mirror of
https://github.com/hwchase17/langchain.git
synced 2025-06-21 14:18:52 +00:00
core[minor]: Add aload to document loader (#19936)
Add aload to document loader
This commit is contained in:
parent
31a641a155
commit
d293431e10
@ -251,7 +251,7 @@ class WebBaseLoader(BaseLoader):
|
|||||||
metadata = _build_metadata(soup, path)
|
metadata = _build_metadata(soup, path)
|
||||||
yield Document(page_content=text, metadata=metadata)
|
yield Document(page_content=text, metadata=metadata)
|
||||||
|
|
||||||
def aload(self) -> List[Document]:
|
def aload(self) -> List[Document]: # type: ignore
|
||||||
"""Load text from the urls in web_path async into Documents."""
|
"""Load text from the urls in web_path async into Documents."""
|
||||||
|
|
||||||
results = self.scrape_all(self.web_paths)
|
results = self.scrape_all(self.web_paths)
|
||||||
|
@ -28,6 +28,10 @@ class BaseLoader(ABC):
|
|||||||
"""Load data into Document objects."""
|
"""Load data into Document objects."""
|
||||||
return list(self.lazy_load())
|
return list(self.lazy_load())
|
||||||
|
|
||||||
|
async def aload(self) -> List[Document]:
|
||||||
|
"""Load data into Document objects."""
|
||||||
|
return [document async for document in self.alazy_load()]
|
||||||
|
|
||||||
def load_and_split(
|
def load_and_split(
|
||||||
self, text_splitter: Optional[TextSplitter] = None
|
self, text_splitter: Optional[TextSplitter] = None
|
||||||
) -> List[Document]:
|
) -> List[Document]:
|
||||||
|
@ -64,3 +64,4 @@ async def test_default_aload() -> None:
|
|||||||
docs = loader.load()
|
docs = loader.load()
|
||||||
assert docs == [Document(page_content="foo"), Document(page_content="bar")]
|
assert docs == [Document(page_content="foo"), Document(page_content="bar")]
|
||||||
assert docs == [doc async for doc in loader.alazy_load()]
|
assert docs == [doc async for doc in loader.alazy_load()]
|
||||||
|
assert docs == await loader.aload()
|
||||||
|
Loading…
Reference in New Issue
Block a user