wehub-resource-sync
c56bef871b
CodeQL / Analyze (python) (push) Has been cancelled
Update Platform Components Table / update (push) Has been cancelled
Docker image release / Build base image (push) Has been cancelled
Sync docs with Docusaurus / sync (push) Has been cancelled
Tests / Check if changed (push) Has been cancelled
Tests / format (push) Has been cancelled
Tests / check-imports (push) Has been cancelled
Tests / Unit / macos-latest (push) Has been cancelled
Tests / Unit / ubuntu-latest (push) Has been cancelled
Tests / Unit / windows-latest (push) Has been cancelled
Tests / mypy (push) Has been cancelled
Tests / Integration / ubuntu-latest (push) Has been cancelled
Tests / Integration / macos-latest (push) Has been cancelled
Tests / Integration / windows-latest (push) Has been cancelled
Tests / notify-slack-on-failure (push) Has been cancelled
Tests / Mark tests as completed (push) Has been cancelled
297 lines
14 KiB
Python
297 lines
14 KiB
Python
# SPDX-FileCopyrightText: 2022-present deepset GmbH <info@deepset.ai>
|
|
#
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
import os
|
|
from unittest.mock import AsyncMock, MagicMock
|
|
|
|
import pytest
|
|
from openai import OpenAIError
|
|
|
|
import haystack.components.embedders.azure_text_embedder as azure_text_embedder_module
|
|
from haystack.components.embedders import AzureOpenAITextEmbedder
|
|
from haystack.utils.azure import default_azure_ad_token_provider
|
|
|
|
|
|
class TestAzureOpenAITextEmbedder:
|
|
def test_init_default(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
|
|
assert embedder.api_key.resolve_value() == "fake-api-key"
|
|
assert embedder.azure_deployment == "text-embedding-ada-002"
|
|
assert embedder.model == "text-embedding-ada-002"
|
|
assert embedder.dimensions is None
|
|
assert embedder.organization is None
|
|
assert embedder.prefix == ""
|
|
assert embedder.suffix == ""
|
|
assert embedder.timeout is None
|
|
assert embedder.max_retries is None
|
|
assert embedder.default_headers == {}
|
|
assert embedder.azure_ad_token_provider is None
|
|
assert embedder.http_client_kwargs is None
|
|
assert embedder.client is None
|
|
assert embedder.async_client is None
|
|
|
|
def test_init_with_zero_max_retries(self, monkeypatch):
|
|
"""Tests that the max_retries init param is set correctly if equal 0"""
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/", max_retries=0)
|
|
|
|
assert embedder.api_key.resolve_value() == "fake-api-key"
|
|
assert embedder.azure_deployment == "text-embedding-ada-002"
|
|
assert embedder.model == "text-embedding-ada-002"
|
|
assert embedder.dimensions is None
|
|
assert embedder.organization is None
|
|
assert embedder.prefix == ""
|
|
assert embedder.suffix == ""
|
|
assert embedder.default_headers == {}
|
|
assert embedder.azure_ad_token_provider is None
|
|
assert embedder.max_retries == 0
|
|
assert embedder.client is None
|
|
assert embedder.async_client is None
|
|
|
|
def test_to_dict_default(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
component = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
data = component.to_dict()
|
|
assert data == {
|
|
"type": "haystack.components.embedders.azure_text_embedder.AzureOpenAITextEmbedder",
|
|
"init_parameters": {
|
|
"api_key": {"env_vars": ["AZURE_OPENAI_API_KEY"], "strict": False, "type": "env_var"},
|
|
"azure_ad_token": {"env_vars": ["AZURE_OPENAI_AD_TOKEN"], "strict": False, "type": "env_var"},
|
|
"azure_deployment": "text-embedding-ada-002",
|
|
"dimensions": None,
|
|
"organization": None,
|
|
"azure_endpoint": "https://example-resource.azure.openai.com/",
|
|
"api_version": "2023-05-15",
|
|
"max_retries": None,
|
|
"timeout": None,
|
|
"prefix": "",
|
|
"suffix": "",
|
|
"default_headers": {},
|
|
"azure_ad_token_provider": None,
|
|
"http_client_kwargs": None,
|
|
},
|
|
}
|
|
|
|
def test_to_dict_with_params(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
component = AzureOpenAITextEmbedder(
|
|
azure_endpoint="https://example-resource.azure.openai.com/",
|
|
azure_deployment="text-embedding-ada-002",
|
|
dimensions=768,
|
|
organization="HaystackCI",
|
|
timeout=60.0,
|
|
max_retries=10,
|
|
prefix="prefix ",
|
|
suffix=" suffix",
|
|
default_headers={"x-custom-header": "custom-value"},
|
|
azure_ad_token_provider=default_azure_ad_token_provider,
|
|
http_client_kwargs={"proxy": "http://example.com:3128", "verify": False},
|
|
)
|
|
data = component.to_dict()
|
|
assert data == {
|
|
"type": "haystack.components.embedders.azure_text_embedder.AzureOpenAITextEmbedder",
|
|
"init_parameters": {
|
|
"api_key": {"env_vars": ["AZURE_OPENAI_API_KEY"], "strict": False, "type": "env_var"},
|
|
"azure_ad_token": {"env_vars": ["AZURE_OPENAI_AD_TOKEN"], "strict": False, "type": "env_var"},
|
|
"azure_deployment": "text-embedding-ada-002",
|
|
"dimensions": 768,
|
|
"organization": "HaystackCI",
|
|
"azure_endpoint": "https://example-resource.azure.openai.com/",
|
|
"api_version": "2023-05-15",
|
|
"max_retries": 10,
|
|
"timeout": 60.0,
|
|
"prefix": "prefix ",
|
|
"suffix": " suffix",
|
|
"default_headers": {"x-custom-header": "custom-value"},
|
|
"azure_ad_token_provider": "haystack.utils.azure.default_azure_ad_token_provider",
|
|
"http_client_kwargs": {"proxy": "http://example.com:3128", "verify": False},
|
|
},
|
|
}
|
|
|
|
def test_from_dict(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
data = {
|
|
"type": "haystack.components.embedders.azure_text_embedder.AzureOpenAITextEmbedder",
|
|
"init_parameters": {
|
|
"api_key": {"env_vars": ["AZURE_OPENAI_API_KEY"], "strict": False, "type": "env_var"},
|
|
"azure_ad_token": {"env_vars": ["AZURE_OPENAI_AD_TOKEN"], "strict": False, "type": "env_var"},
|
|
"azure_deployment": "text-embedding-ada-002",
|
|
"dimensions": None,
|
|
"organization": None,
|
|
"azure_endpoint": "https://example-resource.azure.openai.com/",
|
|
"api_version": "2023-05-15",
|
|
"max_retries": 5,
|
|
"timeout": 30.0,
|
|
"prefix": "",
|
|
"suffix": "",
|
|
"default_headers": {},
|
|
"http_client_kwargs": None,
|
|
},
|
|
}
|
|
component = AzureOpenAITextEmbedder.from_dict(data)
|
|
assert component.azure_deployment == "text-embedding-ada-002"
|
|
assert component.model == "text-embedding-ada-002"
|
|
assert component.azure_endpoint == "https://example-resource.azure.openai.com/"
|
|
assert component.api_version == "2023-05-15"
|
|
assert component.max_retries == 5
|
|
assert component.timeout == 30.0
|
|
assert component.prefix == ""
|
|
assert component.suffix == ""
|
|
assert component.default_headers == {}
|
|
assert component.azure_ad_token_provider is None
|
|
assert component.http_client_kwargs is None
|
|
|
|
def test_from_dict_with_parameters(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
data = {
|
|
"type": "haystack.components.embedders.azure_text_embedder.AzureOpenAITextEmbedder",
|
|
"init_parameters": {
|
|
"api_key": {"env_vars": ["AZURE_OPENAI_API_KEY"], "strict": False, "type": "env_var"},
|
|
"azure_ad_token": {"env_vars": ["AZURE_OPENAI_AD_TOKEN"], "strict": False, "type": "env_var"},
|
|
"azure_deployment": "text-embedding-ada-002",
|
|
"dimensions": 768,
|
|
"organization": "HaystackCI",
|
|
"azure_endpoint": "https://example-resource.azure.openai.com/",
|
|
"api_version": "2023-05-15",
|
|
"max_retries": 10,
|
|
"timeout": 60.0,
|
|
"prefix": "prefix ",
|
|
"suffix": " suffix",
|
|
"default_headers": {"x-custom-header": "custom-value"},
|
|
"azure_ad_token_provider": "haystack.utils.azure.default_azure_ad_token_provider",
|
|
"http_client_kwargs": {"proxy": "http://example.com:3128", "verify": False},
|
|
},
|
|
}
|
|
component = AzureOpenAITextEmbedder.from_dict(data)
|
|
assert component.azure_deployment == "text-embedding-ada-002"
|
|
assert component.model == "text-embedding-ada-002"
|
|
assert component.azure_endpoint == "https://example-resource.azure.openai.com/"
|
|
assert component.api_version == "2023-05-15"
|
|
assert component.max_retries == 10
|
|
assert component.timeout == 60.0
|
|
assert component.prefix == "prefix "
|
|
assert component.suffix == " suffix"
|
|
assert component.default_headers == {"x-custom-header": "custom-value"}
|
|
assert component.azure_ad_token_provider is not None
|
|
assert component.http_client_kwargs == {"proxy": "http://example.com:3128", "verify": False}
|
|
|
|
@pytest.mark.integration
|
|
@pytest.mark.skipif(
|
|
not os.environ.get("AZURE_OPENAI_API_KEY", None) and not os.environ.get("AZURE_OPENAI_ENDPOINT", None),
|
|
reason=(
|
|
"Please export env variables called AZURE_OPENAI_API_KEY containing "
|
|
"the Azure OpenAI key, AZURE_OPENAI_ENDPOINT containing "
|
|
"the Azure OpenAI endpoint URL to run this test."
|
|
),
|
|
)
|
|
def test_run(self):
|
|
# the default model is text-embedding-ada-002 even if we don't specify it, but let's be explicit
|
|
embedder = AzureOpenAITextEmbedder(
|
|
azure_deployment="text-embedding-ada-002", prefix="prefix ", suffix=" suffix", organization="HaystackCI"
|
|
)
|
|
result = embedder.run(text="The food was delicious")
|
|
|
|
assert len(result["embedding"]) == 1536
|
|
assert all(isinstance(x, float) for x in result["embedding"])
|
|
assert result["meta"]["usage"] == {"prompt_tokens": 6, "total_tokens": 6}
|
|
assert "ada" in result["meta"]["model"]
|
|
|
|
|
|
@pytest.fixture
|
|
def mock_azure_clients(monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake")
|
|
sync_cls = MagicMock(name="AzureOpenAI")
|
|
async_cls = MagicMock(name="AsyncAzureOpenAI")
|
|
async_cls.return_value.close = AsyncMock()
|
|
monkeypatch.setattr(azure_text_embedder_module, "AzureOpenAI", sync_cls)
|
|
monkeypatch.setattr(azure_text_embedder_module, "AsyncAzureOpenAI", async_cls)
|
|
return sync_cls, async_cls
|
|
|
|
|
|
class TestComponentLifecycle:
|
|
def test_warm_up_uses_default_timeout_and_max_retries(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
embedder.warm_up()
|
|
assert embedder.client is not None
|
|
assert embedder.client.max_retries == 5
|
|
assert embedder.client.timeout == 30.0
|
|
|
|
def test_warm_up_uses_timeout_and_max_retries_from_parameters(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
embedder = AzureOpenAITextEmbedder(
|
|
azure_endpoint="https://example-resource.azure.openai.com/", timeout=40.0, max_retries=1
|
|
)
|
|
embedder.warm_up()
|
|
assert embedder.client is not None
|
|
assert embedder.client.max_retries == 1
|
|
assert embedder.client.timeout == 40.0
|
|
|
|
def test_warm_up_uses_timeout_and_max_retries_from_env_vars(self, monkeypatch):
|
|
monkeypatch.setenv("AZURE_OPENAI_API_KEY", "fake-api-key")
|
|
monkeypatch.setenv("OPENAI_TIMEOUT", "100")
|
|
monkeypatch.setenv("OPENAI_MAX_RETRIES", "10")
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
embedder.warm_up()
|
|
assert embedder.client is not None
|
|
assert embedder.client.max_retries == 10
|
|
assert embedder.client.timeout == 100.0
|
|
|
|
def test_key_resolved_at_warm_up_not_init(self, monkeypatch):
|
|
monkeypatch.delenv("AZURE_OPENAI_API_KEY", raising=False)
|
|
monkeypatch.delenv("AZURE_OPENAI_AD_TOKEN", raising=False)
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
assert embedder.client is None
|
|
with pytest.raises(OpenAIError):
|
|
embedder.warm_up()
|
|
|
|
def test_sync_lifecycle(self, mock_azure_clients):
|
|
sync_cls, _ = mock_azure_clients
|
|
sync_client = sync_cls.return_value
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
assert embedder.client is None
|
|
assert embedder.async_client is None
|
|
|
|
embedder.warm_up()
|
|
assert embedder.client is sync_cls.return_value
|
|
assert embedder.async_client is None
|
|
|
|
embedder.close()
|
|
sync_client.close.assert_called_once()
|
|
assert embedder.client is None
|
|
|
|
async def test_async_lifecycle(self, mock_azure_clients):
|
|
_, async_cls = mock_azure_clients
|
|
async_client = async_cls.return_value
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
|
|
await embedder.warm_up_async()
|
|
assert embedder.async_client is async_cls.return_value
|
|
assert embedder.client is None
|
|
|
|
await embedder.close_async()
|
|
async_client.close.assert_awaited_once()
|
|
assert embedder.async_client is None
|
|
|
|
async def test_close_is_safe_without_warm_up(self, mock_azure_clients):
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
embedder.close()
|
|
await embedder.close_async()
|
|
assert embedder.client is None
|
|
assert embedder.async_client is None
|
|
|
|
async def test_close_and_close_async_are_independent(self, mock_azure_clients):
|
|
embedder = AzureOpenAITextEmbedder(azure_endpoint="https://example-resource.azure.openai.com/")
|
|
embedder.warm_up()
|
|
await embedder.warm_up_async()
|
|
|
|
embedder.close()
|
|
assert embedder.client is None
|
|
assert embedder.async_client is not None
|
|
|
|
await embedder.close_async()
|
|
assert embedder.async_client is None
|