/usr/include/gamera/boost_limits.hpp is in python-gamera-dev 1:3.4.2+git20160808.1725654-2.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 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 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 | /*
* Copyright (c) 1997
* Silicon Graphics Computer Systems, Inc.
*
* Permission to use, copy, modify, distribute and sell this software
* and its documentation for any purpose is hereby granted without fee,
* provided that the above copyright notice appear in all copies and
* that both that copyright notice and this permission notice appear
* in supporting documentation. Silicon Graphics makes no
* representations about the suitability of this software for any
* purpose. It is provided "as is" without express or implied warranty.
*/
/* NOTE: This is not portable code. Parts of numeric_limits<> are
* inherently machine-dependent, and this file is written for the MIPS
* architecture and the SGI MIPSpro C++ compiler. Parts of it (in
* particular, some of the characteristics of floating-point types)
* are almost certainly incorrect for any other platform.
*/
/* The above comment is almost certainly out of date. This file works
* on systems other than SGI MIPSpro C++ now.
*/
/*
* Revision history:
* 21 Sep 2001:
* Only include <cwchar> if BOOST_NO_CWCHAR is defined. (Darin Adler)
* 10 Aug 2001:
* Added MIPS (big endian) to the big endian family. (Jens Maurer)
* 13 Apr 2001:
* Added powerpc to the big endian family. (Jeremy Siek)
* 5 Apr 2001:
* Added sparc (big endian) processor support (John Maddock).
* Initial sub:
* Modified by Jens Maurer for gcc 2.95 on x86.
*/
#ifndef BOOST_SGI_CPP_LIMITS
#define BOOST_SGI_CPP_LIMITS
#include <climits>
#include <cfloat>
//#include <boost/config.hpp>
//#ifndef BOOST_NO_CWCHAR
//#include <cwchar> // for WCHAR_MIN and WCHAR_MAX
//#endif
#if defined(__sparc) || defined(__sparc__) || defined(__powerpc__) || defined(__ppc__) || defined(__hppa) || defined(_MIPSEB)
#define BOOST_BIG_ENDIAN
#elif defined(__i386__)
#define BOOST_LITTLE_ENDIAN
#else
#error The file boost/detail/limits.hpp needs to be set up for your CPU type.
#endif
namespace std {
enum float_round_style {
round_indeterminate = -1,
round_toward_zero = 0,
round_to_nearest = 1,
round_toward_infinity = 2,
round_toward_neg_infinity = 3
};
enum float_denorm_style {
denorm_indeterminate = -1,
denorm_absent = 0,
denorm_present = 1
};
// The C++ standard (section 18.2.1) requires that some of the members of
// numeric_limits be static const data members that are given constant-
// initializers within the class declaration. On compilers where the
// BOOST_NO_INCLASS_MEMBER_INITIALIZATION macro is defined, it is impossible to write
// a standard-conforming numeric_limits class.
//
// There are two possible workarounds: either initialize the data
// members outside the class, or change them from data members to
// enums. Neither workaround is satisfactory: the former makes it
// impossible to use the data members in constant-expressions, and the
// latter means they have the wrong type and that it is impossible to
// take their addresses. We choose the former workaround.
#ifdef BOOST_NO_INCLASS_MEMBER_INITIALIZATION
# define BOOST_STL_DECLARE_LIMITS_MEMBER(__mem_type, __mem_name, __mem_value) \
enum { __mem_name = __mem_value }
#else /* BOOST_NO_INCLASS_MEMBER_INITIALIZATION */
# define BOOST_STL_DECLARE_LIMITS_MEMBER(__mem_type, __mem_name, __mem_value) \
static const __mem_type __mem_name = __mem_value
#endif /* BOOST_NO_INCLASS_MEMBER_INITIALIZATION */
// Deal with min/max for MinGW
#ifdef min
# undef min
#endif
#ifdef max
# undef max
#endif
// Base class for all specializations of numeric_limits.
template <class __number>
class _Numeric_limits_base {
public:
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_specialized, false);
static __number min() throw() { return __number(); }
static __number max() throw() { return __number(); }
BOOST_STL_DECLARE_LIMITS_MEMBER(int, digits, 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, digits10, 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_signed, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_integer, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_exact, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, radix, 0);
static __number epsilon() throw() { return __number(); }
static __number round_error() throw() { return __number(); }
BOOST_STL_DECLARE_LIMITS_MEMBER(int, min_exponent, 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, min_exponent10, 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, max_exponent, 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, max_exponent10, 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_infinity, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_quiet_NaN, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_signaling_NaN, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(float_denorm_style,
has_denorm,
denorm_absent);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_denorm_loss, false);
static __number infinity() throw() { return __number(); }
static __number quiet_NaN() throw() { return __number(); }
static __number signaling_NaN() throw() { return __number(); }
static __number denorm_min() throw() { return __number(); }
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_iec559, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_bounded, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_modulo, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, traps, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, tinyness_before, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(float_round_style,
round_style,
round_toward_zero);
};
// Base class for integers.
template <class _Int,
_Int __imin,
_Int __imax,
int __idigits = -1>
class _Integer_limits : public _Numeric_limits_base<_Int>
{
public:
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_specialized, true);
static _Int min() throw() { return __imin; }
static _Int max() throw() { return __imax; }
BOOST_STL_DECLARE_LIMITS_MEMBER(int,
digits,
(__idigits < 0) ? (int)(sizeof(_Int) * CHAR_BIT)
- (__imin == 0 ? 0 : 1)
: __idigits);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, digits10, (digits * 301) / 1000);
// log 2 = 0.301029995664...
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_signed, __imin != 0);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_integer, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_exact, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, radix, 2);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_bounded, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_modulo, true);
};
#if defined(BOOST_BIG_ENDIAN)
template<class Number, unsigned int Word>
struct float_helper{
static Number get_word() throw() {
// sizeof(long double) == 16
const unsigned int _S_word[4] = { Word, 0, 0, 0 };
return *reinterpret_cast<const Number*>(&_S_word);
}
};
#else
template<class Number, unsigned int Word>
struct float_helper{
static Number get_word() throw() {
// sizeof(long double) == 12, but only 10 bytes significant
const unsigned int _S_word[4] = { 0, 0, 0, Word };
return *reinterpret_cast<const Number*>(
reinterpret_cast<const char *>(&_S_word)+16-
(sizeof(Number) == 12 ? 10 : sizeof(Number)));
}
};
#endif
// Base class for floating-point numbers.
template <class __number,
int __Digits, int __Digits10,
int __MinExp, int __MaxExp,
int __MinExp10, int __MaxExp10,
unsigned int __InfinityWord,
unsigned int __QNaNWord, unsigned int __SNaNWord,
bool __IsIEC559,
float_round_style __RoundStyle>
class _Floating_limits : public _Numeric_limits_base<__number>
{
public:
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_specialized, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, digits, __Digits);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, digits10, __Digits10);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_signed, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, radix, 2);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, min_exponent, __MinExp);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, max_exponent, __MaxExp);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, min_exponent10, __MinExp10);
BOOST_STL_DECLARE_LIMITS_MEMBER(int, max_exponent10, __MaxExp10);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_infinity, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_quiet_NaN, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_signaling_NaN, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(float_denorm_style,
has_denorm,
denorm_indeterminate);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, has_denorm_loss, false);
static __number infinity() throw() {
return float_helper<__number, __InfinityWord>::get_word();
}
static __number quiet_NaN() throw() {
return float_helper<__number,__QNaNWord>::get_word();
}
static __number signaling_NaN() throw() {
return float_helper<__number,__SNaNWord>::get_word();
}
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_iec559, __IsIEC559);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, is_bounded, true);
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, traps, false /* was: true */ );
BOOST_STL_DECLARE_LIMITS_MEMBER(bool, tinyness_before, false);
BOOST_STL_DECLARE_LIMITS_MEMBER(float_round_style, round_style, __RoundStyle);
};
// Class numeric_limits
// The unspecialized class.
template<class T>
class numeric_limits : public _Numeric_limits_base<T> {};
// Specializations for all built-in integral types.
template<>
class numeric_limits<bool>
: public _Integer_limits<bool, false, true, 0>
{};
template<>
class numeric_limits<char>
: public _Integer_limits<char, CHAR_MIN, CHAR_MAX>
{};
template<>
class numeric_limits<signed char>
: public _Integer_limits<signed char, SCHAR_MIN, SCHAR_MAX>
{};
template<>
class numeric_limits<unsigned char>
: public _Integer_limits<unsigned char, 0, UCHAR_MAX>
{};
#ifndef BOOST_NO_INTRINSIC_WCHAR_T
template<>
class numeric_limits<wchar_t>
#if !defined(WCHAR_MAX) || !defined(WCHAR_MIN)
#if defined(_WIN32) || defined(__CYGWIN__)
: public _Integer_limits<wchar_t, 0, USHRT_MAX>
#elif defined(__hppa)
// wchar_t has "unsigned int" as the underlying type
: public _Integer_limits<wchar_t, 0, UINT_MAX>
#else
// assume that wchar_t has "int" as the underlying type
: public _Integer_limits<wchar_t, INT_MIN, INT_MAX>
#endif
#else
// we have WCHAR_MIN and WCHAR_MAX defined, so use it
: public _Integer_limits<wchar_t, WCHAR_MIN, WCHAR_MAX>
#endif
{};
#endif
template<>
class numeric_limits<short>
: public _Integer_limits<short, SHRT_MIN, SHRT_MAX>
{};
template<>
class numeric_limits<unsigned short>
: public _Integer_limits<unsigned short, 0, USHRT_MAX>
{};
template<>
class numeric_limits<int>
: public _Integer_limits<int, INT_MIN, INT_MAX>
{};
template<>
class numeric_limits<unsigned int>
: public _Integer_limits<unsigned int, 0, UINT_MAX>
{};
template<>
class numeric_limits<long>
: public _Integer_limits<long, LONG_MIN, LONG_MAX>
{};
template<>
class numeric_limits<unsigned long>
: public _Integer_limits<unsigned long, 0, ULONG_MAX>
{};
#ifdef __GNUC__
// Some compilers have long long, but don't define the
// LONGLONG_MIN and LONGLONG_MAX macros in limits.h. This
// assumes that long long is 64 bits.
#if !defined(LONGLONG_MAX) && !defined(ULONGLONG_MAX)
# define ULONGLONG_MAX 0xffffffffffffffffLLU
# define LONGLONG_MAX 0x7fffffffffffffffLL
#endif
#if !defined(LONGLONG_MIN)
# define LONGLONG_MIN (-LONGLONG_MAX - 1)
#endif
template<>
class numeric_limits<long long>
: public _Integer_limits<long long, LONGLONG_MIN, LONGLONG_MAX>
{};
template<>
class numeric_limits<unsigned long long>
: public _Integer_limits<unsigned long long, 0, ULONGLONG_MAX>
{};
#endif /* __GNUC__ */
// Specializations for all built-in floating-point type.
template<> class numeric_limits<float>
: public _Floating_limits<float,
FLT_MANT_DIG, // Binary digits of precision
FLT_DIG, // Decimal digits of precision
FLT_MIN_EXP, // Minimum exponent
FLT_MAX_EXP, // Maximum exponent
FLT_MIN_10_EXP, // Minimum base 10 exponent
FLT_MAX_10_EXP, // Maximum base 10 exponent
#if defined(BOOST_BIG_ENDIAN)
0x7f80 << (sizeof(int)*CHAR_BIT-16), // Last word of +infinity
0x7f81 << (sizeof(int)*CHAR_BIT-16), // Last word of quiet NaN
0x7fc1 << (sizeof(int)*CHAR_BIT-16), // Last word of signaling NaN
#else
0x7f800000u, // Last word of +infinity
0x7f810000u, // Last word of quiet NaN
0x7fc10000u, // Last word of signaling NaN
#endif
true, // conforms to iec559
round_to_nearest>
{
public:
static float min() throw() { return FLT_MIN; }
static float denorm_min() throw() { return FLT_MIN; }
static float max() throw() { return FLT_MAX; }
static float epsilon() throw() { return FLT_EPSILON; }
static float round_error() throw() { return 0.5f; } // Units: ulps.
};
template<> class numeric_limits<double>
: public _Floating_limits<double,
DBL_MANT_DIG, // Binary digits of precision
DBL_DIG, // Decimal digits of precision
DBL_MIN_EXP, // Minimum exponent
DBL_MAX_EXP, // Maximum exponent
DBL_MIN_10_EXP, // Minimum base 10 exponent
DBL_MAX_10_EXP, // Maximum base 10 exponent
#if defined(BOOST_BIG_ENDIAN)
0x7ff0 << (sizeof(int)*CHAR_BIT-16), // Last word of +infinity
0x7ff1 << (sizeof(int)*CHAR_BIT-16), // Last word of quiet NaN
0x7ff9 << (sizeof(int)*CHAR_BIT-16), // Last word of signaling NaN
#else
0x7ff00000u, // Last word of +infinity
0x7ff10000u, // Last word of quiet NaN
0x7ff90000u, // Last word of signaling NaN
#endif
true, // conforms to iec559
round_to_nearest>
{
public:
static double min() throw() { return DBL_MIN; }
static double denorm_min() throw() { return DBL_MIN; }
static double max() throw() { return DBL_MAX; }
static double epsilon() throw() { return DBL_EPSILON; }
static double round_error() throw() { return 0.5; } // Units: ulps.
};
template<> class numeric_limits<long double>
: public _Floating_limits<long double,
LDBL_MANT_DIG, // Binary digits of precision
LDBL_DIG, // Decimal digits of precision
LDBL_MIN_EXP, // Minimum exponent
LDBL_MAX_EXP, // Maximum exponent
LDBL_MIN_10_EXP,// Minimum base 10 exponent
LDBL_MAX_10_EXP,// Maximum base 10 exponent
#if defined(BOOST_BIG_ENDIAN)
0x7ff0 << (sizeof(int)*CHAR_BIT-16), // Last word of +infinity
0x7ff1 << (sizeof(int)*CHAR_BIT-16), // Last word of quiet NaN
0x7ff9 << (sizeof(int)*CHAR_BIT-16), // Last word of signaling NaN
#else
0x7fff8000u, // Last word of +infinity
0x7fffc000u, // Last word of quiet NaN
0x7fff9000u, // Last word of signaling NaN
#endif
false, // Doesn't conform to iec559
round_to_nearest>
{
public:
static long double min() throw() { return LDBL_MIN; }
static long double denorm_min() throw() { return LDBL_MIN; }
static long double max() throw() { return LDBL_MAX; }
static long double epsilon() throw() { return LDBL_EPSILON; }
static long double round_error() throw() { return 4; } // Units: ulps.
};
} // namespace std
#endif /* BOOST_SGI_CPP_LIMITS */
// Local Variables:
// mode:C++
// End:
|