Support plan_inv for ScaledPlan's
#77
Conversation
| # functionally identical plans | ||
| for P in [plan_rfft(x, dims), inv(plan_irfft(ry, size(x, dims), dims)), | ||
| AbstractFFTs.plan_inv(plan_irfft(ry, size(x, dims), dims))] | ||
| @test eltype(P) <: Real |
There was a problem hiding this comment.
Other than add the loop over P, the only other change I made was to this assertion. Previously the assertion was that the eltype should be Int here. That doesn't seem right (for FFTW the eltype would be Float64); rather than thinking too hard about what the test plan implementation should be doing (it's probably underspecified), I just made the check a little looser.
Codecov Report
@@ Coverage Diff @@
## master #77 +/- ##
==========================================
+ Coverage 83.09% 84.13% +1.04%
==========================================
Files 2 2
Lines 207 208 +1
==========================================
+ Hits 172 175 +3
+ Misses 35 33 -2
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. |
|
To clarify what's happening here, it's a one-line fix for the downstream issue. I added a few more tests to cover |
Fixes a downstream issue in FFTW's tests caused by #72