feat: improve pinecone tests (langchain-ai#2806)

Improve the integration tests for Pinecone by adding an `.env.example`
file for local testing. Additionally, add some dev dependencies
specifically for integration tests.

This change also helps me understand how Pinecone deals with certain
things, see related issues
langchain-ai#2484
langchain-ai#2816

pyproject.toml

@@ -104,6 +104,9 @@ deeplake = "^3.2.21"
 torch = "^1.0.0"
 chromadb = "^0.3.21"
 tiktoken = "^0.3.3"
+python-dotenv = "^1.0.0"
+gptcache = "^0.1.9"
+promptlayer = "^0.1.80"
 
 [tool.poetry.group.lint.dependencies]
 ruff = "^0.0.249"

tests/README.md

@@ -39,6 +39,11 @@ cd tests/integration_tests/vectorstores/docker-compose
 docker-compose -f elasticsearch.yml up
 ```
 
+### Prepare environment variables for local testing:
+
+- copy `tests/.env.example` to `tests/.env`
+- set variables in `tests/.env` file, e.g `OPENAI_API_KEY`
+
 Additionally, it's important to note that some integration tests may require certain
 environment variables to be set, such as `OPENAI_API_KEY`. Be sure to set any required
 environment variables before running the tests to ensure they run correctly.
@@ -54,7 +59,9 @@ cassettes. You can use the --vcr-record=none command-line option to disable reco
 new cassettes. Here's an example:
 
 ```bash
+pytest --log-cli-level=10 tests/integration_tests/vectorstores/test_pinecone.py --vcr-record=none
 pytest tests/integration_tests/vectorstores/test_elasticsearch.py --vcr-record=none
+
 ```
 
 ### Run some tests with coverage:

tests/integration_tests/.env.example

@@ -0,0 +1,9 @@
+# openai
+# your api key from https://platform.openai.com/account/api-keys
+OPENAI_API_KEY=
+
+# pinecone
+# your api key from left menu "API Keys" in https://app.pinecone.io
+PINECONE_API_KEY=your_pinecone_api_key_here
+# your pinecone environment from left menu "API Keys" in https://app.pinecone.io
+PINECONE_ENVIRONMENT=us-west4-gcp

tests/integration_tests/conftest.py

@@ -1,10 +1,31 @@
 import os
+from pathlib import Path
 
 import pytest
 
 # Getting the absolute path of the current file's directory
 ABS_PATH = os.path.dirname(os.path.abspath(__file__))
 
+# Getting the absolute path of the project's root directory
+PROJECT_DIR = os.path.abspath(os.path.join(ABS_PATH, os.pardir, os.pardir))
+
+
+# Loading the .env file if it exists
+def _load_env() -> None:
+    dotenv_path = os.path.join(PROJECT_DIR, "tests", "integration_tests", ".env")
+    if os.path.exists(dotenv_path):
+        from dotenv import load_dotenv
+
+        load_dotenv(dotenv_path)
+
+
+_load_env()
+
+
+@pytest.fixture(scope="module")
+def test_dir() -> Path:
+    return Path(os.path.join(PROJECT_DIR, "tests", "integration_tests"))
+
 
 # This fixture returns a string containing the path to the cassette directory for the
 # current module
@@ -15,18 +36,3 @@ def vcr_cassette_dir(request: pytest.FixtureRequest) -> str:
         "cassettes",
         os.path.basename(request.module.__file__).replace(".py", ""),
     )
-
-
-# This fixture returns a dictionary containing filter_headers options
-# for replacing certain headers with dummy values during cassette playback
-# Specifically, it replaces the authorization header with a dummy value to
-# prevent sensitive data from being recorded in the cassette.
-@pytest.fixture(scope="module")
-def vcr_config() -> dict:
-    return {
-        "filter_headers": [
-            ("authorization", "authorization-DUMMY"),
-            ("X-OpenAI-Client-User-Agent", "X-OpenAI-Client-User-Agent-DUMMY"),
-            ("User-Agent", "User-Agent-DUMMY"),
-        ],
-    }

tests/integration_tests/vectorstores/conftest.py

@@ -1,15 +1,49 @@
 import os
-from typing import Generator, List
+from typing import Generator, List, Union
 
 import pytest
+from vcr.request import Request
 
 from langchain.document_loaders import TextLoader
+from langchain.embeddings import OpenAIEmbeddings
 from langchain.schema import Document
 from langchain.text_splitter import CharacterTextSplitter
 
 
-# Define a fixture that yields a generator object returning a list of documents
+# This fixture returns a dictionary containing filter_headers options
+# for replacing certain headers with dummy values during cassette playback
+# Specifically, it replaces the authorization header with a dummy value to
+# prevent sensitive data from being recorded in the cassette.
+# It also filters request to certain hosts (specified in the `ignored_hosts` list)
+# to prevent data from being recorded in the cassette.
 @pytest.fixture(scope="module")
+def vcr_config() -> dict:
+    skipped_host = ["pinecone.io"]
+
+    def before_record_response(response: dict) -> Union[dict, None]:
+        return response
+
+    def before_record_request(request: Request) -> Union[Request, None]:
+        for host in skipped_host:
+            if request.host.startswith(host) or request.host.endswith(host):
+                return None
+        return request
+
+    return {
+        "before_record_request": before_record_request,
+        "before_record_response": before_record_response,
+        "filter_headers": [
+            ("authorization", "authorization-DUMMY"),
+            ("X-OpenAI-Client-User-Agent", "X-OpenAI-Client-User-Agent-DUMMY"),
+            ("Api-Key", "Api-Key-DUMMY"),
+            ("User-Agent", "User-Agent-DUMMY"),
+        ],
+        "ignore_localhost": True,
+    }
+
+
+# Define a fixture that yields a generator object returning a list of documents
+@pytest.fixture(scope="function")
 def documents() -> Generator[List[Document], None, None]:
     """Return a generator that yields a list of documents."""
 
@@ -23,3 +57,18 @@ def documents() -> Generator[List[Document], None, None]:
 
     # Yield the documents split into chunks
     yield text_splitter.split_documents(documents)
+
+
+@pytest.fixture(scope="function")
+def texts() -> Generator[List[str], None, None]:
+    # Load the documents from a file located in the fixtures directory
+    documents = TextLoader(
+        os.path.join(os.path.dirname(__file__), "fixtures", "sharks.txt")
+    ).load()
+
+    yield [doc.page_content for doc in documents]
+
+
+@pytest.fixture(scope="module")
+def embedding_openai() -> OpenAIEmbeddings:
+    return OpenAIEmbeddings()

tests/integration_tests/vectorstores/test_elasticsearch.py

@@ -21,6 +21,11 @@
 
 
 class TestElasticsearch:
+    @classmethod
+    def setup_class(cls) -> None:
+        if not os.getenv("OPENAI_API_KEY"):
+            raise ValueError("OPENAI_API_KEY environment variable is not set")
+
     @pytest.fixture(scope="class", autouse=True)
     def elasticsearch_url(self) -> Union[str, Generator[str, None, None]]:
         """Return the elasticsearch url."""
@@ -34,15 +39,6 @@ def elasticsearch_url(self) -> Union[str, Generator[str, None, None]]:
             # print(index_name)
             es.indices.delete(index=index_name)
 
-    @pytest.fixture(scope="class", autouse=True)
-    def openai_api_key(self) -> Union[str, Generator[str, None, None]]:
-        """Return the OpenAI API key."""
-        openai_api_key = os.getenv("OPENAI_API_KEY")
-        if not openai_api_key:
-            raise ValueError("OPENAI_API_KEY environment variable is not set")
-
-        yield openai_api_key
-
     def test_similarity_search_without_metadata(self, elasticsearch_url: str) -> None:
         """Test end to end construction and search without metadata."""
         texts = ["foo", "bar", "baz"]
@@ -67,15 +63,17 @@ def test_similarity_search_with_metadata(self, elasticsearch_url: str) -> None:
 
     @pytest.mark.vcr(ignore_localhost=True)
     def test_default_index_from_documents(
-        self, documents: List[Document], openai_api_key: str, elasticsearch_url: str
+        self,
+        documents: List[Document],
+        embedding_openai: OpenAIEmbeddings,
+        elasticsearch_url: str,
     ) -> None:
         """This test checks the construction of a default
         ElasticSearch index using the 'from_documents'."""
-        embedding = OpenAIEmbeddings(openai_api_key=openai_api_key)
 
         elastic_vector_search = ElasticVectorSearch.from_documents(
             documents=documents,
-            embedding=embedding,
+            embedding=embedding_openai,
             elasticsearch_url=elasticsearch_url,
         )
 
@@ -86,16 +84,18 @@ def test_default_index_from_documents(
 
     @pytest.mark.vcr(ignore_localhost=True)
     def test_custom_index_from_documents(
-        self, documents: List[Document], openai_api_key: str, elasticsearch_url: str
+        self,
+        documents: List[Document],
+        embedding_openai: OpenAIEmbeddings,
+        elasticsearch_url: str,
     ) -> None:
         """This test checks the construction of a custom
         ElasticSearch index using the 'from_documents'."""
 
         index_name = f"custom_index_{uuid.uuid4().hex}"
-        embedding = OpenAIEmbeddings(openai_api_key=openai_api_key)
         elastic_vector_search = ElasticVectorSearch.from_documents(
             documents=documents,
-            embedding=embedding,
+            embedding=embedding_openai,
             elasticsearch_url=elasticsearch_url,
             index_name=index_name,
         )
@@ -110,15 +110,17 @@ def test_custom_index_from_documents(
 
     @pytest.mark.vcr(ignore_localhost=True)
     def test_custom_index_add_documents(
-        self, documents: List[Document], openai_api_key: str, elasticsearch_url: str
+        self,
+        documents: List[Document],
+        embedding_openai: OpenAIEmbeddings,
+        elasticsearch_url: str,
     ) -> None:
         """This test checks the construction of a custom
         ElasticSearch index using the 'add_documents'."""
 
         index_name = f"custom_index_{uuid.uuid4().hex}"
-        embedding = OpenAIEmbeddings(openai_api_key=openai_api_key)
         elastic_vector_search = ElasticVectorSearch(
-            embedding=embedding,
+            embedding=embedding_openai,
             elasticsearch_url=elasticsearch_url,
             index_name=index_name,
         )