transformd

Transform a dictionary to another dictionary, but keep the same shape based on a spec.

What is a spec?

It is a string (or sequence of strings) that specifies what "parts" of the dictionary should be included in a new dictionary that is returned from the transform function. A spec uses dot-notation to traverse into the dictionary. It can also use indexes if the object is a list.

Examples

from transformd import Transformer

data = {
    "library": {
        "name": "Main St Library",
        "location": {
            "street": "123 Main St",
            "city": "New York City",
            "state": "NY",
        },
        "books": [
            {
                "title": "The Grapes of Wrath",
                "author": {"first_name": "John", "last_name": "Steinbeck"},
            },
            {
                "title": "Slaughterhouse-Five",
                "author": {"first_name": "Kurt", "last_name": "Vonnegut"},
            },
        ],
    }
}

assert Transformer(data).transform(spec="library.name") == {
    "library": {
        "name": "Main St Library"
    }
}

assert Transformer(data).transform(spec=("library.name", "library.location.state")) == {
    "library": {
        "name": "Main St Library",
        "location": {
            "state": "NY"
        },
    }
}

assert Transformer(data).transform(spec=("library.books.0.title", "library.books.1")) == {
    "library": {
        "books": [
            {
                "title": "The Grapes of Wrath",
            },
            {
                "title": "Slaughterhouse-Five",
                "author": {"first_name": "Kurt", "last_name": "Vonnegut"},
            },
        ],
    }
}

Why?

I needed this functionality for Unicorn, but could not find a suitable library. After writing the code, I thought maybe it would be useful for someone else. 🤷

Run tests

rye sync && rye run test

Inspiration

• Django Templates for the dot-notation inspiration
• A lot of existing JSON-related tools, but especially glom, jello, jq, and gron; all of which did not quite do what I wanted, but were useful on the journey