test_add_texts.py — langchain Source File

Source Code

from __future__ import annotations

import uuid

import pytest  # type: ignore[import-not-found]
from langchain_core.documents import Document

from langchain_qdrant import Qdrant
from tests.integration_tests.common import (
    ConsistentFakeEmbeddings,
    assert_documents_equals,
)


@pytest.mark.parametrize("batch_size", [1, 64])
@pytest.mark.parametrize("vector_name", [None, "my-vector"])
def test_qdrant_add_documents_extends_existing_collection(
    batch_size: int, vector_name: str | None
) -> None:
    """Test end to end construction and search."""
    texts = ["foo", "bar", "baz"]
    docsearch: Qdrant = Qdrant.from_texts(
        texts,
        ConsistentFakeEmbeddings(),
        location=":memory:",
        batch_size=batch_size,
        vector_name=vector_name,
    )

    new_texts = ["foobar", "foobaz"]
    docsearch.add_documents(
        [Document(page_content=content) for content in new_texts], batch_size=batch_size
    )
    output = docsearch.similarity_search("foobar", k=1)
    # ConsistentFakeEmbeddings return the same query embedding as the first document
    # embedding computed in `embedding.embed_documents`. Thus, "foo" embedding is the
    # same as "foobar" embedding
    assert_documents_equals(output, [Document(page_content="foobar")])


@pytest.mark.parametrize("batch_size", [1, 64])
def test_qdrant_add_texts_returns_all_ids(batch_size: int) -> None:
    """Test end to end Qdrant.add_texts returns unique ids."""
    docsearch: Qdrant = Qdrant.from_texts(
        ["foobar"],
        ConsistentFakeEmbeddings(),
        location=":memory:",
        batch_size=batch_size,
    )

    ids = docsearch.add_texts(["foo", "bar", "baz"])
    assert len(ids) == 3
    assert len(set(ids)) == 3


@pytest.mark.parametrize("vector_name", [None, "my-vector"])
def test_qdrant_add_texts_stores_duplicated_texts(vector_name: str | None) -> None:
    """Test end to end Qdrant.add_texts stores duplicated texts separately."""
    from qdrant_client import QdrantClient
    from qdrant_client.http import models as rest
// ... (79 more lines)