Uh oh!
There was an error while loading. Please reload this page.
[Ruby 2.5] Integer#pow accepts modulo argument for calculating modular exponentiation - #573
Conversation
| 2.pow(40).should eql 1099511627776 | ||
| end | ||
| it "overflows the answer to a bignum transparantly" do |
| context "second argument is passed" do | ||
| it "returns modulo of self raised to the given power" do | ||
| 2.pow(5, 12).should == 8 |
There was a problem hiding this comment.
It would be nice to have a few extra examples here, notably with Bignum-range values and other numeric types.
| 2.pow(62, 5843009213693952).should eql 1 | ||
| 2.pow(63, 5843009213693953).should eql 1 | ||
| 2.pow(64, 5843009213693954).should eql 1 | ||
| end |
There was a problem hiding this comment.
looks like a bug:2.pow(61, 9213693951) # => 1
but(2 ** 61) % 9213693951 # => 5432161352
There was a problem hiding this comment.
Yes indeed, could you add the example to https://bugs.ruby-lang.org/issues/14259?
There was a problem hiding this comment.
In this PR, you can then specify the right behavior (eql 147138137) and guard it with:
ruby_bug'#13669','2.5'...'2.5.1'do
...| it "returns a complex number when negative and raised to a fractional power" do | ||
| (-8).pow(1.0/3) .should be_close(Complex(1, 1.73205), TOLERANCE) | ||
| (-8).pow(Rational(1, 3)).should be_close(Complex(1, 1.73205), TOLERANCE) | ||
| end |
There was a problem hiding this comment.
Could we share the 1-argument spec with Integer#**?
Or is it difficult?
There was a problem hiding this comment.
Planned to do this in a separate PR, but OK, will make it here
| describe "Integer#**" do | ||
| it_behaves_like :integer_exponent, :** | ||
| end |
There was a problem hiding this comment.
Currently there is Fixnum#** and Bignum#** specs, but it sounds best to follow the 2.4 unified Integer structure.
I filed an issue to address this in general: #578.
eregon
commented
Jan 2, 2018
This looks great, thank you for the new specs! |
Share some cases with Integer#**
Based on specs for Fixnum#** and https://github.com/ruby/ruby/blob/trunk/test/ruby/test_numeric.rb#L388
https://bugs.ruby-lang.org/issues/11003, https://bugs.ruby-lang.org/issues/12508