deprecate rand(::Tuple) #25429
deprecate rand(::Tuple) #25429
Conversation
rfourquet
added
domain:randomness
rfourquet
force-pushed
the
rf/rand/depr-tuple
branch
2 times, most recently
from
1d8e045
to
8561baf
Compare
base/deprecated.jl
Outdated
| @@ -1138,6 +1138,10 @@ import .Random: srand | |||
| @deprecate srand(filename::AbstractString, n::Integer=4) srand(read!(filename, Vector{UInt32}(uninitialized, Int(n)))) | |||
| @deprecate MersenneTwister(filename::AbstractString) srand(MersenneTwister(0), read!(filename, Vector{UInt32}(uninitialized, Int(4)))) | |||
|
|
|||
| # PR #25429 | |||
| @deprecate rand(r::AbstractRNG, dims::Dims) rand(r, Float64, dims) | |||
There was a problem hiding this comment.
Does this one really need to be deprecated? I thought a dims tuple as the second argument was ok (and there are still some other methods with that).
There was a problem hiding this comment.
in rand([rng], [S], [dims]), the conflict is when using t::Tuple either for S or for dims, whether rng is specified or not. In rand(rng, (1, 2, 3)), the conflict is the same as in rand((1, 2, 3)). A tuple as the second argument is ok when rng is not specifed, as in rand(Float64, (1, 2, 3)).
and there are still some other methods with that
I must have missed that, but no test catched this apparently...
There was a problem hiding this comment.
I see. In that case this seems fine as-is.
rfourquet
force-pushed
the
rf/rand/depr-tuple
branch
from
95c64fd
to
c4aba4d
Compare
|
The Appveyor failure seems to be a timeout (the job lasted 2 hours), and the travis failure is a problem with |
Pre-requisite for #25278.