Help refactoring deprecated Task() to Channels

New to Julia
4

Sure enough, simple and works.
I’m gonna work it a bit. In go I’d typically have a goroutine per file or per batch of files. I imagine i’d have to rework it quite a bit if I wanted to run a channel on multiple cpu to large sets of files, per the discourse post here: reading-and-processing-data-files-concurrently

Thank you again.

I did noticed if I put the Channel() do block under scope of the for (root, dirs, files) in walkdir(path) block i get an error.

ERROR: MethodError: no method matching start(::Void)
Closest candidates are:
  start(::SimpleVector) at essentials.jl:258
  start(::Base.MethodList) at reflection.jl:560
  start(::ExponentialBackOff) at error.jl:107

Here is the refactored bit:

function readFile(path::String)
    Channel() do chan
        for (root, dirs, files) in walkdir(path)
            for filename in files 
                if !(filename in SKIPFILES) && filename[1] != '.'
                    fullname = joinpath(root, filename)
                    if isfile(fullname)
                        pastHeader, lines = false, Vector{String}()
                        open(fullname) do f 
                            for line in eachline(f)
                                if !isvalid(line)
                                    line = decode(convert(Array{UInt8,1}, line), "LATIN1")
                                end
                                line = chomp(line)
                                if pastHeader
                                    push!(lines, line)
                                elseif endof(line) == 0
                                    pastHeader = true
                                end
                            end
                        end
                        content = join(lines, NEWLINE)
                        put!(chan, (fullname,content))
                    end
                end
            end
        end
    end
end

function addData!(df::DataFrame, path::String, classification::String)
    for (filename, text) in readFile(path)
        push!(df, @data([text, classification, filename]))
    end
end

function buildDataSet(sources)
    df = DataFrame(text = Vector{String}(), class = Vector{String}(), index = Vector{String}())
    for (path, classification) in sources
        addData!(df, joinpath(SPAMROOT, path), classification)
    end
    return df
end