Cubic roots

[NAThompson]

No description provided.

[NAThompson]

@jzmaddock, @mborland : Could you take a look?

[NAThompson]

@jzmaddock : I can't build the docs anymore; does the CI build the docs? I'm a bit worried about the unicode in the .qbk file.

[jzmaddock]

@NAThompson : I've just killed all the CI except the most recent, that batch of pushes had created a huge backlog...

[jzmaddock]

I can't build the docs anymore; does the CI build the docs? I'm a bit worried about the unicode in the .qbk file.

Yes, the CircleCI job builds the docs.

However, a mark one eyeball over the generated docs is still a good idea. How come you can't build the docs?

[NAThompson]

@NAThompson : I've just killed all the CI except the most recent, that batch of pushes had created a huge backlog...

Bah, my bad. I still contend the default should be to kill the jobs if you force push an update!

(It's the universe that's wrong, not me.)

[NAThompson]

How come you can't build the docs.

$ ../../b2 doc
In file included from ../../boost/math/policies/policy.hpp:11:
../../boost/math/tools/mp.hpp:22:18: warning: variadic templates are a C++11 extension [-Wc++11-extensions]
template<typename... T>

I can't for the life of me convince b2 that it's allowed to use C++11.

[NAThompson]

The failing checks are the known issues in the Bernoulli numbers.

[jzmaddock]

The failing checks are the known issues in the Bernoulli numbers.

What known issues?

It's just annoying to have failures in CI as it makes it so much harder to evaluate PR's.

[NAThompson]

@jzmaddock : I grepped through the log and found this:

../../../boost/math/special_functions/bernoulli.hpp:74:24: note: in instantiation of function template specialization 'boost::math::bernoulli_b2n<boost::multiprecision::number<boost::multiprecision::backends::cpp_bin_float<50, boost::multiprecision::backends::digit_base_10, void, int, 0, 0>, boost::multiprecision::et_off>, boost::math::policies::policy<boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy, boost::math::policies::default_policy> >' requested here
1305
   return boost::math::bernoulli_b2n<T>(i, policies::policy<>());
1306
                       ^
1307
bernoulli_no_atomic_fail.cpp:14:71: note: in instantiation of function template specialization 'boost::math::bernoulli_b2n<boost::multiprecision::number<boost::multiprecision::backends::cpp_bin_float<50, boost::multiprecision::backends::digit_base_10, void, int, 0, 0>, boost::multiprecision::et_off> >' requested here
1308
   check_result<boost::multiprecision::cpp_bin_float_50>(boost::math::bernoulli_b2n<boost::multiprecision::cpp_bin_float_50>(4));
1309
                                                                      ^
1310
1 error generated.

And . . . its an expected error. So my bad; I need to submit a fixup commit.