Jspidy.jl

module Jspidy

using DataFrames

# import rawdata
function get()
	if !(isfile(pwd() * "/raw_data.csv") && int(strftime("%H", time()))-int(strftime("%H", mtime(pwd() * "/raw_data.csv"))) < 1 && time() - mtime(pwd() * "/raw_data.csv") < 3600)
		download("http://www.gw2spidy.com/api/v0.9/csv/all-items/all", pwd()*"/raw_data.csv");
	end
	df = readtable(pwd() * "/raw_data.csv");
	return df;
end

# process everything
function process{T<:DataFrame}(df::T; 
	header::Array{Symbol,1}=[:name],
	sby::Symbol=:margin,
	minsale::Int32=10000,
	minoffer::Int32=10000,
	minmargin::Int32=10,
	rev_opt::Bool=true)

#	@assert prod({in(header[i], names(df)) for i=1:length(header)});
	@assert in(sby, [names(df); :netprice; :margin; :margin_percent]);

	sub_df = df[convert(BitArray, {df[i, :sale_availability] .> minsale && df[i, :offer_availability] .> minoffer for i=1:size(df,1)}), names(df)];

	c1 = DataFrame(netprice = round(sub_df[:,:min_sale_unit_price]*0.85, 2));
	c2 = DataFrame(margin = round((c1[1] - sub_df[:, :max_offer_unit_price]),2));
	c3 = DataFrame(margin_percent = round(c2[1] ./ c1[1]*100 ,2));

	return sort!([sub_df c1 c2 c3][c2[1] .>= minmargin, :], cols=sby, rev=rev_opt)[[header; :netprice; :margin; :margin_percent]];
end

#write csv to disk
function write{T<:DataFrame}(df::T)
	writetable(*(pwd() , "/gw2spidy ", strftime("%d.%m.%Y %HUhr", TmStruct(time())), ".csv"), df);
end

end