Skip to content

Cape Python API with Coordinator#

This document describes how to use Cape Python with Cape Coordinator. It builds on the Cape Python quickstart guide and the Cape Coordinator installation guides.

Quickstart#

Prerequisites#

  • Cape Coordinator installed. You need the password and email that you set during Cape Coordinator installation.
  • Cape Python installed.

We recommend working through the Cape Python quickstart guide and reading the Cape Coordinator documentation before following the steps below.

Create your first project#

Authenticate with Cape Coordinator by running:

$ CAPE_PASSWORD=<PASSWORD> cape login --email <EMAIL>

Replace <PASSWORD> and <EMAIL> with the values set during Cape Coordinator installation.

Cape Coordinator uses projects to manage which version of a policy is applied to a dataset. To create your first project, run:

$ cape projects create first-project "Hello Project World"

You can now add a policy to the project from a policy specification file.

Copy the following YAML to a file called first-policy.yaml:

version: 1
rules:
# Set the column name
- match:
    name: weight
    actions:
    - transform:
        # This example shows an unnamed transformation.
        # It tells the policy runner to:
        # (1) Apply the transformation numeric-rounding
        # (2) Round to one decimal place
        type: numeric-rounding
        dtype: Double
        precision: 1

Update the project with the first version of policy:

$ cape projects update --from-spec first-policy.yaml first-project

Get an API token#

To connect to Cape Coordinator from Cape Python you need an API token. Obtain the token by running:

$ cape tokens create

You'll see output like:

A token for cape_user@mycape.com has been created!

Token:    <REDACTED TOKEN>

‼ Remember: Please keep the token safe and share it only over secure channels.

Copy the token from the CLI. You will need it in the next section.

Write the policy application script#

The following script is very similar to the Cape Python quickstart guide, except in this case we're connecting to Cape Coordinator to retrieve the policy.

Create a coordinator-policy.py file in your project, with the following content:

import cape_privacy as cape
import pandas as pd

# In the Coordinator installation instructions you should have run the
# Coordinator on port 8080. If not, edit the line below with the proper
# port.
client = cape.Client("http://localhost:8080")
client.login("<PASTE TOKEN HERE>")

p_dict = client.get_policy("first-project")
policy = cape.parse_policy(p_dict)

# Create a simple Pandas DataFrame
df = pd.DataFrame([114.432, 134.622, 142.984], columns=["weight"])

df = cape.apply_policy(policy, df, inplace=False)
print(df.head())

Make sure to replace <PASTE TOKEN HERE> with your API token.

Run your transformations#

In coordinator-policy.py we create a dataset programatically, so there are no further steps to load a dataset.

Run the policy application script and view the output:

$ python coordinator-policy.py