Merge pull request #11 from blink1073/INTPYTHON-642

INTPYTHON-642 Ensure client certificates

Author: blink1073

pymongo_voyageai/utils.py

@@ -1,7 +1,9 @@
 import io
+import ssl
 import urllib.request
 from typing import Any
 
+import certifi
 from PIL import Image
 
 from .document import ImageDocument
@@ -104,7 +106,8 @@ def url_to_images(
         storage.close()
     # For all other files, use the native download.
     if source is None:
-        with urllib.request.urlopen(url) as response:
+        ssl_context = ssl.create_default_context(cafile=certifi.where())
+        with urllib.request.urlopen(url, context=ssl_context) as response:
             source = io.BytesIO(response.read())
 
     if url.endswith(".parquet"):

pyproject.toml

@@ -29,6 +29,7 @@ classifiers = [
 ]
 dependencies = [
     "boto3>=1.37.25",
+    "certifi>=2025.4.26",
     "langchain-mongodb>=0.6.0",
     "pymongo>=4.12.0",
     "voyageai>=0.3.0",

tests/test_client_integration.py

@@ -1,7 +1,9 @@
 import os
+import ssl
 import urllib.request
 from collections.abc import Generator
 
+import certifi
 import numpy as np
 import pytest
 from bson import ObjectId
@@ -82,7 +84,8 @@ def test_pdf_pages_storage(client: PyMongoVoyageAI):
     url = "https://www.fdrlibrary.org/documents/356632/390886/readingcopy.pdf"
     storage = client._storage
     object_name = f"{ObjectId()}.pdf"
-    with urllib.request.urlopen(url) as response:
+    ssl_context = ssl.create_default_context(cafile=certifi.where())
+    with urllib.request.urlopen(url, context=ssl_context) as response:
         storage.client.upload_fileobj(response, storage.root_location, object_name)
     url = f"s3://{storage.root_location}/{object_name}"
     images = client.url_to_images(url)
@@ -100,7 +103,8 @@ def test_pdf_pages_custom_storage(client: PyMongoVoyageAI):
     url = "https://www.fdrlibrary.org/documents/356632/390886/readingcopy.pdf"
     storage = client._storage
     object_name = f"{ObjectId()}.pdf"
-    with urllib.request.urlopen(url) as response:
+    ssl_context = ssl.create_default_context(cafile=certifi.where())
+    with urllib.request.urlopen(url, context=ssl_context) as response:
         storage.client.upload_fileobj(response, storage.root_location, object_name)
     url = f"s3://{storage.root_location}/{object_name}"
     client._storage = MemoryStorage()

tests/test_client_unit.py

@@ -1,10 +1,12 @@
 import os
+import ssl
 import tempfile
 import urllib.request
 from collections.abc import Generator
 from pathlib import Path
 from typing import Any
 
+import certifi
 import pytest
 
 from pymongo_voyageai import MemoryStorage, PyMongoVoyageAI
@@ -89,7 +91,8 @@ def test_pdf_pages(client: PyMongoVoyageAI):
 def test_pdf_pages_storage(client: PyMongoVoyageAI):
     query = "The consequences of a dictator's peace"
     url = "https://www.fdrlibrary.org/documents/356632/390886/readingcopy.pdf"
-    with urllib.request.urlopen(url) as response:
+    ssl_context = ssl.create_default_context(cafile=certifi.where())
+    with urllib.request.urlopen(url, context=ssl_context) as response:
         data = response.read()
     with tempfile.NamedTemporaryFile(suffix=".pdf") as fp:
         fp.write(data)