void NumericLiteralParser::checkSeparator(SourceLocation TokLoc,
const char *Pos,
CheckSeparatorKind IsAfterDigits) {
if (IsAfterDigits == CSK_AfterDigits) {
if (Pos == ThisTokBegin)
return;
--Pos;
} else if (Pos == ThisTokEnd)
return;
if (isDigitSeparator(*Pos))
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, Pos - ThisTokBegin),
diag::err_digit_separator_not_between_digits)
<< IsAfterDigits;
}
/// decimal integer: [0-9] [0-9_]*
/// binary integer: [01] [01_]* [bB]
/// binary integer: "0b" [01_]+
/// octal integer: [0-7] [0-7_]* [qQoO]
/// hex integer: [0-9] [0-9a-fA-F_]* [hH]
/// hex integer: [$] [0-9] [0-9a-fA-F_]*
/// hex integer: "0x" [0-9a-fA-F_]+
///
/// decimal float: [0-9]+ [.] [0-9]* ([eE] [-+]? [0-9]+)?
/// decimal float: [0-9]+ [eE] [-+]? [0-9]+
/// hex float: "0x" [0-9a-fA-F_]* [.] [0-9a-fA-F]* ([pP] [-+]? [0-9]+)?
/// hex float: "0x" [0-9a-fA-F_]+ [pP] [-+]? [0-9]+
///
NasmNumericParser::NasmNumericParser(llvm::StringRef str,
SourceLocation loc,
Preprocessor& pp)
: NumericParser(str)
{
// This routine assumes that the range begin/end matches the regex for
// integer and FP constants, and assumes that the byte at "*end" is both
// valid and not part of the regex. Because of this, it doesn't have to
// check for 'overscan' in various places.
assert(!isalnum(*m_digits_end) && *m_digits_end != '.' &&
*m_digits_end != '_' && "Lexer didn't maximally munch?");
const char* s = str.begin();
bool float_ok = false;
// Look for key radix flags (prefixes and suffixes)
if (*s == '$')
{
m_radix = 16;
++s;
}
else if (m_digits_end[-1] == 'b' || m_digits_end[-1] == 'B')
{
m_radix = 2;
--m_digits_end;
}
else if (m_digits_end[-1] == 'q' || m_digits_end[-1] == 'Q' ||
m_digits_end[-1] == 'o' || m_digits_end[-1] == 'O')
{
m_radix = 8;
--m_digits_end;
}
else if (m_digits_end[-1] == 'h' || m_digits_end[-1] == 'H')
{
m_radix = 16;
--m_digits_end;
}
else if (*s == '0' && (s[1] == 'x' || s[1] == 'X') &&
(isxdigit(s[2]) || s[2] == '.'))
{
m_radix = 16;
float_ok = true; // C99-style hex floating point
s += 2;
}
else if (*s == '0' && (s[1] == 'b' || s[1] == 'B') &&
(s[2] == '0' || s[2] == '1'))
{
m_radix = 2;
s += 2;
}
else
{
// Otherwise it's a decimal or float
m_radix = 10;
float_ok = true;
}
m_digits_begin = s;
switch (m_radix)
{
case 2: s = SkipBinaryDigits(s); break;
case 8: s = SkipOctalDigits(s); break;
case 10: s = SkipDigits(s); break;
case 16: s = SkipHexDigits(s); break;
}
if (s == m_digits_end)
{
// Done.
}
else if (isxdigit(*s) && (!float_ok || (*s != 'e' && *s != 'E')))
{
unsigned int err;
switch (m_radix)
{
case 2: err = diag::err_invalid_binary_digit; break;
case 8: err = diag::err_invalid_octal_digit; break;
case 10: err = diag::err_invalid_decimal_digit; break;
case 16:
default:
assert(false && "unexpected radix");
err = diag::err_invalid_decimal_digit;
break;
}
pp.Diag(pp.AdvanceToTokenCharacter(loc, s-str.begin()), err)
<< std::string(s, s+1);
m_had_error = true;
return;
}
else if (*s == '.' && float_ok)
{
++s;
m_is_float = true;
if (m_radix == 16)
s = SkipHexDigits(s);
else
s = SkipDigits(s);
}
if (float_ok &&
((m_radix == 10 && (*s == 'e' || *s == 'E')) ||
(m_radix == 16 && (*s == 'p' || *s == 'P'))))
{
// Float exponent
const char* exponent = s;
++s;
m_is_float = true;
if (*s == '+' || *s == '-') // sign
++s;
const char* first_non_digit = SkipDigits(s);
if (first_non_digit == s)
{
pp.Diag(pp.AdvanceToTokenCharacter(loc, exponent-str.begin()),
diag::err_exponent_has_no_digits);
m_had_error = true;
return;
}
s = first_non_digit;
}
// Report an error if there are any.
if (s != m_digits_end)
{
pp.Diag(pp.AdvanceToTokenCharacter(loc, s-str.begin()),
m_is_float ? diag::err_invalid_suffix_float_constant :
diag::err_invalid_suffix_integer_constant)
<< std::string(s, str.end());
m_had_error = true;
return;
}
}
/// decimal integer: [1-9] [0-9]*
/// binary integer: "0" [bB] [01]+
/// octal integer: "0" [0-7]*
/// hex integer: "0" [xX] [0-9a-fA-F]+
///
/// float: "0" [a-zA-Z except bB or xX]
/// [-+]? [0-9]* ([.] [0-9]*)? ([eE] [-+]? [0-9]+)?
///
GasNumericParser::GasNumericParser(llvm::StringRef str,
SourceLocation loc,
Preprocessor& pp,
bool force_float)
: NumericParser(str)
{
// This routine assumes that the range begin/end matches the regex for
// integer and FP constants, and assumes that the byte at "*end" is both
// valid and not part of the regex. Because of this, it doesn't have to
// check for 'overscan' in various places.
assert(!isalnum(*m_digits_end) && *m_digits_end != '.' &&
"Lexer didn't maximally munch?");
const char* s = str.begin();
// Look for key radix prefixes
if (force_float)
{
// forced decimal float; skip the prefix if present
m_radix = 10;
m_is_float = true;
if (*s == '0' && isalpha(s[1]))
s += 2;
}
else if (*s == '0' && (s[1] == 'x' || s[1] == 'X'))
{
m_radix = 16;
s += 2;
}
else if (*s == '0' && (s[1] == 'b' || s[1] == 'B'))
{
m_radix = 2;
s += 2;
}
else if (*s == '0' && isalpha(s[1]))
{
// it's a decimal float; skip the prefix
m_radix = 10;
s += 2;
m_is_float = true;
}
else if (*s == '0')
{
// It's an octal integer
m_radix = 8;
}
else
{
// Otherwise it's a decimal
m_radix = 10;
}
m_digits_begin = s;
switch (m_radix)
{
case 2: s = SkipBinaryDigits(s); break;
case 8: s = SkipOctalDigits(s); break;
case 10: s = SkipDigits(s); break;
case 16: s = SkipHexDigits(s); break;
}
if (s == m_digits_end)
{
// Done.
}
else if (isxdigit(*s) && (!m_is_float || (*s != 'e' && *s != 'E')))
{
unsigned int err;
switch (m_radix)
{
case 2: err = diag::err_invalid_binary_digit; break;
case 8: err = diag::err_invalid_octal_digit; break;
case 10: err = diag::err_invalid_decimal_digit; break;
case 16:
default:
assert(false && "unexpected radix");
err = diag::err_invalid_decimal_digit;
break;
}
pp.Diag(pp.AdvanceToTokenCharacter(loc, s-str.begin()), err)
<< std::string(s, s+1);
m_had_error = true;
return;
}
else if (m_is_float)
{
if (*s == '-' || *s == '+')
{
++s;
s = SkipDigits(s);
}
if (*s == '.')
{
++s;
s = SkipDigits(s);
}
if (*s == 'e' || *s == 'E')
{
// Float exponent
const char* exponent = s;
++s;
if (*s == '+' || *s == '-') // sign
++s;
const char* first_non_digit = SkipDigits(s);
if (first_non_digit == s)
{
pp.Diag(pp.AdvanceToTokenCharacter(loc, exponent-str.begin()),
diag::err_exponent_has_no_digits);
m_had_error = true;
return;
}
s = first_non_digit;
}
}
// Report an error if there are any.
if (s != m_digits_end)
{
pp.Diag(pp.AdvanceToTokenCharacter(loc, s-str.begin()),
m_is_float ? diag::err_invalid_suffix_float_constant :
diag::err_invalid_suffix_integer_constant)
<< std::string(s, str.end());
m_had_error = true;
return;
}
}
/// ParseNumberStartingWithZero - This method is called when the first character
/// of the number is found to be a zero. This means it is either an octal
/// number (like '04') or a hex number ('0x123a') a binary number ('0b1010') or
/// a floating point number (01239.123e4). Eat the prefix, determining the
/// radix etc.
void NumericLiteralParser::ParseNumberStartingWithZero(SourceLocation TokLoc) {
assert(s[0] == '0' && "Invalid method call");
s++;
int c1 = s[0];
// Handle a hex number like 0x1234.
if ((c1 == 'x' || c1 == 'X') && (isHexDigit(s[1]) || s[1] == '.')) {
s++;
assert(s < ThisTokEnd && "didn't maximally munch?");
radix = 16;
DigitsBegin = s;
s = SkipHexDigits(s);
bool noSignificand = (s == DigitsBegin);
if (s == ThisTokEnd) {
// Done.
} else if (*s == '.') {
s++;
saw_period = true;
const char *floatDigitsBegin = s;
checkSeparator(TokLoc, s, CSK_BeforeDigits);
s = SkipHexDigits(s);
noSignificand &= (floatDigitsBegin == s);
}
if (noSignificand) {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, s - ThisTokBegin),
diag::err_hexconstant_requires_digits);
hadError = true;
return;
}
// A binary exponent can appear with or with a '.'. If dotted, the
// binary exponent is required.
if (*s == 'p' || *s == 'P') {
checkSeparator(TokLoc, s, CSK_AfterDigits);
const char *Exponent = s;
s++;
saw_exponent = true;
if (*s == '+' || *s == '-') s++; // sign
const char *first_non_digit = SkipDigits(s);
if (first_non_digit == s) {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, Exponent-ThisTokBegin),
diag::err_exponent_has_no_digits);
hadError = true;
return;
}
checkSeparator(TokLoc, s, CSK_BeforeDigits);
s = first_non_digit;
if (!PP.getLangOpts().HexFloats)
PP.Diag(TokLoc, diag::ext_hexconstant_invalid);
} else if (saw_period) {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, s-ThisTokBegin),
diag::err_hexconstant_requires_exponent);
hadError = true;
}
return;
}
// Handle simple binary numbers 0b01010
if ((c1 == 'b' || c1 == 'B') && (s[1] == '0' || s[1] == '1')) {
// 0b101010 is a C++1y / GCC extension.
PP.Diag(TokLoc,
PP.getLangOpts().CPlusPlus14
? diag::warn_cxx11_compat_binary_literal
: PP.getLangOpts().CPlusPlus
? diag::ext_binary_literal_cxx14
: diag::ext_binary_literal);
++s;
assert(s < ThisTokEnd && "didn't maximally munch?");
radix = 2;
DigitsBegin = s;
s = SkipBinaryDigits(s);
if (s == ThisTokEnd) {
// Done.
} else if (isHexDigit(*s)) {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, s-ThisTokBegin),
diag::err_invalid_binary_digit) << StringRef(s, 1);
hadError = true;
}
// Other suffixes will be diagnosed by the caller.
return;
}
// For now, the radix is set to 8. If we discover that we have a
// floating point constant, the radix will change to 10. Octal floating
// point constants are not permitted (only decimal and hexadecimal).
radix = 8;
DigitsBegin = s;
s = SkipOctalDigits(s);
if (s == ThisTokEnd)
return; // Done, simple octal number like 01234
// If we have some other non-octal digit that *is* a decimal digit, see if
// this is part of a floating point number like 094.123 or 09e1.
if (isDigit(*s)) {
const char *EndDecimal = SkipDigits(s);
if (EndDecimal[0] == '.' || EndDecimal[0] == 'e' || EndDecimal[0] == 'E') {
s = EndDecimal;
radix = 10;
}
}
// If we have a hex digit other than 'e' (which denotes a FP exponent) then
// the code is using an incorrect base.
if (isHexDigit(*s) && *s != 'e' && *s != 'E') {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, s-ThisTokBegin),
diag::err_invalid_octal_digit) << StringRef(s, 1);
hadError = true;
return;
}
if (*s == '.') {
s++;
radix = 10;
saw_period = true;
checkSeparator(TokLoc, s, CSK_BeforeDigits);
s = SkipDigits(s); // Skip suffix.
}
if (*s == 'e' || *s == 'E') { // exponent
checkSeparator(TokLoc, s, CSK_AfterDigits);
const char *Exponent = s;
s++;
radix = 10;
saw_exponent = true;
if (*s == '+' || *s == '-') s++; // sign
const char *first_non_digit = SkipDigits(s);
if (first_non_digit != s) {
checkSeparator(TokLoc, s, CSK_BeforeDigits);
s = first_non_digit;
} else {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, Exponent-ThisTokBegin),
diag::err_exponent_has_no_digits);
hadError = true;
return;
}
}
}
/// integer-constant: [C99 6.4.4.1]
/// decimal-constant integer-suffix
/// octal-constant integer-suffix
/// hexadecimal-constant integer-suffix
/// binary-literal integer-suffix [GNU, C++1y]
/// user-defined-integer-literal: [C++11 lex.ext]
/// decimal-literal ud-suffix
/// octal-literal ud-suffix
/// hexadecimal-literal ud-suffix
/// binary-literal ud-suffix [GNU, C++1y]
/// decimal-constant:
/// nonzero-digit
/// decimal-constant digit
/// octal-constant:
/// 0
/// octal-constant octal-digit
/// hexadecimal-constant:
/// hexadecimal-prefix hexadecimal-digit
/// hexadecimal-constant hexadecimal-digit
/// hexadecimal-prefix: one of
/// 0x 0X
/// binary-literal:
/// 0b binary-digit
/// 0B binary-digit
/// binary-literal binary-digit
/// integer-suffix:
/// unsigned-suffix [long-suffix]
/// unsigned-suffix [long-long-suffix]
/// long-suffix [unsigned-suffix]
/// long-long-suffix [unsigned-sufix]
/// nonzero-digit:
/// 1 2 3 4 5 6 7 8 9
/// octal-digit:
/// 0 1 2 3 4 5 6 7
/// hexadecimal-digit:
/// 0 1 2 3 4 5 6 7 8 9
/// a b c d e f
/// A B C D E F
/// binary-digit:
/// 0
/// 1
/// unsigned-suffix: one of
/// u U
/// long-suffix: one of
/// l L
/// long-long-suffix: one of
/// ll LL
///
/// floating-constant: [C99 6.4.4.2]
/// TODO: add rules...
///
NumericLiteralParser::NumericLiteralParser(StringRef TokSpelling)
: ThisTokBegin(TokSpelling.begin()), ThisTokEnd(TokSpelling.end()) {
// This routine assumes that the range begin/end matches the regex for integer
// and FP constants (specifically, the 'pp-number' regex), and assumes that
// the byte at "*end" is both valid and not part of the regex. Because of
// this, it doesn't have to check for 'overscan' in various places.
assert(!isPreprocessingNumberBody(*ThisTokEnd) && "didn't maximally munch?");
s = DigitsBegin = ThisTokBegin;
saw_exponent = false;
saw_period = false;
saw_ud_suffix = false;
isLong = false;
isUnsigned = false;
isLongLong = false;
isFloat = false;
isImaginary = false;
MicrosoftInteger = 0;
hadError = false;
SourceLocation TokLoc;
if (*s == '0') { // parse radix
ParseNumberStartingWithZero(TokLoc);
if (hadError)
return;
} else { // the first digit is non-zero
radix = 10;
s = SkipDigits(s);
if (s == ThisTokEnd) {
// Done.
} else if (isHexDigit(*s) && !(*s == 'e' || *s == 'E')) {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, s - ThisTokBegin),
diag::err_invalid_decimal_digit) << StringRef(s, 1);
hadError = true;
return;
} else if (*s == '.') {
checkSeparator(TokLoc, s, CSK_AfterDigits);
s++;
saw_period = true;
checkSeparator(TokLoc, s, CSK_BeforeDigits);
s = SkipDigits(s);
}
if ((*s == 'e' || *s == 'E')) { // exponent
checkSeparator(TokLoc, s, CSK_AfterDigits);
const char *Exponent = s;
s++;
saw_exponent = true;
if (*s == '+' || *s == '-') s++; // sign
checkSeparator(TokLoc, s, CSK_BeforeDigits);
const char *first_non_digit = SkipDigits(s);
if (first_non_digit != s) {
s = first_non_digit;
} else {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, Exponent - ThisTokBegin),
diag::err_exponent_has_no_digits);
hadError = true;
return;
}
}
}
SuffixBegin = s;
checkSeparator(TokLoc, s, CSK_AfterDigits);
// Parse the suffix. At this point we can classify whether we have an FP or
// integer constant.
bool isFPConstant = isFloatingLiteral();
const char *ImaginarySuffixLoc = nullptr;
// Loop over all of the characters of the suffix. If we see something bad,
// we break out of the loop.
for (; s != ThisTokEnd; ++s) {
switch (*s) {
case 'f': // FP Suffix for "float"
case 'F':
if (!isFPConstant) break; // Error for integer constant.
if (isFloat || isLong) break; // FF, LF invalid.
isFloat = true;
continue; // Success.
case 'u':
case 'U':
if (isFPConstant) break; // Error for floating constant.
if (isUnsigned) break; // Cannot be repeated.
isUnsigned = true;
continue; // Success.
case 'l':
case 'L':
if (isLong || isLongLong) break; // Cannot be repeated.
if (isFloat) break; // LF invalid.
// Check for long long. The L's need to be adjacent and the same case.
if (s[1] == s[0]) {
assert(s + 1 < ThisTokEnd && "didn't maximally munch?");
if (isFPConstant) break; // long long invalid for floats.
isLongLong = true;
++s; // Eat both of them.
} else {
isLong = true;
}
continue; // Success.
case 'i':
case 'I':
if (PP.getLangOpts().MicrosoftExt) {
if (isLong || isLongLong || MicrosoftInteger)
break;
if (!isFPConstant) {
// Allow i8, i16, i32, i64, and i128.
switch (s[1]) {
case '8':
s += 2; // i8 suffix
MicrosoftInteger = 8;
break;
case '1':
if (s[2] == '6') {
s += 3; // i16 suffix
MicrosoftInteger = 16;
} else if (s[2] == '2' && s[3] == '8') {
s += 4; // i128 suffix
MicrosoftInteger = 128;
}
break;
case '3':
if (s[2] == '2') {
s += 3; // i32 suffix
MicrosoftInteger = 32;
}
break;
case '6':
if (s[2] == '4') {
s += 3; // i64 suffix
MicrosoftInteger = 64;
}
break;
default:
break;
}
}
if (MicrosoftInteger) {
assert(s <= ThisTokEnd && "didn't maximally munch?");
break;
}
}
// "i", "if", and "il" are user-defined suffixes in C++1y.
if (*s == 'i' && PP.getLangOpts().CPlusPlus14)
break;
// fall through.
case 'j':
case 'J':
if (isImaginary) break; // Cannot be repeated.
isImaginary = true;
ImaginarySuffixLoc = s;
continue; // Success.
}
// If we reached here, there was an error or a ud-suffix.
break;
}
if (s != ThisTokEnd) {
// FIXME: Don't bother expanding UCNs if !tok.hasUCN().
expandUCNs(UDSuffixBuf, StringRef(SuffixBegin, ThisTokEnd - SuffixBegin));
if (isValidUDSuffix(PP.getLangOpts(), UDSuffixBuf)) {
// Any suffix pieces we might have parsed are actually part of the
// ud-suffix.
isLong = false;
isUnsigned = false;
isLongLong = false;
isFloat = false;
isImaginary = false;
MicrosoftInteger = 0;
saw_ud_suffix = true;
return;
}
// Report an error if there are any.
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc, SuffixBegin - ThisTokBegin),
isFPConstant ? diag::err_invalid_suffix_float_constant :
diag::err_invalid_suffix_integer_constant)
<< StringRef(SuffixBegin, ThisTokEnd-SuffixBegin);
hadError = true;
return;
}
if (isImaginary) {
PP.Diag(PP.AdvanceToTokenCharacter(TokLoc,
ImaginarySuffixLoc - ThisTokBegin),
diag::ext_imaginary_constant);
}
}