community: If load() has been overridden, use it in default lazy_load() (#18690)

This commit is contained in:
Christophe Bornet
2024-03-07 17:52:19 +01:00
committed by GitHub
parent 6cd7607816
commit 4a7d73b39d
3 changed files with 29 additions and 4 deletions

View File

@@ -1,43 +0,0 @@
"""Test Base Schema of documents."""
from typing import Iterator
from langchain_core.documents import Document
from langchain_community.document_loaders.base import BaseBlobParser, BaseLoader
from langchain_community.document_loaders.blob_loaders import Blob
def test_base_blob_parser() -> None:
"""Verify that the eager method is hooked up to the lazy method by default."""
class MyParser(BaseBlobParser):
"""A simple parser that returns a single document."""
def lazy_parse(self, blob: Blob) -> Iterator[Document]:
"""Lazy parsing interface."""
yield Document(
page_content="foo",
)
parser = MyParser()
assert isinstance(parser.lazy_parse(Blob(data="who?")), Iterator)
# We're verifying that the eager method is hooked up to the lazy method by default.
docs = parser.parse(Blob(data="who?"))
assert len(docs) == 1
assert docs[0].page_content == "foo"
async def test_default_aload() -> None:
class FakeLoader(BaseLoader):
def lazy_load(self) -> Iterator[Document]:
yield from [
Document(page_content="foo"),
Document(page_content="bar"),
]
loader = FakeLoader()
docs = loader.load()
assert docs == [Document(page_content="foo"), Document(page_content="bar")]
assert docs == [doc async for doc in loader.alazy_load()]