diff --git a/services/website_collector.py b/services/website_collector.py index b1e84fb..2fd9ae7 100644 --- a/services/website_collector.py +++ b/services/website_collector.py @@ -18,7 +18,7 @@ logger = logging.getLogger(__name__) def _clean_html(raw_html: str) -> str: if not raw_html: return "" - text = re.sub(r'<[^>]+>', ' ', raw_html) + text = re.sub(r'<[^>]+>', '', raw_html) text = html.unescape(text) return re.sub(r'\s+', ' ', text).strip()