jzmaddock
255dbc9201
Coverage: erf, exclude block known to be covered by __float128 tests.
2025-04-21 15:46:18 +01:00
jzmaddock
2600de8431
Coverage: Digamma exclude another MP-only function.
2025-04-21 15:45:30 +01:00
jzmaddock
760fca3743
Remove MP test case: our test data isn't precise enough.
...
Disable MP expm1 from coverage check.
2025-04-21 12:14:39 +01:00
jzmaddock
ce9c72d350
expint: coverage cosmetic change.
2025-04-21 09:56:43 +01:00
jzmaddock
2bd29ae5db
erf: exclude __float128 case from coverage.
2025-04-21 09:55:08 +01:00
jzmaddock
b0a9ffb231
Correct erf reflection logic (fixes CUDA change).
2025-04-21 09:52:55 +01:00
jzmaddock
56f56115f9
Remove duplicated error handling from ellint_rj.
2025-04-21 09:41:53 +01:00
jzmaddock
63d559cc9a
Digamma coverage tidy up.
2025-04-21 09:24:32 +01:00
jzmaddock
54b057bb51
Modernise log1p/expm1 to mostly use their std:: equivalents.
...
Add support for <cstdfloat> types as well.
Extend tests for better coverage.
2025-04-19 13:35:23 +01:00
jzmaddock
b0200e41d1
Correct macro name for standalone.
2025-04-18 21:02:52 +01:00
jzmaddock
00870090bd
Correct macro name for standalone usage.
2025-04-18 19:59:23 +01:00
jzmaddock
8107e55faf
Merge branch 'develop' into improve_coverage_3
...
Fixed Conflicts:
include/boost/math/special_functions/detail/bessel_ik.hpp
include/boost/math/special_functions/detail/bessel_j1.hpp
include/boost/math/special_functions/detail/bessel_k0.hpp
include/boost/math/special_functions/detail/bessel_k1.hpp
include/boost/math/special_functions/detail/igamma_large.hpp
include/boost/math/special_functions/detail/lgamma_small.hpp
include/boost/math/special_functions/expint.hpp
include/boost/math/special_functions/expm1.hpp
include/boost/math/special_functions/gamma.hpp
include/boost/math/special_functions/log1p.hpp
include/boost/math/special_functions/trigamma.hpp
test/Jamfile.v2
test/test_expint.cpp
2025-04-18 19:45:05 +01:00
jzmaddock
01dbedcb0a
Merge pull request #1260 from boostorg/issue1247
...
Prevent spurious numeric underflow in ibeta_series.
2025-04-18 11:26:51 +01:00
jzmaddock
7a41ad8f1f
Merge pull request #1181 from fsb4000/fix_warnings
...
Fix warning C4701: "potentially uninitialized local variable used"
2025-04-18 09:49:51 +01:00
jzmaddock
54a558451a
Prevent spurious numeric underflow in ibeta_series.
...
Fixes #1247 .
2025-04-17 19:51:24 +01:00
jzmaddock
deb439132f
Update MacOS error rate for new test.
...
Add better calculation for y.
2025-04-17 11:17:28 +01:00
jzmaddock
8c7ff156e4
Readjust heuristics.
2025-04-16 19:20:27 +01:00
jzmaddock
e19554c55d
Fix for inverse ibeta with large a,b.
...
In this case Temme's method may fail as the incomplete gamma can't cope, but as the function changes over from 0 to 1 very rapidly, we can just use the saddle point as a reasonable starting location for iteration.
See https://github.com/scipy/scipy/issues/21725 .
2025-04-16 17:57:20 +01:00
jzmaddock
a0fd863c1f
Make variadic functions C++26 safe.
2025-04-14 12:32:45 +01:00
jzmaddock
ddd803296f
Merge pull request #1257 from boostorg/issue1198
...
Collected fixes for incomplete beta in the asymptotic large a/b region
2025-04-14 12:17:54 +01:00
jzmaddock
a84766d160
Use correct namespace!
2025-04-13 11:43:09 +01:00
jzmaddock
c9c27c5e13
Correct changed Bessel logic.
2025-04-13 10:12:15 +01:00
John Maddock
b98153a5a2
Merge branch 'develop' into issue1198
2025-04-12 15:35:39 +01:00
ckormanyos
ad2111fccd
Merge branch 'ci_dep' into issue1255
2025-04-12 15:07:33 +02:00
jzmaddock
8e5df72a92
Add workaround for std::float64_t in large constants.
2025-04-12 12:38:54 +01:00
jzmaddock
14b2059b35
Correct beta call so it's not ambiguous.
2025-04-11 18:01:32 +01:00
ckormanyos
d07b79d377
Handle issue 1255 and update docs
2025-04-11 12:49:36 +02:00
jzmaddock
20b5692940
Use the correct comparison in bessel_ik.hpp.
2025-04-10 18:54:17 +01:00
jzmaddock
dd59c01ac8
Fixes for #1198 .
...
Use asymptotic expansion for very large a,b and add test cases.
Add some corrections to improve numeric stability.
Add better error handling of cases that explode so we get evaluation_error's.
2025-04-10 18:52:35 +01:00
jzmaddock
4d58e93b5d
Fix first test case from https://github.com/boostorg/math/issues/1198 and add test case.
2025-03-28 18:19:42 +00:00
Warren Weckesser
2fda707186
MAINT: Remove an obsolete comment in skew_normal.hpp.
2025-01-24 21:18:22 -05:00
Warren Weckesser
242d36c3c5
MAINT: Remove a comment about the number of distributions.
...
Apparently this comment is not being maintained.
2025-01-24 21:16:32 -05:00
Pavel P
dfa0bd6ee6
Fix check for clang warning -Wnan-infinity-disabled
2025-01-23 17:55:02 +02:00
Alejandro Baranek
2f3c7985f7
Update gamma.hpp ( #1236 )
...
* Update gamma.hpp
Missing name for Lanczos parameter in gamma_imp_final
* Update gamma.hpp
* Change location of void and annotate
---------
Co-authored-by: Matt Borland <matt@mattborland.com >
2025-01-21 16:15:52 -05:00
Warren Weckesser
7d6f32045a
Simplify the formula for the CDF of the Cauchy distribution.
...
The Cauchy CDF function may be expressed as atan2(1, -x)/pi.
2025-01-16 23:07:27 -05:00
Youssef LAKHAL
adaf1bfd06
Fix constexpr cmath floor of 1 returning 0 #1232
2025-01-12 15:54:47 +01:00
ckormanyos
56aadecb62
Repair typo in max_digits10
2024-12-23 14:21:34 +01:00
ckormanyos
2a7bef3410
Query cyl_bessel_k for cpp_double_double
2024-12-22 14:43:19 +01:00
Matt Borland
52b029ffa1
Merge pull request #1225 from boostorg/fix_warn
2024-12-19 15:48:25 -05:00
ckormanyos
bf7fb3782b
Remove wrong constexpr and revert bessel_j change
2024-12-19 19:41:36 +01:00
ckormanyos
214e19cc51
Apply cyl_bessel_k0 logic change 113 bits
2024-12-19 18:33:28 +01:00
Matt Borland
b101f106cc
Change minimum clang version for -Wnan-infinity-disabled
2024-12-17 06:35:24 -05:00
Matt Borland
699d79b54f
Merge pull request #1222 from pps83/develop-inf-warn2
...
Fix clang warning when compiling with -ffast-math: use of infinity is undefined behavior due to the currently enabled floating-point options
2024-12-16 16:18:07 -05:00
Alexander Grund
16f09c5bce
Suppress unused parameter warnings
2024-12-06 16:33:43 +01:00
Pavel P
289e1813d8
Fix clang warning when compiling with -ffast-math: use of infinity is undefined behavior due to the currently enabled floating-point options
...
Fixes the following warning:
```
C:\boost_1_86_0\boost/math/ccmath/isinf.hpp(29,69): warning : use of infinity is undefined behavior due to the currently enabled floating-point options [-Wnan-infinity-disabled]
C:\boost_1_86_0\boost/math/ccmath/isinf.hpp(29,25): warning : use of infinity is undefined behavior due to the currently enabled floating-point options [-Wnan-infinity-disabled]
C:\boost_1_86_0\boost/math/ccmath/ldexp.hpp(50,38): note: in instantiation of function template specialization 'boost::math::ccmath::isinf<float>' requested here
C:\boost_1_86_0\boost/math/ccmath/ldexp.hpp(69,33): note: in instantiation of function template specialization 'boost::math::ccmath::ldexp<float, true>' requested here
```
2024-11-28 15:40:10 +02:00
Matt Borland
8f843fd131
Merge branch 'master' into develop
2024-10-29 08:51:15 -04:00
Pavel P
2cd9c4d835
Fix icx warning: explicit comparison with infinity in fast floating point mode with intel compiler ( #1218 )
...
fixes #1217
2024-10-28 15:46:58 -04:00
NickelWenzel
097b73c9bf
fix: define BOOST_MATH_NO_LONG_DOUBLE_MATH_FUNCTIONS for emscripten ( #1216 )
2024-10-24 09:01:17 -04:00
Matt Borland
a6afa23eb2
Make sure we don't activate float128 use when explicitly disabled
2024-10-10 09:11:44 -04:00
Matt Borland
7bd5348abc
Remove using statements for complex functions
2024-09-18 09:59:26 -04:00
