Merge branch 'main' into main

CMakeLists.txt

@@ -1,6 +1,6 @@
 cmake_minimum_required(VERSION 3.9)
 
-project(fast_float VERSION 5.3.0 LANGUAGES CXX)
+project(fast_float VERSION 6.0.0 LANGUAGES CXX)
 option(FASTFLOAT_TEST "Enable tests" OFF)
 if(FASTFLOAT_TEST)
   enable_testing()

CONTRIBUTORS

@@ -7,4 +7,4 @@ Fabio Pellacini
 Lénárd Szolnoki
 Jan Pharago
 Maya Warrier
-Taha Khokhar
+Taha Khokhar

README.md

@@ -4,17 +4,21 @@
 [![Ubuntu 22.04 CI (GCC 11)](https://github.com/fastfloat/fast_float/actions/workflows/ubuntu22.yml/badge.svg)](https://github.com/fastfloat/fast_float/actions/workflows/ubuntu22.yml)
 
 The fast_float library provides fast header-only implementations for the C++ from_chars
-functions for `float` and `double` types.  These functions convert ASCII strings representing
-decimal values (e.g., `1.3e10`) into binary types. We provide exact rounding (including
+functions for `float` and `double` types as well as integer types.  These functions convert ASCII strings representing decimal values (e.g., `1.3e10`) into binary types. We provide exact rounding (including
 round to even). In our experience, these `fast_float` functions many times faster than comparable number-parsing functions from existing C++ standard libraries.
 
-Specifically, `fast_float` provides the following two functions with a C++17-like syntax (the library itself only requires C++11):
+Specifically, `fast_float` provides the following two functions to parse floating-point numbers with a C++17-like syntax (the library itself only requires C++11):
 
 ```C++
 from_chars_result from_chars(const char* first, const char* last, float& value, ...);
 from_chars_result from_chars(const char* first, const char* last, double& value, ...);
 ```
 
+You can also parse integer types:
+
+
+
+
 The return type (`from_chars_result`) is defined as the struct:
 ```C++
 struct from_chars_result {
@@ -103,6 +107,43 @@ We support Visual Studio, macOS, Linux, freeBSD. We support big and little endia
 
 We assume that the rounding mode is set to nearest (`std::fegetround() == FE_TONEAREST`).
 
+
+## Integer types
+
+You can also parse integer types using different bases (e.g., 2, 10, 16). The following code will
+print the number 22250738585072012 three times:
+
+
+```C++
+  uint64_t i;
+  const char str[] = "22250738585072012";
+  auto answer = fast_float::from_chars(str, str + strlen(str), i);
+  if (answer.ec != std::errc()) {
+    std::cerr << "parsing failure\n";
+    return EXIT_FAILURE;
+  }
+  std::cout << "parsed the number "<< i << std::endl;
+
+  const char binstr[] = "1001111000011001110110111001001010110100111000110001100";
+
+  answer = fast_float::from_chars(binstr, binstr + strlen(binstr), i, 2);
+  if (answer.ec != std::errc()) {
+    std::cerr << "parsing failure\n";
+    return EXIT_FAILURE;
+  }
+  std::cout << "parsed the number "<< i << std::endl;
+
+
+  const char hexstr[] = "4f0cedc95a718c";
+
+  answer = fast_float::from_chars(hexstr, hexstr + strlen(hexstr), i, 16);
+  if (answer.ec != std::errc()) {
+    std::cerr << "parsing failure\n";
+    return EXIT_FAILURE;
+  }
+  std::cout << "parsed the number "<< i << std::endl;
+```
+
 ## C++20: compile-time evaluation (constexpr)
 
 In C++20, you may use `fast_float::from_chars` to parse strings
@@ -264,7 +305,7 @@ The fast_float library provides a performance similar to that of the [fast_doubl
 
 ## Users
 
-The fast_float library is used by [Apache Arrow](https://github.com/apache/arrow/pull/8494) where it multiplied the number parsing speed by two or three times. It is also used by [Yandex ClickHouse](https://github.com/ClickHouse/ClickHouse) and by [Google Jsonnet](https://github.com/google/jsonnet).
+The fast_float library is used by [Apache Arrow](https://github.com/apache/arrow/pull/8494) where it multiplied the number parsing speed by two or three times. It is also used by [ClickHouse](https://github.com/ClickHouse/ClickHouse) and by [Google Jsonnet](https://github.com/google/jsonnet). It is part of GCC (as of GCC 12). It is part of WebKit (Safari).
 
 
 ## How fast is it?
@@ -330,7 +371,11 @@ the command line help.
 
 You may directly download automatically generated single-header files:
 
-https://github.com/fastfloat/fast_float/releases/download/v5.3.0/fast_float.h
+https://github.com/fastfloat/fast_float/releases/download/v6.0.0/fast_float.h
+
+## RFC 7159 
+
+If you need support for RFC 7159 (JSON standard), you may want to consider using the [fast_double_parser](https://github.com/lemire/fast_double_parser/) library instead.
 
 ## Credit
 

include/fast_float/ascii_number.h

@@ -5,6 +5,7 @@
 #include <cstdint>
 #include <cstring>
 #include <iterator>
+#include <limits>
 #include <type_traits>
 
 #include "float_common.h"
@@ -115,7 +116,7 @@ FASTFLOAT_SIMD_RESTORE_WARNINGS
 #if defined(_MSC_VER) && _MSC_VER <= 1900
 template <typename UC>
 #else
-template <typename UC, FASTFLOAT_ENABLE_IF(!has_simd_opt<UC>())>
+template <typename UC, FASTFLOAT_ENABLE_IF(!has_simd_opt<UC>()) = 0>
 #endif
 // dummy for compile
 uint64_t simd_read8_to_u64(UC const*) {
@@ -223,15 +224,15 @@ FASTFLOAT_SIMD_RESTORE_WARNINGS
 #if defined(_MSC_VER) && _MSC_VER <= 1900
 template <typename UC>
 #else
-template <typename UC, FASTFLOAT_ENABLE_IF(!has_simd_opt<UC>())>
+template <typename UC, FASTFLOAT_ENABLE_IF(!has_simd_opt<UC>()) = 0>
 #endif
 // dummy for compile
 bool simd_parse_if_eight_digits_unrolled(UC const*, uint64_t&) {
   return 0;
 }
 
 
-template <typename UC, FASTFLOAT_ENABLE_IF(!std::is_same<UC, char>::value)>
+template <typename UC, FASTFLOAT_ENABLE_IF(!std::is_same<UC, char>::value) = 0>
 fastfloat_really_inline FASTFLOAT_CONSTEXPR20
 void loop_parse_if_eight_digits(const UC*& p, const UC* const pend, uint64_t& i) {
   if (!has_simd_opt<UC>()) {
@@ -439,6 +440,106 @@ parsed_number_string_t<UC> parse_number_string(UC const *p, UC const * pend, par
   return answer;
 }
 
+template <typename T, typename UC>
+fastfloat_really_inline FASTFLOAT_CONSTEXPR20
+from_chars_result_t<UC> parse_int_string(UC const* p, UC const* pend, T& value, int base)
+{
+  from_chars_result_t<UC> answer;
+
+  UC const* const first = p;
+
+  bool negative = (*p == UC('-'));
+  if (!std::is_signed<T>::value && negative) {
+    answer.ec = std::errc::invalid_argument;
+    answer.ptr = first;
+    return answer;
+  }
+#ifdef FASTFLOAT_ALLOWS_LEADING_PLUS // disabled by default
+  if ((*p == UC('-')) || (*p == UC('+'))) {
+#else
+  if (*p == UC('-')) {
+#endif
+    ++p;
+  }
+
+  UC const* const start_num = p;
+  while (*p == UC('0')) { 
+    ++p; 
+  }
+  const bool has_leading_zeros = p > start_num;
+
+  UC const* const start_digits = p;
+
+  uint64_t i = 0;
+  if (base == 10) {
+    loop_parse_if_eight_digits(p, pend, i); // use SIMD if possible
+  }
+  while (p != pend) {
+    uint8_t digit = ch_to_digit(*p);
+    if (digit >= base) {
+      break;
+    }
+    i = uint64_t(base) * i + digit; // might overflow, check this later
+    p++; 
+  }
+
+  size_t digit_count = size_t(p - start_digits);
+
+  if (digit_count == 0) {
+    if (has_leading_zeros) {
+      value = 0;
+      answer.ec = std::errc();
+      answer.ptr = p;
+    }
+    else {
+      answer.ec = std::errc::invalid_argument;
+      answer.ptr = first;
+    }
+    return answer; 
+  }
+
+  answer.ptr = p;
+
+  // check u64 overflow
+  size_t max_digits = max_digits_u64(base);
+  if (digit_count > max_digits) {
+    answer.ec = std::errc::result_out_of_range;
+    return answer;
+  }
+  // this check can be eliminated for all other types, but they will all require a max_digits(base) equivalent
+  if (digit_count == max_digits && i < min_safe_u64(base)) {
+    answer.ec = std::errc::result_out_of_range;
+    return answer;
+  }
+
+  // check other types overflow
+  if (!std::is_same<T, uint64_t>::value) {
+    if (i > uint64_t(std::numeric_limits<T>::max()) + uint64_t(negative)) {
+      answer.ec = std::errc::result_out_of_range;
+      return answer;
+    }
+  }
+
+  if (negative) {
+#ifdef FASTFLOAT_VISUAL_STUDIO
+#pragma warning(push)
+#pragma warning(disable: 4146) 
+#endif
+    // this weird workaround is required because:
+    // - converting unsigned to signed when its value is greater than signed max is UB pre-C++23.
+    // - reinterpret_casting (~i + 1) would work, but it is not constexpr
+    // this is always optimized into a neg instruction.
+    value = T(-std::numeric_limits<T>::max() - T(i - std::numeric_limits<T>::max()));
+#ifdef FASTFLOAT_VISUAL_STUDIO
+#pragma warning(pop)
+#endif
+  }
+  else { value = T(i); }
+
+  answer.ec = std::errc();
+  return answer;
+}
+
 } // namespace fast_float
 
 #endif

include/fast_float/fast_float.h

@@ -24,7 +24,7 @@ namespace fast_float {
  * to determine whether we allow the fixed point and scientific notation respectively.
  * The default is  `fast_float::chars_format::general` which allows both `fixed` and `scientific`.
  */
-template<typename T, typename UC = char>
+template<typename T, typename UC = char, typename = FASTFLOAT_ENABLE_IF(is_supported_float_type<T>())>
 FASTFLOAT_CONSTEXPR20
 from_chars_result_t<UC> from_chars(UC const * first, UC const * last,
                              T &value, chars_format fmt = chars_format::general)  noexcept;
@@ -36,6 +36,12 @@ template<typename T, typename UC = char>
 FASTFLOAT_CONSTEXPR20
 from_chars_result_t<UC> from_chars_advanced(UC const * first, UC const * last,
                                       T &value, parse_options_t<UC> options)  noexcept;
+/**
+* from_chars for integer types.
+*/
+template <typename T, typename UC = char, typename = FASTFLOAT_ENABLE_IF(!is_supported_float_type<T>())>
+FASTFLOAT_CONSTEXPR20
+from_chars_result_t<UC> from_chars(UC const * first, UC const * last, T& value, int base = 10) noexcept;
 
 } // namespace fast_float
 #include "parse_number.h"

include/fast_float/float_common.h

@@ -134,7 +134,7 @@ using parse_options = parse_options_t<char>;
 #define FASTFLOAT_NEON 1
 #endif
 
-#if defined(FASTFLOAT_SSE2) || defined(FASTFLOAT_ARM64)
+#if defined(FASTFLOAT_SSE2) || defined(FASTFLOAT_NEON)
 #define FASTFLOAT_HAS_SIMD 1
 #endif
 
@@ -173,7 +173,7 @@ using parse_options = parse_options_t<char>;
 // rust style `try!()` macro, or `?` operator
 #define FASTFLOAT_TRY(x) { if (!(x)) return false; }
 
-#define FASTFLOAT_ENABLE_IF(...) typename std::enable_if<(__VA_ARGS__), int>::type = 0
+#define FASTFLOAT_ENABLE_IF(...) typename std::enable_if<(__VA_ARGS__), int>::type
 
 
 namespace fast_float {
@@ -186,6 +186,20 @@ fastfloat_really_inline constexpr bool cpp20_and_in_constexpr() {
 #endif
 }
 
+template <typename T>
+fastfloat_really_inline constexpr bool is_supported_float_type() {
+  return std::is_same<T, float>::value || std::is_same<T, double>::value;
+}
+
+template <typename UC>
+fastfloat_really_inline constexpr bool is_supported_char_type() {
+  return
+    std::is_same<UC, char>::value ||
+    std::is_same<UC, wchar_t>::value ||
+    std::is_same<UC, char16_t>::value ||
+    std::is_same<UC, char32_t>::value;
+}
+
 // Compares two ASCII strings in a case insensitive manner.
 template <typename UC>
 inline FASTFLOAT_CONSTEXPR14 bool
@@ -270,10 +284,10 @@ uint64_t umul128_generic(uint64_t ab, uint64_t cd, uint64_t *hi) {
   uint64_t ad = emulu((uint32_t)(ab >> 32), (uint32_t)cd);
   uint64_t bd = emulu((uint32_t)ab, (uint32_t)cd);
   uint64_t adbc = ad + emulu((uint32_t)ab, (uint32_t)(cd >> 32));
-  uint64_t adbc_carry = !!(adbc < ad);
+  uint64_t adbc_carry = (uint64_t)(adbc < ad);
   uint64_t lo = bd + (adbc << 32);
   *hi = emulu((uint32_t)(ab >> 32), (uint32_t)(cd >> 32)) + (adbc >> 32) +
-        (adbc_carry << 32) + !!(lo < bd);
+        (adbc_carry << 32) + (uint64_t)(lo < bd);
   return lo;
 }
 
@@ -674,6 +688,69 @@ constexpr char32_t const * str_const_inf<char32_t>()
 {
     return U"infinity";
 }
+
+
+template <typename = void>
+struct int_luts {
+  static constexpr uint8_t chdigit[] = {
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 255, 255, 255, 255, 255, 255,
+    255, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24,
+    25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 255, 255, 255, 255, 255,
+    255, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24,
+    25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
+    255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255
+  };
+
+  static constexpr size_t maxdigits_u64[] = {
+    64, 41, 32, 28, 25, 23, 22, 21,
+    20, 19, 18, 18, 17, 17, 16, 16,
+    16, 16, 15, 15, 15, 15, 14, 14,
+    14, 14, 14, 14, 14, 13, 13, 13,
+    13, 13, 13
+  };
+
+  static constexpr uint64_t min_safe_u64[] = {
+    9223372036854775808ull, 12157665459056928801ull, 4611686018427387904, 7450580596923828125, 4738381338321616896,
+    3909821048582988049, 9223372036854775808ull, 12157665459056928801ull, 10000000000000000000ull, 5559917313492231481,
+    2218611106740436992, 8650415919381337933, 2177953337809371136, 6568408355712890625, 1152921504606846976, 
+    2862423051509815793, 6746640616477458432, 15181127029874798299ull, 1638400000000000000, 3243919932521508681,
+    6221821273427820544, 11592836324538749809ull, 876488338465357824, 1490116119384765625, 2481152873203736576,
+    4052555153018976267, 6502111422497947648, 10260628712958602189ull, 15943230000000000000ull, 787662783788549761,
+    1152921504606846976, 1667889514952984961, 2386420683693101056, 3379220508056640625, 4738381338321616896
+  };
+};
+
+template <typename T>
+constexpr uint8_t int_luts<T>::chdigit[];
+
+template <typename T>
+constexpr size_t int_luts<T>::maxdigits_u64[];
+
+template <typename T>
+constexpr uint64_t int_luts<T>::min_safe_u64[];
+
+template <typename UC>
+fastfloat_really_inline
+constexpr uint8_t ch_to_digit(UC c) { return int_luts<>::chdigit[static_cast<unsigned char>(c)]; }
+
+fastfloat_really_inline
+constexpr size_t max_digits_u64(int base) { return int_luts<>::maxdigits_u64[base - 2]; }
+
+// If a u64 is exactly max_digits_u64() in length, this is
+// the value below which it has definitely overflowed. 
+fastfloat_really_inline
+constexpr uint64_t min_safe_u64(int base) { return int_luts<>::min_safe_u64[base - 2]; }
+
 } // namespace fast_float
 
 #endif