Julia 1.0 update #4
Conversation
lostella
left a comment
lostella
left a comment
There was a problem hiding this comment.
We should simplify how adjoint operators are handled
src/calculus/HCAT.jl
Outdated
| # build mul!(y, H.A[1], b[H.idxs[1]]) | ||
| bb = :(b[H.idxs[1]]) | ||
| else | ||
| # staked operator |
lostella
left a comment
lostella
left a comment
There was a problem hiding this comment.
Very minor comments/request. Congratulations, this must have required some patience!
| return (domain...,) | ||
| end | ||
| codomainType(L::HCAT) = codomainType.(L.A[1]) | ||
| codomainType(L::HCAT) = codomainType.(Ref(L.A[1])) |
There was a problem hiding this comment.
Why this Ref? I'm just curious, I don't now how that works
There was a problem hiding this comment.
from the 0.7 release notes
Previously, broadcast defaulted to treating its arguments as scalars if they were not arrays. This behavior is deprecated, and in the future broadcast will default to iterating over all its arguments. Wrap arguments you wish to be treated as scalars with Ref() or a 1-tuple. Package developers can choose to allow a non-iterable type T to always behave as a scalar by implementing broadcastable(x::T) = Ref(x) (#26212).
I'm still digesting this though...
src/linearoperators/LMatrixOp.jl
Outdated
|
|
||
| struct LMatrixOp{T, A <: Union{AbstractVector,AbstractMatrix}, | ||
| B <:Union{RowVector,AbstractMatrix}} <: LinearOperator | ||
| B <:Union{Adjoint,AbstractMatrix}} <: LinearOperator |
There was a problem hiding this comment.
I don't get it: Adjoint <: AbstractMatrix, so this Union seems redundant.
|
Another request:
|
Codecov Report
@@ Coverage Diff @@
## master #4 +/- ##
========================================
+ Coverage 92.26% 97% +4.73%
========================================
Files 44 44
Lines 1564 1201 -363
========================================
- Hits 1443 1165 -278
+ Misses 121 36 -85
Continue to review full report at Codecov.
|
3 participants
Making compatible with Julia 0.7 and 1.0