Sobhan Mohammadpour

anyhow point is, sparse for already sparse matrices works, if you need SparseMatrixCSC, just call SparseMatrixCSC on the thing.

I think you don't even need to do that, see https://github.com/JuliaSparse/SparseArrays.jl/blob/d938744fcd2d3a879a7299530392400ae4feca27/src/sparsematrix.jl#L734, calling SparseMatrixCSC, should just "work".

@ViralBShah what do you think about implementing the same for all special matrices (symmetric, transpose, diagonal, etc)? this could give rise to fast generic implementations for many of the functions...

i made https://github.com/SobhanMP/SparseExtra.jl for now, guess i'll add a bunch of other stuff while i'm at it...

note that it's not faster than the last function (the pattern used almost everywhere in Sparsearrays). accessing via the getcolptr is faster than naming a variable (i don't know why)....

@ViralBShah i've updated the readme of my repo, feel free to check it out

@ViralBShah should this stay in SparseExtra.jl (and close this) or should i make this a pull request (in contrast to a draft)? this would make closing https://github.com/JuliaSparse/SparseArrays.jl/issues/23, https://github.com/JuliaSparse/SparseArrays.jl/issues/83, and possibly...

in term of API it's done (at least the outward facing part) but there are three things worth considering. - skip_col/skip_row_to skip to the next element - 1 so that...

@fredrikekre @KristofferC @dkarrasch Any comments on the API in https://sobhanmp.github.io/SparseExtra.jl/iternz/, which we are discussing bringing into SparseArrays.jl?

@ViralBShah @dkarrasch @fredrikekre @KristofferC ping