diff --git a/crawlers/crawl_today.py b/crawlers/crawl_today.py index edc1dcf..4f3325d 100644 --- a/crawlers/crawl_today.py +++ b/crawlers/crawl_today.py @@ -179,7 +179,14 @@ class YichensTodaySpider(PaginationSpider): created_at = None m = re.search(r"(\d{4}/\d{1,2}/\d{1,2}\s+\d{1,2}:\d{2}:\d{2})", page_text) if m: - created_at = m.group(1).replace("/", "-") + date_str = m.group(1).replace("/", "-") + try: + from datetime import datetime + dt = datetime.strptime(date_str, "%Y-%m-%d %H:%M:%S") + if 2020 <= dt.year <= 2030: + created_at = date_str + except: + pass phones = re.findall(r'1[3-9]\d{9,10}', page_text) contact = ",".join(dict.fromkeys(phones[:5])) if phones else None