test_from_texts.py — langchain Source File

Source Code

from __future__ import annotations

import os
import uuid

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

from langchain_qdrant import Qdrant
from langchain_qdrant.vectorstores import QdrantException
from tests.integration_tests.common import (
    ConsistentFakeEmbeddings,
    assert_documents_equals,
)
from tests.integration_tests.fixtures import (
    qdrant_locations,
)


@pytest.mark.parametrize("qdrant_location", qdrant_locations())
async def test_qdrant_from_texts_stores_duplicated_texts(qdrant_location: str) -> None:
    """Test end to end Qdrant.afrom_texts stores duplicated texts separately."""
    collection_name = uuid.uuid4().hex

    vec_store = await Qdrant.afrom_texts(
        ["abc", "abc"],
        ConsistentFakeEmbeddings(),
        collection_name=collection_name,
        location=qdrant_location,
    )

    client = vec_store.client
    assert client.count(collection_name).count == 2


@pytest.mark.parametrize("batch_size", [1, 64])
@pytest.mark.parametrize("vector_name", [None, "my-vector"])
@pytest.mark.parametrize("qdrant_location", qdrant_locations())
async def test_qdrant_from_texts_stores_ids(
    batch_size: int, vector_name: str | None, qdrant_location: str
) -> None:
    """Test end to end Qdrant.afrom_texts stores provided ids."""
    collection_name = uuid.uuid4().hex
    ids = [
        "fa38d572-4c31-4579-aedc-1960d79df6df",
        "cdc1aa36-d6ab-4fb2-8a94-56674fd27484",
    ]
    vec_store = await Qdrant.afrom_texts(
        ["abc", "def"],
        ConsistentFakeEmbeddings(),
        ids=ids,
        collection_name=collection_name,
        batch_size=batch_size,
        vector_name=vector_name,
        location=qdrant_location,
    )

    client = vec_store.client
    assert client.count(collection_name).count == 2
    stored_ids = [point.id for point in client.scroll(collection_name)[0]]
// ... (209 more lines)