unwind/tests/test_imdb.py

import bz2
from pathlib import Path
from unittest.mock import AsyncMock

import bs4
import pytest

from unwind import imdb
from unwind.imdb import imdb_rating_from_score, score_from_imdb_rating

testsdir = Path(__file__).parent
fixturesdir = testsdir / "fixtures"


@pytest.mark.parametrize("rating", (x / 10 for x in range(10, 101)))
def test_rating_conversion(rating: float):
    assert rating == imdb_rating_from_score(score_from_imdb_rating(rating))


@pytest.mark.parametrize("score", range(0, 101))
def test_score_conversion(score: int):
    # Because our score covers 101 discrete values and IMDb's rating only 91
    # discrete values, the mapping is non-injective, i.e. 10 values can't be
    # mapped uniquely.
    non_injective = set(range(5, 100, 10))
    if score in non_injective:
        pytest.skip(f"Score cannot be mapped back correctly: {score}")

    assert score == score_from_imdb_rating(imdb_rating_from_score(score))


@pytest.mark.asyncio
async def test_load_most_popular_100(monkeypatch):
    with bz2.open(fixturesdir / "most_popular_100.html.bz2", "rb") as f:
        html = f.read()
    soup = bs4.BeautifulSoup(html, "html5lib")

    monkeypatch.setattr(imdb, "asoup_from_url", AsyncMock(return_value=soup))

    movie_ids = await imdb.load_most_popular_100()
    assert len(movie_ids) == 100
    assert all(id_.startswith("tt") for id_ in movie_ids)


@pytest.mark.asyncio
async def test_load_bottom_100(monkeypatch):
    with bz2.open(fixturesdir / "bottom_100.html.bz2", "rb") as f:
        html = f.read()
    soup = bs4.BeautifulSoup(html, "html5lib")

    monkeypatch.setattr(imdb, "asoup_from_url", AsyncMock(return_value=soup))

    movie_ids = await imdb.load_bottom_100()
    assert len(movie_ids) == 100
    assert all(id_.startswith("tt") for id_ in movie_ids)


@pytest.mark.asyncio
async def test_load_top_250(monkeypatch):
    with bz2.open(fixturesdir / "top250.gql.json.bz2", "rb") as f:
        jsonstr = f.read()

    monkeypatch.setattr(imdb, "adownload", AsyncMock(return_value=jsonstr))

    movie_ids = await imdb.load_top_250()
    assert len(movie_ids) == 250
    assert all(id_.startswith("tt") for id_ in movie_ids)


@pytest.mark.asyncio
async def test_load_ratings_page(monkeypatch):
    with bz2.open(fixturesdir / "ratings-ur655321.html.bz2", "rb") as f:
        html = f.read()
    soup = bs4.BeautifulSoup(html, "html5lib")

    monkeypatch.setattr(imdb, "asoup_from_url", AsyncMock(return_value=soup))

    page = await imdb.load_ratings_page("fakeurl")
    assert len(page.ratings) == 100
    assert page.imdb_user_id is not None
    assert page.imdb_user_id == "ur655321"
    assert page.imdb_user_name == "AlexUltra"
    assert page.next_page_url is not None
    assert page.next_page_url.startswith("/user/ur655321/ratings?")