How would I set a generalized condition that can be passed in to index a dataframe without explicitly defining the dataframe name

General Usage
1

Here is what I have now:

df = df[df[Symbol(“PURCHASE”)].==“SUCCESS”, :]

what I’d like to do is set a success condition variable and have something that works like the code below:

success_condition = Symbol(“PURCHASE”)].==“SUCCESS”

df = df[df[success_condition]]

The key is that it should be able to be passed into any dataframe, not just one named df. I need it generalized if possible.

Thank you for your time and help!

2

First, please remember to quote your code with backticks.

Strictly speaking I’m not sure you can do what you want. The Symbol("PURCHASE") only refers to a column of a DataFrame when it is actually passed in to one as df[!,Symbol("PURCHASE")]. Outside of that it will just look like :PURCHASE, which is just a variable of type Symbol.

What you could do is to define a function like so

findsuccess(v) = findall( x -> x == "SUCCESS",v)

and then pass to it the DataFrame column findsuccess(df[!,Symbol("PURCHASE")])

1 Like
3

An easy way is to use Query.jl:

func = x -> x.PURCHASE == "SUCCESS"
df |> @filter(func(_))
4

Or just filter(x -> x.PURCHASE == "SUCCESS", df). You can store x -> x.PURCHASE in a variable to apply it to any data frame.