test_from_texts.py — langchain Source File

Source Code

from __future__ import annotations

import tempfile
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


def test_qdrant_from_texts_stores_duplicated_texts() -> None:
    """Test end to end Qdrant.from_texts stores duplicated texts separately."""
    from qdrant_client import QdrantClient

    collection_name = uuid.uuid4().hex

    with tempfile.TemporaryDirectory() as tmpdir:
        vec_store = Qdrant.from_texts(
            ["abc", "abc"],
            ConsistentFakeEmbeddings(),
            collection_name=collection_name,
            path=str(tmpdir),
        )
        del vec_store

        client = QdrantClient(path=str(tmpdir))
        assert client.count(collection_name).count == 2


@pytest.mark.parametrize("batch_size", [1, 64])
@pytest.mark.parametrize("vector_name", [None, "my-vector"])
def test_qdrant_from_texts_stores_ids(batch_size: int, vector_name: str | None) -> None:
    """Test end to end Qdrant.from_texts stores provided ids."""
    from qdrant_client import QdrantClient

    collection_name = uuid.uuid4().hex
    with tempfile.TemporaryDirectory() as tmpdir:
        ids = [
            "fa38d572-4c31-4579-aedc-1960d79df6df",
            "cdc1aa36-d6ab-4fb2-8a94-56674fd27484",
        ]
        vec_store = Qdrant.from_texts(
            ["abc", "def"],
            ConsistentFakeEmbeddings(),
            ids=ids,
            collection_name=collection_name,
            path=str(tmpdir),
            batch_size=batch_size,
            vector_name=vector_name,
        )
        del vec_store

        client = QdrantClient(path=str(tmpdir))
// ... (228 more lines)