mirror of
https://github.com/hwchase17/langchain.git
synced 2026-06-09 10:17:00 +00:00
fix(mistralai): handle HTTP errors in async embed documents (#33187)
The async embed function does not properly handle HTTP errors. For instance with large batches, Mistral AI returns `Too many inputs in request, split into more batches.` in a 400 error. This leads to a KeyError in `response.json()["data"]` l.288 This PR fixes the issue by: - calling `response.raise_for_status()` before returning - adding a retry similarly to what is done in the synchronous counterpart `embed_documents` I also added an integration test, but willing to move it to unit tests if more relevant.
This commit is contained in:
@@ -1,5 +1,11 @@
|
||||
"""Test MistralAI Embedding."""
|
||||
|
||||
from unittest.mock import patch
|
||||
|
||||
import httpx
|
||||
import pytest
|
||||
import tenacity
|
||||
|
||||
from langchain_mistralai import MistralAIEmbeddings
|
||||
|
||||
|
||||
@@ -29,6 +35,21 @@ async def test_mistralai_embedding_documents_async() -> None:
|
||||
assert len(output[0]) == 1024
|
||||
|
||||
|
||||
async def test_mistralai_embedding_documents_http_error_async() -> None:
|
||||
"""Test MistralAI embeddings for documents."""
|
||||
documents = ["foo bar", "test document"]
|
||||
embedding = MistralAIEmbeddings(max_retries=0)
|
||||
mock_response = httpx.Response(
|
||||
status_code=400,
|
||||
request=httpx.Request("POST", url=embedding.async_client.base_url),
|
||||
)
|
||||
with (
|
||||
patch.object(embedding.async_client, "post", return_value=mock_response),
|
||||
pytest.raises(tenacity.RetryError),
|
||||
):
|
||||
await embedding.aembed_documents(documents)
|
||||
|
||||
|
||||
async def test_mistralai_embedding_query_async() -> None:
|
||||
"""Test MistralAI embeddings for query."""
|
||||
document = "foo bar"
|
||||
|
||||
Reference in New Issue
Block a user