static Cursor maybeLexMachineBasicBlock(Cursor C, MIToken &Token,
ErrorCallbackType ErrorCallback) {
bool IsReference = C.remaining().startswith("%bb.");
if (!IsReference && !C.remaining().startswith("bb."))
return None;
auto Range = C;
unsigned PrefixLength = IsReference ? 4 : 3;
C.advance(PrefixLength); // Skip '%bb.' or 'bb.'
if (!isdigit(C.peek())) {
Token.reset(MIToken::Error, C.remaining());
ErrorCallback(C.location(), "expected a number after '%bb.'");
return C;
}
auto NumberRange = C;
while (isdigit(C.peek()))
C.advance();
StringRef Number = NumberRange.upto(C);
unsigned StringOffset = PrefixLength + Number.size(); // Drop '%bb.<id>'
// TODO: The format bb.<id>.<irname> is supported only when it's not a
// reference. Once we deprecate the format where the irname shows up, we
// should only lex forward if it is a reference.
if (C.peek() == '.') {
C.advance(); // Skip '.'
++StringOffset;
while (isIdentifierChar(C.peek()))
C.advance();
}
Token.reset(IsReference ? MIToken::MachineBasicBlock
: MIToken::MachineBasicBlockLabel,
Range.upto(C))
.setIntegerValue(APSInt(Number))
.setStringValue(Range.upto(C).drop_front(StringOffset));
return C;
}
StringRef llvm::lexMIToken(StringRef Source, MIToken &Token,
ErrorCallbackType ErrorCallback) {
auto C = skipComment(skipWhitespace(Cursor(Source)));
if (C.isEOF()) {
Token.reset(MIToken::Eof, C.remaining());
return C.remaining();
}
if (Cursor R = maybeLexIntegerOrScalarType(C, Token))
return R.remaining();
if (Cursor R = maybeLexMachineBasicBlock(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexIdentifier(C, Token))
return R.remaining();
if (Cursor R = maybeLexJumpTableIndex(C, Token))
return R.remaining();
if (Cursor R = maybeLexStackObject(C, Token))
return R.remaining();
if (Cursor R = maybeLexFixedStackObject(C, Token))
return R.remaining();
if (Cursor R = maybeLexConstantPoolItem(C, Token))
return R.remaining();
if (Cursor R = maybeLexSubRegisterIndex(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexIRBlock(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexIRValue(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexRegister(C, Token))
return R.remaining();
if (Cursor R = maybeLexGlobalValue(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexExternalSymbol(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexHexadecimalLiteral(C, Token))
return R.remaining();
if (Cursor R = maybeLexNumericalLiteral(C, Token))
return R.remaining();
if (Cursor R = maybeLexExlaim(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexSymbol(C, Token))
return R.remaining();
if (Cursor R = maybeLexNewline(C, Token))
return R.remaining();
if (Cursor R = maybeLexEscapedIRValue(C, Token, ErrorCallback))
return R.remaining();
if (Cursor R = maybeLexStringConstant(C, Token, ErrorCallback))
return R.remaining();
Token.reset(MIToken::Error, C.remaining());
ErrorCallback(C.location(),
Twine("unexpected character '") + Twine(C.peek()) + "'");
return C.remaining();
}
static Cursor maybeLexNewline(Cursor C, MIToken &Token) {
if (!isNewlineChar(C.peek()))
return None;
auto Range = C;
C.advance();
Token.reset(MIToken::Newline, Range.upto(C));
return C;
}
static Cursor maybeLexIntegerType(Cursor C, MIToken &Token) {
if (C.peek() != 'i' || !isdigit(C.peek(1)))
return None;
auto Range = C;
C.advance(); // Skip 'i'
while (isdigit(C.peek()))
C.advance();
Token.reset(MIToken::IntegerType, Range.upto(C));
return C;
}
static Cursor lexVirtualRegister(Cursor C, MIToken &Token) {
auto Range = C;
C.advance(); // Skip '%'
auto NumberRange = C;
while (isdigit(C.peek()))
C.advance();
Token.reset(MIToken::VirtualRegister, Range.upto(C))
.setIntegerValue(APSInt(NumberRange.upto(C)));
return C;
}
static Cursor maybeLexExlaim(Cursor C, MIToken &Token,
ErrorCallbackType ErrorCallback) {
if (C.peek() != '!')
return None;
auto Range = C;
C.advance(1);
if (isdigit(C.peek()) || !isIdentifierChar(C.peek())) {
Token.reset(MIToken::exclaim, Range.upto(C));
return C;
}
while (isIdentifierChar(C.peek()))
C.advance();
StringRef StrVal = Range.upto(C);
Token.reset(getMetadataKeywordKind(StrVal), StrVal);
if (Token.isError())
ErrorCallback(Token.location(),
"use of unknown metadata keyword '" + StrVal + "'");
return C;
}
static Cursor maybeLexIdentifier(Cursor C, MIToken &Token) {
if (!isalpha(C.peek()) && C.peek() != '_')
return None;
auto Range = C;
while (isIdentifierChar(C.peek()))
C.advance();
auto Identifier = Range.upto(C);
Token.reset(getIdentifierKind(Identifier), Identifier)
.setStringValue(Identifier);
return C;
}
static Cursor lexName(Cursor C, MIToken &Token, MIToken::TokenKind Type,
unsigned PrefixLength, ErrorCallbackType ErrorCallback) {
auto Range = C;
C.advance(PrefixLength);
if (C.peek() == '"') {
if (Cursor R = lexStringConstant(C, ErrorCallback)) {
StringRef String = Range.upto(R);
Token.reset(Type, String)
.setOwnedStringValue(
unescapeQuotedString(String.drop_front(PrefixLength)));
return R;
}
Token.reset(MIToken::Error, Range.remaining());
return Range;
}
while (isIdentifierChar(C.peek()))
C.advance();
Token.reset(Type, Range.upto(C))
.setStringValue(Range.upto(C).drop_front(PrefixLength));
return C;
}
static Cursor maybeLexHexadecimalLiteral(Cursor C, MIToken &Token) {
if (C.peek() != '0' || (C.peek(1) != 'x' && C.peek(1) != 'X'))
return None;
Cursor Range = C;
C.advance(2);
unsigned PrefLen = 2;
if (isValidHexFloatingPointPrefix(C.peek())) {
C.advance();
PrefLen++;
}
while (isxdigit(C.peek()))
C.advance();
StringRef StrVal = Range.upto(C);
if (StrVal.size() <= PrefLen)
return None;
if (PrefLen == 2)
Token.reset(MIToken::HexLiteral, Range.upto(C));
else // It must be 3, which means that there was a floating-point prefix.
Token.reset(MIToken::FloatingPointLiteral, Range.upto(C));
return C;
}
static Cursor maybeLexHexFloatingPointLiteral(Cursor C, MIToken &Token) {
if (C.peek() != '0' || C.peek(1) != 'x')
return None;
Cursor Range = C;
C.advance(2); // Skip '0x'
if (isValidHexFloatingPointPrefix(C.peek()))
C.advance();
while (isxdigit(C.peek()))
C.advance();
Token.reset(MIToken::FloatingPointLiteral, Range.upto(C));
return C;
}
static Cursor maybeLexIndex(Cursor C, MIToken &Token, StringRef Rule,
MIToken::TokenKind Kind) {
if (!C.remaining().startswith(Rule) || !isdigit(C.peek(Rule.size())))
return None;
auto Range = C;
C.advance(Rule.size());
auto NumberRange = C;
while (isdigit(C.peek()))
C.advance();
Token.reset(Kind, Range.upto(C)).setIntegerValue(APSInt(NumberRange.upto(C)));
return C;
}
static Cursor maybeLexRegister(Cursor C, MIToken &Token) {
if (C.peek() != '%')
return None;
if (isdigit(C.peek(1)))
return lexVirtualRegister(C, Token);
auto Range = C;
C.advance(); // Skip '%'
while (isRegisterChar(C.peek()))
C.advance();
Token.reset(MIToken::NamedRegister, Range.upto(C))
.setStringValue(Range.upto(C).drop_front(1)); // Drop the '%'
return C;
}
static Cursor maybeLexEscapedIRValue(Cursor C, MIToken &Token,
ErrorCallbackType ErrorCallback) {
if (C.peek() != '`')
return None;
auto Range = C;
C.advance();
auto StrRange = C;
while (C.peek() != '`') {
if (C.isEOF() || isNewlineChar(C.peek())) {
ErrorCallback(
C.location(),
"end of machine instruction reached before the closing '`'");
Token.reset(MIToken::Error, Range.remaining());
return C;
}
C.advance();
}
StringRef Value = StrRange.upto(C);
C.advance();
Token.reset(MIToken::QuotedIRValue, Range.upto(C)).setStringValue(Value);
return C;
}
static Cursor maybeLexNumericalLiteral(Cursor C, MIToken &Token) {
if (!isdigit(C.peek()) && (C.peek() != '-' || !isdigit(C.peek(1))))
return None;
auto Range = C;
C.advance();
while (isdigit(C.peek()))
C.advance();
if (C.peek() == '.')
return lexFloatingPointLiteral(Range, C, Token);
StringRef StrVal = Range.upto(C);
Token.reset(MIToken::IntegerLiteral, StrVal).setIntegerValue(APSInt(StrVal));
return C;
}
static Cursor maybeLexSymbol(Cursor C, MIToken &Token) {
MIToken::TokenKind Kind;
unsigned Length = 1;
if (C.peek() == ':' && C.peek(1) == ':') {
Kind = MIToken::coloncolon;
Length = 2;
} else
Kind = symbolToken(C.peek());
if (Kind == MIToken::Error)
return None;
auto Range = C;
C.advance(Length);
Token.reset(Kind, Range.upto(C));
return C;
}
static Cursor lexFloatingPointLiteral(Cursor Range, Cursor C, MIToken &Token) {
C.advance();
// Skip over [0-9]*([eE][-+]?[0-9]+)?
while (isdigit(C.peek()))
C.advance();
if ((C.peek() == 'e' || C.peek() == 'E') &&
(isdigit(C.peek(1)) ||
((C.peek(1) == '-' || C.peek(1) == '+') && isdigit(C.peek(2))))) {
C.advance(2);
while (isdigit(C.peek()))
C.advance();
}
Token.reset(MIToken::FloatingPointLiteral, Range.upto(C));
return C;
}
static Cursor maybeLexGlobalValue(Cursor C, MIToken &Token,
ErrorCallbackType ErrorCallback) {
if (C.peek() != '@')
return None;
if (!isdigit(C.peek(1)))
return lexName(C, Token, MIToken::NamedGlobalValue, /*PrefixLength=*/1,
ErrorCallback);
auto Range = C;
C.advance(1); // Skip the '@'
auto NumberRange = C;
while (isdigit(C.peek()))
C.advance();
Token.reset(MIToken::GlobalValue, Range.upto(C))
.setIntegerValue(APSInt(NumberRange.upto(C)));
return C;
}
static Cursor maybeLexIntegerOrScalarType(Cursor C, MIToken &Token) {
if ((C.peek() != 'i' && C.peek() != 's' && C.peek() != 'p') ||
!isdigit(C.peek(1)))
return None;
char Kind = C.peek();
auto Range = C;
C.advance(); // Skip 'i', 's', or 'p'
while (isdigit(C.peek()))
C.advance();
Token.reset(Kind == 'i'
? MIToken::IntegerType
: (Kind == 's' ? MIToken::ScalarType : MIToken::PointerType),
Range.upto(C));
return C;
}
static Cursor maybeLexIndexAndName(Cursor C, MIToken &Token, StringRef Rule,
MIToken::TokenKind Kind) {
if (!C.remaining().startswith(Rule) || !isdigit(C.peek(Rule.size())))
return None;
auto Range = C;
C.advance(Rule.size());
auto NumberRange = C;
while (isdigit(C.peek()))
C.advance();
StringRef Number = NumberRange.upto(C);
unsigned StringOffset = Rule.size() + Number.size();
if (C.peek() == '.') {
C.advance();
++StringOffset;
while (isIdentifierChar(C.peek()))
C.advance();
}
Token.reset(Kind, Range.upto(C))
.setIntegerValue(APSInt(Number))
.setStringValue(Range.upto(C).drop_front(StringOffset));
return C;
}
