langchain/libs/partners/exa/tests/integration_tests/test_retriever.py
Ishan Goswami f16456139b
exa docs and python package update (#31307)
Added support for new Exa API features. Updated Exa docs and python
package (langchain-exa).

Description

Added support for new Exa API features in the langchain-exa package:
- Added max_characters option for text content
- Added support for summary and custom summary prompts
- Added livecrawl option with "always", "fallback", "never" settings
- Added "auto" option for search type
- Updated documentation and tests

Dependencies
- No new dependencies required. Using existing features from exa-py.

twitter: @theishangoswami

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-05-21 21:33:30 -04:00

45 lines
1.5 KiB
Python

from langchain_core.documents import (
Document, # type: ignore[import-not-found, import-not-found]
)
from langchain_exa import ExaSearchRetriever
def test_exa_retriever() -> None:
retriever = ExaSearchRetriever()
res = retriever.invoke("best time to visit japan")
print(res) # noqa: T201
assert len(res) == 10 # default k
assert isinstance(res, list)
assert isinstance(res[0], Document)
def test_exa_retriever_highlights() -> None:
retriever = ExaSearchRetriever(highlights=True)
res = retriever.invoke("best time to visit japan")
print(res) # noqa: T201
assert isinstance(res, list)
assert isinstance(res[0], Document)
highlights = res[0].metadata["highlights"]
highlight_scores = res[0].metadata["highlight_scores"]
assert isinstance(highlights, list)
assert isinstance(highlight_scores, list)
assert isinstance(highlights[0], str)
assert isinstance(highlight_scores[0], float)
def test_exa_retriever_advanced_features() -> None:
retriever = ExaSearchRetriever(
k=3, text_contents_options={"max_characters": 1000}, summary=True, type="auto"
)
res = retriever.invoke("best time to visit japan")
print(res) # noqa: T201
assert len(res) == 3 # requested k=3
assert isinstance(res, list)
assert isinstance(res[0], Document)
# Verify summary is in metadata
assert "summary" in res[0].metadata
assert isinstance(res[0].metadata["summary"], str)
# Verify text was limited
assert len(res[0].page_content) <= 1000