from pydantic_settings import BaseSettings class ScraperConfig(BaseSettings): prothom_alo_suffixes: list[str] = ["opinion","politics","bangladesh/crime",'world','business','sports','entertainment','chakri','lifestyle'] prothom_alo_baseurl: str = 'https://www.prothomalo.com/' naya_diganta_baseurl: str = 'https://dailynayadiganta.com/archive' masud_rana_series_url: str = "https://www.ebanglalibrary.com/series/%E0%A6%AE%E0%A6%BE%E0%A6%B8%E0%A7%81%E0%A6%A6-%E0%A6%B0%E0%A6%BE%E0%A6%A8%E0%A6%BE-%E0%A6%B8%E0%A6%BF%E0%A6%B0%E0%A6%BF%E0%A6%9C/" meghe_dhaka_tara: str = "https://www.ebanglalibrary.com/books/%e0%a6%ae%e0%a7%87%e0%a6%98%e0%a7%87-%e0%a6%a2%e0%a6%be%e0%a6%95%e0%a6%be-%e0%a6%a4%e0%a6%be%e0%a6%b0%e0%a6%be-%e0%a6%b6%e0%a6%95%e0%a7%8d%e0%a6%a4%e0%a6%bf%e0%a6%aa%e0%a6%a6-%e0%a6%b0%e0%a6%be/" class DataConfig(BaseSettings): model_name: str = "noor-raghib-12/bangla-bert-base-finetuned-ner-generated_data-finetuned-ner-adversarial_data" label_list: list[str] =[ "O", "B-HYPH", "I-HYPH", "B-COMM", "B-QUES", "B-EXCLAM", "B-DARI", "B-COLL", "B-SEMICOL", ] label_to_punc: dict = { "B-COMM": ",", "B-QUES": "?", "B-EXCLAM": "!", "B-DARI":"ред", "B-COLL": ":", "B-SEMICOL": ";" } scraper_utils = ScraperConfig() data_config = DataConfig()