Spaces:
Sleeping
Sleeping
| 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 find_soup_text_with_default(soup, element: str, find_search_map: Mapping[str, str]): | |
| find_result = soup.find(element, find_search_map) | |
| if not find_result: | |
| return "" | |
| return find_result.text.strip() | |
| def parse_player_div(player_div): | |
| return { | |
| "Date/Time": player_div.find("div", {"class": "PlayerNewsPost-date"}).get("data-date"), | |
| "Name": find_soup_text_with_default(player_div, "div", {"class": "PlayerNewsPost-name"}), | |
| "Team": find_soup_text_with_default(player_div, "span", {"class": "PlayerNewsPost-team-abbr"}).upper(), | |
| "Position": find_soup_text_with_default(player_div, "span", {"class": "PlayerNewsPost-position"}).title(), | |
| "Headline": find_soup_text_with_default(player_div, "div", {"class": "PlayerNewsPost-headline"}), | |
| "Analysis": find_soup_text_with_default(player_div, "div", {"class": "PlayerNewsPost-analysis"}), | |
| } | |
| 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_div_list = soup.find_all("div", {"class": "PlayerNewsPost"}) | |
| if not player_div_list: | |
| return pd.DataFrame() | |
| parsed_player_list = [parse_player_div(d) for d in player_div_list] | |
| df = pd.DataFrame(parsed_player_list) | |
| 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) | |