Clone the repository

git clone https://github.com/FJLendinez/fastrack.git
cd fastrack
cp .env_example .env

Edit your .env file with custom settings

Just run:

docker-compose up -d
docker exec -it fastrack_api python manage.py migrate

install dependencies

pip install -r requirements.txt
python manage.py migrate

and run

uvicorn main:app --reload

Use /a.js endpoint to have the client functionality

To can track on every web framework with ease, fastrack track automatically using

window.addEventListener('beforeunload', window.fastrack_trackview);

When beforeunload event is triggered, you can use fastrack_trackview() function to track the current page and time spent on it.

var first_load = true
router.beforeEach((from, to, next) => {
    if (first_load) {
        first_load = false
    } else {
        window.fastrack_trackview()
    }
    return next() 
})

Thanks to router guards we can control how to track, but it always launch beforeEach guard at the begining so we need to flag the first call.

To assign collected info to an email, we can use the function fastrack_identify(email) when he login, add the email in a form or something to relate this email with fastrack tracker variables.

When fastrack_trackview() is called, it will do a request to track some variables automatically but you can add custom data for your bussiness logic using window.fastrack_metadata = {"myobject": "for view A", "is": ["a", "standard", "json"]}

Here we have 4 ways to deep into:

• Use directly the postgres DB with tools like Metabase to accesss and filter
• Use python manage.py shell and query with Django ORM
• Use /track?email=<email to get tracking> to get his navigation history (to use it we must define a SECURITY KEY)
• Use /analytics endpoint

This endpoint require this query params:

• operations a comma separated string to define what operations we need (admited values: max, min, avg, count)

  Example: ?operations=max,count

• operation_value a string to define the database field of the operation

  Example: ?operations=max&operation_value=time_spent (max of time spent)

• groupby a string to define the database field to group the operation

  Example: ?operations=avg&operation_value=time_spent&groupby=history_uuid (average of time_spent per history_uuid)

• database field you can aditionally add filter for each database field

  Example: ?operations=avg&operation_value=time_spent&groupby=history_uuid&url=/important_page (average of time_spent per history_uuid on /important_page)

• database field with lookup you can aditionally add lookups for each database field Admited lookups field

  Example: ?operations=avg&operation_value=time_spent&groupby=history_uuid&url__icontains=/blog/ (average of time_spent per history_uuid on all pages that contains /blog/)

• field lookup for metadata we can use lookups to filter metadata

  Example: Supose that we track on metadata {"info": "important"} and {"info": "meh"} on our blog and {"potatoes": "are great"} on another route We can use ?operations=avg&operation_value=time_spent&groupby=history_uuid&metadata__info=important to get the average time spent per history_uuid on pages that have {"info": "important"} on it metadata

In the environment file (.env) add:

PRIVATE_ACCESS_KEYS=mykeys,areseparated,bycommas

and then use one of the defined keys as header x-access-key

curl -X GET "<fastrackendpoint>/track?email=myemail%40email.com" -H  "accept: application/json" -H  "x-access-key: mykeys"