The Tracker Python API Client provides a simple Python interface for the Tracker GraphQL API, with the aim of allowing users to easily integrate data from Tracker into existing workflows and platforms. It allows access to the JSON data served by the API without requiring specific knowledge of GraphQL or the Tracker API. This is done by providing an object-oriented interface to execute canned queries against the API using gql. Responses are formatted to remove pagination related structures, and to ensure useful keys are always present.
The client will soon be available to install as a package via pip or pipenv. Until then, follow the instructions for developers below.
Install pipenv if you don't already have it. The following instructions assume you are using pipenv.
Make sure you have pulled the most recent version from the repo, then run:
pipenv install --devYou must have a Tracker account that is a member of one or more organizations to make use of the Python client. You can manage your account in the Tracker web interface.
The client will attempt to draw credentials from its environment in order to obtain an authentication token. Pipenv makes this easy to set up by importing environment variables from a .env file present in this directory whenever pipenv run or pipenv shell are used. The .env file can be created like so:
cat <<'EOF' > .env
[email protected]
TRACKER_PASS=YOURPASSWORDHERE
EOFYou should be mindful that setting these variables manually can result in credentials being stored in your shell command history.
Start by importing the Client class (see below for examples). Instantiating Client will connect to Tracker and authenticate with your credentials, at which point you can begin getting Domain and Organization objects using that Client instance's methods. You can then operate on these objects to get the data you need from Tracker.
Note the use of a list comprehension inside the call to print. Calling print on a container uses contained objects' __repr__ rather than __str__, so the list comprehension simply serves to make the output more human-friendly.
>>> from tracker_client.client import Client
>>> client = Client()
>>> domain_list = client.get_domains()
>>> print([str(domain) for domain in domain_list])
["foo.bar", "foo.bar.baz", "fizz.buzz", "buzz.bang", "ab.cd.ef"]The following examples assume the package has been imported as above and a Client has been instantiated as client.
>>> foobar = client.get_domain("foo.bar")
>>> print(foobar.get_dmarc_summary("september", 2020))
{
"foo.bar": {
"month": "SEPTEMBER",
"year": "2020",
"categoryPercentages": {
"fullPassPercentage": 87,
"passSpfOnlyPercentage": 0,
"passDkimOnlyPercentage": 6,
"failPercentage": 8,
"totalMessages": 10534
}
}
}>>> foo = client.get_organization("Foo Bar")
>>> print(foo.get_summary())
{
"FOO": {
"domainCount": 10,
"summaries": {
"web": {
"total": 10,
"categories": [
{
"name": "pass",
"count": 1,
"percentage": 10
},
{
"name": "fail",
"count": 9,
"percentage": 90
}
]
},
"mail": {
"total": 10,
"categories": [
{
"name": "pass",
"count": 5,
"percentage": 50
},
{
"name": "fail",
"count": 5,
"percentage": 50
}
]
}
}
}
}>>> foobar = client.get_domain("foo.bar")
>>> print(foobar.get_domain_status())
{
"foo.bar": {
"lastRan": "2021-01-23 22:33:26.921529",
"status": {
"https": "FAIL",
"ssl": "FAIL",
"dmarc": "PASS",
"dkim": "PASS",
"spf": "PASS"
}
}
}Supposing all your domains, except "foo.bar", properly implement DMARC:
>>> import json
>>> client = Client()
>>> domain_list = client.get_domains()
>>> dmarc_fails = []
>>> for domain in domain_list:
... status = json.loads(domain.get_status())
... if status[domain.domain_name]["status"]["dmarc"] == "FAIL":
... dmarc_fails.append(domain.domain_name)
...
>>> print(dmarc_fails)
["foo.bar"]NOTE: Because of gql limitations, the client is not currently compatible with IPython or Jupyter.
The client comes with documentation generated by Sphinx from the docstrings in the source code. Because of this, they are not included in the Github repo. To build the docs you will first need to install Sphinx.
The simplest way to ensure you have Sphinx installed is to install the client's dev dependencies with pipenv:
pipenv install --devIf you don't wish to use pipenv, see the Sphinx installation instructions.
Once Sphinx is installed, navigate to the /docs directory located in the Python client project root and run:
pipenv run make htmlOr if you installed Sphinx outside pipenv, just run make html.
You will find the built docs in /build/html within the docs directory. Sphinx can also build to other formats! See here for the available options.
Pytest is used for testing. To run tests, run the following in the project root (the folder containing this README.md):
pipenv run pytestAlternatively, if you are already in a pipenv shell, just run pytest.
If tests are failing with ModuleNotFoundError, make sure tracker_client/ is on your PYTHONPATH. The .env file used to store your credentials is a good way to set this.
When additions or significant changes are made, check test coverage with:
pipenv run pytest --cov=tracker_client