiply(UniqueRef<AST::Expression>&& previous) ->
> 
> auto Parser::parsePossiblePrefix(bool *isEffectful) -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    if (auto prefix = tryTypes({
>+    if (auto prefix = tryTypes<
>         Lexer::Token::Type::PlusPlus,
>         Lexer::Token::Type::MinusMinus,
>         Lexer::Token::Type::Plus,
>@@ -1774,7 +1816,7 @@ auto Parser::parsePossiblePrefix(bool *isEffectful) -> Expected<UniqueRef<AST::E
>         Lexer::Token::Type::And,
>         Lexer::Token::Type::At,
>         Lexer::Token::Type::Star
>-    })) {
>+    >()) {
>         PARSE(next, PossiblePrefix);
> 
>         switch (prefix->type) {
>@@ -1877,7 +1919,7 @@ auto Parser::parsePossibleSuffix(bool *isEffectful) -> Expected<UniqueRef<AST::E
> auto Parser::parseCallExpression() -> Expected<UniqueRef<AST::Expression>, Error>
> {
>     CONSUME_TYPE(name, Identifier);
>-    auto callName = name->stringView.toString();
>+    auto callName = name->stringView(m_lexer).toString();
> 
>     CONSUME_TYPE(leftParenthesis, LeftParenthesis);
> 
>@@ -1899,7 +1941,7 @@ auto Parser::parseCallExpression() -> Expected<UniqueRef<AST::Expression>, Error
> 
> auto Parser::parseTerm() -> Expected<UniqueRef<AST::Expression>, Error>
> {
>-    auto type = consumeTypes({
>+    auto type = consumeTypes<
>         Lexer::Token::Type::IntLiteral,
>         Lexer::Token::Type::UintLiteral,
>         Lexer::Token::Type::FloatLiteral,
>@@ -1907,26 +1949,25 @@ auto Parser::parseTerm() -> Expected<UniqueRef<AST::Expression>, Error>
>         Lexer::Token::Type::True,
>         Lexer::Token::Type::False,
>         Lexer::Token::Type::Identifier,
>-        Lexer::Token::Type::LeftParenthesis
>-    });
>+        Lexer::Token::Type::LeftParenthesis>();
>     if (!type)
>         return Unexpected<Error>(type.error());
> 
>     switch (type->type) {
>     case Lexer::Token::Type::IntLiteral: {
>-        auto value = intLiteralToInt(type->stringView);
>+        auto value = intLiteralToInt(type->stringView(m_lexer));
>         if (!value)
>             return Unexpected<Error>(value.error());
>         return { makeUniqueRef<AST::IntegerLiteral>(WTFMove(*type), *value) };
>     }
>     case Lexer::Token::Type::UintLiteral: {
>-        auto value = uintLiteralToUint(type->stringView);
>+        auto value = uintLiteralToUint(type->stringView(m_lexer));
>         if (!value)
>             return Unexpected<Error>(value.error());
>         return { makeUniqueRef<AST::UnsignedIntegerLiteral>(WTFMove(*type), *value) };
>     }
>     case Lexer::Token::Type::FloatLiteral: {
>-        auto value = floatLiteralToFloat(type->stringView);
>+        auto value = floatLiteralToFloat(type->stringView(m_lexer));
>         if (!value)
>             return Unexpected<Error>(value.error());
>         return { makeUniqueRef<AST::FloatLiteral>(WTFMove(*type), *value) };
>@@ -1938,7 +1979,7 @@ auto Parser::parseTerm() -> Expected<UniqueRef<AST::Expression>, Error>
>     case Lexer::Token::Type::False:
>         return { makeUniqueRef<AST::BooleanLiteral>(WTFMove(*type), false) };
>     case Lexer::Token::Type::Identifier: {
>-        auto name = type->stringView.toString();
>+        auto name = type->stringView(m_lexer).toString();
>         return { makeUniqueRef<AST::VariableReference>(WTFMove(*type), WTFMove(name)) };
>     }
>     default: {
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.h b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.h
>index b4482a4469edc15e533190630741f768791db747..db9961d87116337fbb0a1ef87d5e268ba3438a59 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLParser.h
>@@ -143,11 +143,14 @@ private:
>     Unexpected<Error> fail(const String& message, TryToPeek = TryToPeek::Yes);
>     Expected<Lexer::Token, Error> peek();
>     Expected<Lexer::Token, Error> peekFurther();
>-    bool peekTypes(const Vector<Lexer::Token::Type>&);
>+    template <Lexer::Token::Type... types>
>+    bool peekTypes();
>     Optional<Lexer::Token> tryType(Lexer::Token::Type);
>-    Optional<Lexer::Token> tryTypes(const Vector<Lexer::Token::Type>&);
>+    template <Lexer::Token::Type... types>
>+    Optional<Lexer::Token> tryTypes();
>     Expected<Lexer::Token, Error> consumeType(Lexer::Token::Type);
>-    Expected<Lexer::Token, Error> consumeTypes(const Vector<Lexer::Token::Type>&);
>+    template <Lexer::Token::Type... types>
>+    Expected<Lexer::Token, Error> consumeTypes();
> 
>     Expected<Variant<int, unsigned>, Error> consumeIntegralLiteral();
>     Expected<unsigned, Error> consumeNonNegativeIntegralLiteral();
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPreserveVariableLifetimes.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPreserveVariableLifetimes.cpp
>index 3c2eb10eab11a962830f8cb3a2b00bf26950ed8b..9d81efd7ab0dd5da214d350ff055136a3cf6b7f7 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPreserveVariableLifetimes.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPreserveVariableLifetimes.cpp
>@@ -93,7 +93,7 @@ private:
> 
> static ALWAYS_INLINE Lexer::Token anonymousToken(Lexer::Token::Type type)
> {
>-    return Lexer::Token { { }, 0, type };
>+    return Lexer::Token { 0, 0, type };
> }
> 
> class PreserveLifetimes : public Visitor {

Actions: View | Formatted Diff | Diff

Attachments on bug 199520: 373517 | 373518 | 373524