toddrob99/nflapi

NFL API

The nflapi library provides an interface for the NFL.com api. It essentially acts like a browser, accessing the same functionality available through nfl.com.

Installing

The library has not yet been released on PyPI, since it's still in a pretty rough state. There's a distinct risk that this might not ever change. To install, you can use pip's ability to to install directly from a git repository:

pip install -e git+git@github.com:rasher/nflapi.git#egg=nflapi

Examples

Basic boilerplate to getting going:

>>> from nflapi import NFL
>>> nfl = NFL(ua="nflapi example script")

That's it. You can now use the nfl object to query various objects off the NFL api. A couple of helpers are added to perform common queries and return the results as reasonable objects.

Getting the current season and week:

>>> week = nfl.schedule.current_week()
>>> print(("{w.current_season[default]}"
           " {w.current_season_type[default]}"
           " {w.current_week[default]}").format(w=week))
2018 PRE 2

Getting standings information:

>>> records = nfl.standings.current()
>>> for team, record in records:
...     print("{t.nick_name} {r.overall_pct:.3f}".format(t=team, r=record))
...
Cardinals 0.500

Have a look at [nflcli.py](nflapi/cli/nflcli.py) for more examples.

You can also query the API directly if you're feeling adventurous.:

>>> op = Operation(shield.Viewer)
>>> player = op.viewer.player(id='3213464f-4c05-8566-2020-85b506da0baa')
>>> player.person.display_name()
>>> nfl.query(player).viewer.player.person.display_name
'Nick Foles'

If you want to get even more wild, you can pass in a raw graphql query. In that case you'll have to deal with the raw JSON response:

>>> query = """{
viewer{
  league{
    gamesByWeek(week_seasonValue: 2020,
                week_seasonType: REG,
                week_weekValue: 2) {
      id
      gameTime
      awayTeam{
        abbreviation
        }
        homeTeam{
          abbreviation
        }
      }
    }
  }
}"""
>>> games = nfl.endpoint(query)
>>> games['data']['viewer']['league']['gamesByWeek'][0]['gameTime']
'2020-09-22T00:15:00.000Z'

TODOs

• Helpers for more object types
• A smarter approach to default field selections
• Documentation
• Tests, in a perfect world