Unexpected Read Behavior on list of JSON

[kohehir]

Describe the bug When creating a DataFrame from a list of JSON objects with key value pair data, the DataFrame ignores keys and populates the DataFrame using the order of the values passed. Below is a toy example of this behavior.

To Reproduce

json_data = [{ A: "A", B: "B", C: "C"},
             { A: "A", B: "B", C: "C"},
             { C: "C", B: "B", A: "A" },
             {A: "A", C: "C"}]

df = new dfd.DataFrame(json_data)
table(df)

This generates the below table.

Expected behavior More logical output would be the below:

[risenW]

@kohehir Reading JSON into DataFrames does not use the object key, we instead use the position. Going forward, we may add an option to use keyed objects.

[skwasha]

Unfortunately, this makes usage with JSON almost pointless. JSON by definition is not required to be ordered. From the source JSON.org An object is an unordered set of name/value pairs.

[risenW]

Unfortunately, this makes usage with JSON almost pointless. JSON by definition is not required to be ordered. From the source JSON.org An object is an unordered set of name/value pairs.

I’ve come to realize this as well. I’ve added it as a priority to the next update.

[mazenashgar]

@skwasha This issue is still happening on the toJSON function, was there a specific configuration for the toJSON call?

[skwasha]

@skwasha This issue is still happening on the toJSON function, was there a specific configuration for the toJSON call?

I think perhaps you meant to mention someone else?

[mazenashgar]

My apologies @skwasha I meant to mention @risenW