feat(google_trends.py): add google trends data loader

Closes #72

backend-python/media_impact_monitor/data_loaders/web/google_trends.py

@@ -0,0 +1,28 @@
+"""
+Documentation of the Google Trends API:
+- https://github.com/GeneralMills/pytrends
+- https://searchanalysisguide.blogspot.com/2013/04/google-trends-what-is-partial-data.html
+
+There is also regional data, useful for synthetic control.
+For the last 90 days, data is also available with daily resolution; otherwise only weekly.
+"""
+
+from time import sleep
+
+from media_impact_monitor.util.cache import cache
+from pytrends.request import TrendReq
+
+
+@cache
+def get_google_trends_counts(query: str):
+    PyTrends = TrendReq(hl="de-DE", tz=60)
+    PyTrends.build_payload([query], timeframe="today 5-y", geo="DE")
+    df = PyTrends.interest_over_time()
+    df = (
+        df[~df["isPartial"]]
+        .drop(columns=["isPartial"])
+        .rename(columns={query: "count"})
+    )
+    # when rate limit is reached, this should be 60 seconds according to https://github.com/GeneralMills/pytrends
+    sleep(1)
+    return df

backend-python/media_impact_monitor/data_loaders/web/google_trends_test.py

@@ -0,0 +1,15 @@
+from media_impact_monitor.data_loaders.web.google_trends import get_google_trends_counts
+
+
+def test_get_google_trends_counts():
+    df = get_google_trends_counts("corona")
+    assert not df.empty
+    assert df.columns == ["count"]
+    assert df.index.name == "date"
+    assert df.index.is_monotonic_increasing
+    assert df["count"].dtype == int
+    assert df["count"].min() >= 0
+    assert df["count"].max() >= 0
+    assert df["count"].max() == 100
+    assert df["count"].sum() >= 0
+    assert df["count"].sum() <= 100 * len(df)