Add 24 hour player news

src/pages/10_Player_News.py

@@ -0,0 +1,50 @@
+import datetime
+import streamlit as st
+
+from config import DEFAULT_ICON
+from shared_page import common_page_config
+
+from queries.nbcsports.player_news import get_player_news_window_hours
+
+
+@st.cache_data(ttl=60 * 60 * 24)
+def load_data():
+    data = get_player_news_window_hours(24)
+    teams_list = sorted(filter(None, data.Team.unique()))
+    position_list = data.Position.unique()
+    data_load_time_str = datetime.datetime.utcnow().strftime("%m/%d/%Y %I:%M %p")
+    return data, teams_list, position_list, data_load_time_str
+
+
+def get_page():
+    page_title = "Player News - Last 24 Hours"
+    st.set_page_config(page_title=page_title, page_icon=DEFAULT_ICON, layout="wide")
+    common_page_config()
+    st.title(page_title)
+    if st.button("Refresh Data"):
+        st.cache_data.clear()
+    data, teams_list, position_list, data_load_time_str = load_data()
+    st.write(f"Data loaded as of: {data_load_time_str} UTC")
+
+    teams_selected = st.multiselect("Team:", teams_list, placeholder="Select a team to filter") or teams_list
+
+    with st.container():
+        filtered_data = data[(data.Team.isin(teams_selected))]
+        st.dataframe(
+            filtered_data,
+            hide_index=True,
+            height=35 * (len(filtered_data) + 1) + 12,
+            use_container_width=True,
+            column_order=[
+                "Date/Time",
+                "Name",
+                "Team",
+                "Position",
+                "Headline",
+            ],
+            column_config={"Date/Time": st.column_config.DatetimeColumn()},
+        )
+
+
+if __name__ == "__main__":
+    get_page()

src/queries/nbcsports/player_news.py

@@ -0,0 +1,51 @@
+from bs4 import BeautifulSoup
+import datetime
+import pandas as pd
+import requests
+from typing import Mapping
+
+NEWS_URL = "https://www.nbcsports.com/fantasy/football/player-news"
+
+
+def get_text_from_find_all(soup, element: str, find_search_map: Mapping[str, str]):
+    find_list = soup.find_all(element, find_search_map)
+    assert find_list
+    text_list = [x.text.strip() for x in find_list]
+    return text_list
+
+
+def get_nfl_player_news(page_number: int = 1) -> pd.DataFrame:
+    url = f"{NEWS_URL}?p={page_number}"
+    request_page = requests.get(url)
+    soup = BeautifulSoup(request_page.content)
+    player_names_list = get_text_from_find_all(soup, "div", {"class": "PlayerNewsPost-name"})
+    team_abbr_list = get_text_from_find_all(soup, "span", {"class": "PlayerNewsPost-team-abbr"})
+    position_list = get_text_from_find_all(soup, "span", {"class": "PlayerNewsPost-position"})
+    headline_list = get_text_from_find_all(soup, "div", {"class": "PlayerNewsPost-headline"})
+    analysis_list = get_text_from_find_all(soup, "div", {"class": "PlayerNewsPost-analysis"})
+    datetime_div_list = soup.find_all("div", {"class": "PlayerNewsPost-date"})
+    assert datetime_div_list
+    datetime_list = [x["data-date"] for x in datetime_div_list]
+    assert (
+        len(player_names_list) == len(team_abbr_list) == len(position_list) == len(headline_list) == len(analysis_list)
+    )
+    df = pd.DataFrame(
+        zip(datetime_list, player_names_list, team_abbr_list, position_list, headline_list, analysis_list),
+        columns=["Date/Time", "Name", "Team", "Position", "Headline", "Analysis"],
+    )
+    df["Date/Time"] = pd.to_datetime(df["Date/Time"])
+    return df
+
+
+def get_player_news_window_hours(hours: int = 1):
+    end_date = datetime.datetime.now(datetime.timezone.utc) - datetime.timedelta(hours=hours)
+    page = 1
+    max_pages = 20
+    date_reached = False
+    df_list = []
+    while page < max_pages and not date_reached:
+        last_news = get_nfl_player_news(page)
+        df_list.append(last_news)
+        date_reached = min(last_news["Date/Time"]) < end_date
+        page += 1
+    return pd.concat(df_list)

tests/contract/test_nbcsports_player_news.py

@@ -0,0 +1,13 @@
+import pytest
+
+from queries.nbcsports import player_news
+
+
+@pytest.mark.parametrize("page_number", [(1), (2)])
+def test_get_nfl_player_news(page_number: int):
+    _ = player_news.get_nfl_player_news(page_number)
+
+
+@pytest.mark.parametrize("hours", [(1), (10)])
+def test_get_player_news_window_hours(hours: int):
+    _ = player_news.get_player_news_window_hours(hours)