two version of R2R are hereHEADmaster

1 files changed, 15 insertions, 0 deletions

diff --git a/R2R/r2r/parsers/text/html_parser.py b/R2R/r2r/parsers/text/html_parser.py
new file mode 100755
index 00000000..9c663fbe
--- /dev/null
+++ b/R2R/r2r/parsers/text/html_parser.py
@@ -0,0 +1,15 @@
+from typing import AsyncGenerator
+
+from bs4 import BeautifulSoup
+
+from r2r.base.abstractions.document import DataType
+from r2r.base.parsers.base_parser import AsyncParser
+
+
+class HTMLParser(AsyncParser[DataType]):
+    """A parser for HTML data."""
+
+    async def ingest(self, data: DataType) -> AsyncGenerator[str, None]:
+        """Ingest HTML data and yield text."""
+        soup = BeautifulSoup(data, "html.parser")
+        yield soup.get_text()