Feature/azure client

docs/docs/snippets/cli_options.md

@@ -1,4 +1,5 @@
 <!-- --8<------ [start:options] -->
+
 ```console
 Usage: readmeai [OPTIONS]
 
@@ -9,7 +10,7 @@ Options:
   -a, --align [center|left|right]
                                   align for the README.md file header
                                   sections.
-  --api [anthropic|gemini|ollama|openai|offline]
+  --api [anthropic|gemini|ollama|openai|azure|offline]
                                   LLM API service provider to power the README
                                   file generation.
   -bc, --badge-color TEXT         Primary color (hex code or name) to use for
@@ -50,4 +51,5 @@ Options:
                                   generated for the README file.
   --help                          Show this message and exit.
 ```
+
 <!-- --8<------ [end:options] -->

readmeai/models/azure.py

@@ -0,0 +1,84 @@
+"""Azure OpenAI model handler for ReadmeAI."""
+import os
+from typing import Any
+
+import aiohttp
+import openai
+from openai import AsyncAzureOpenAI
+
+from readmeai.config.settings import ConfigLoader
+from readmeai.extractors.models import RepositoryContext
+from readmeai.models.openai_ import OpenAIHandler
+from readmeai.models.tokens import token_handler
+
+
+class AzureOpenAIHandler(OpenAIHandler):
+    """Handler for Azure OpenAI models."""
+
+    def __init__(self, config_loader: ConfigLoader, context: RepositoryContext) -> None:
+        super().__init__(config_loader, context)
+        self._model_settings()
+
+    def _model_settings(self):
+        self.host_name = os.getenv("AZURE_ENDPOINT")
+        self.max_tokens = self.config.llm.tokens
+        self.model = os.getenv("AZURE_MODEL")
+        self.top_p = self.config.llm.top_p
+
+        self.client = AsyncAzureOpenAI(
+            azure_endpoint=self.host_name,
+            api_key=os.getenv("AZURE_API_KEY"),
+            api_version=os.getenv("AZURE_API_VERSION"),
+        )
+
+    async def _make_request(
+            self,
+            index: str | None,
+            prompt: str | None,
+            tokens: int | None,
+            repo_files: Any,
+    ):
+        """Process requests to OpenAI API, with retries and error handling."""
+
+        try:
+            if prompt is None:
+                raise ValueError("Prompt cannot be None")
+
+            prompt = await token_handler(
+                config=self.config,
+                index=index,
+                prompt=prompt,
+                tokens=tokens,
+            )
+            if not prompt:
+                raise ValueError("Token handler returned empty prompt")
+
+            if index == "file_summary":
+                self.max_tokens = 100
+
+            parameters = await self._build_payload(prompt)
+
+            # Just await the create call directly
+            response = await self.client.chat.completions.create(**parameters)
+            content = response.choices[0].message.content
+
+            if not content:
+                raise ValueError("Empty response from API")
+
+            self._logger.info(
+                f"Response from {self.config.llm.api.capitalize()} for '{index}': {content}",
+            )
+            return index, content
+
+        except (
+                aiohttp.ClientError,
+                aiohttp.ClientResponseError,
+                aiohttp.ClientConnectorError,
+                openai.OpenAIError,
+        ) as e:
+            self._logger.error(f"Error processing request for '{index}': {e!r}")
+            raise  # Re-raise for retry decorator
+
+        except Exception as e:
+            self._logger.error(f"Unexpected error for '{index}': {e!r}")
+            return index, self.placeholder

readmeai/models/enums.py

@@ -10,6 +10,7 @@ class LLMAuthKeys(str, Enum):
     GOOGLE_API_KEY = "GOOGLE_API_KEY"
     OLLAMA_HOST = "OLLAMA_HOST"
     OPENAI_API_KEY = "OPENAI_API_KEY"
+    AZURE_API_KEY = "AZURE_API_KEY"
 
 
 class LLMProviders(str, Enum):
@@ -22,6 +23,7 @@ class LLMProviders(str, Enum):
     OLLAMA = "ollama"
     OPENAI = "openai"
     OFFLINE = "offline"
+    AZURE = "azure"
 
 
 class AnthropicModels(str, Enum):

readmeai/models/factory.py

@@ -4,11 +4,12 @@
 from readmeai.core.errors import UnsupportedServiceError
 from readmeai.extractors.models import RepositoryContext
 from readmeai.models.anthropic import AnthropicHandler
+from readmeai.models.azure import AzureOpenAIHandler
 from readmeai.models.base import BaseModelHandler
 from readmeai.models.enums import LLMProviders
 from readmeai.models.gemini import GeminiHandler
 from readmeai.models.offline import OfflineHandler
-from readmeai.models.openai import OpenAIHandler
+from readmeai.models.openai_ import OpenAIHandler
 
 
 class ModelFactory:
@@ -22,18 +23,15 @@ class ModelFactory:
         LLMProviders.OLLAMA.value: OpenAIHandler,
         LLMProviders.OPENAI.value: OpenAIHandler,
         LLMProviders.OFFLINE.value: OfflineHandler,
+        LLMProviders.AZURE.value: AzureOpenAIHandler,
     }
 
     @staticmethod
-    def get_backend(
-        config: ConfigLoader, context: RepositoryContext
-    ) -> BaseModelHandler:
+    def get_backend(config: ConfigLoader, context: RepositoryContext) -> BaseModelHandler:
         """Retrieves configured LLM API handler instance."""
         llm_service = ModelFactory._model_map.get(config.config.llm.api)
 
         if llm_service is None:
-            raise UnsupportedServiceError(
-                f"Unsupported LLM provider: {config.config.llm.api}"
-            )
+            raise UnsupportedServiceError(f"Unsupported LLM provider: {config.config.llm.api}")
 
         return llm_service(config, context)

tests/cli/test_main.py

@@ -1,8 +1,8 @@
-from unittest.mock import MagicMock, patch
-
 import pytest
 from _pytest._py.path import LocalPath
 from click.testing import CliRunner
+from unittest.mock import MagicMock, patch
+
 from readmeai.cli.main import main
 from readmeai.config.settings import ConfigLoader
 
@@ -24,13 +24,13 @@ def mock_readme_agent():
 
 
 def test_main_command_basic(
-    cli_runner: CliRunner,
-    mock_config_loader: ConfigLoader,
-    output_file_path: str,
+        cli_runner: CliRunner,
+        mock_config_loader: ConfigLoader,
+        output_file_path: str,
 ):
     with patch(
-        "readmeai.config.settings.ConfigLoader",
-        return_value=mock_config_loader,
+            "readmeai.config.settings.ConfigLoader",
+            return_value=mock_config_loader,
     ):
         result = cli_runner.invoke(
             main,
@@ -46,9 +46,9 @@ def test_main_command_basic(
 
 
 def test_main_command_all_options(
-    cli_runner: CliRunner,
-    mock_config_loader: ConfigLoader,
-    output_file_path: str,
+        cli_runner: CliRunner,
+        mock_config_loader: ConfigLoader,
+        output_file_path: str,
 ):
     mock_config = mock_config_loader
     mock_config.config.git.repository = "https://github.com/eli64s/readme-ai-streamlit"
@@ -108,61 +108,62 @@ def test_version_option(cli_runner: CliRunner):
     [
         ("--align", "invalid", "Invalid value for '-a' / '--align'"),
         (
-            "--api",
-            "invalid",
-            "Invalid value for '--api': 'invalid' is not one of 'anthropic', 'gemini', 'ollama', 'openai', 'offline'.",
+                "--api",
+                "invalid",
+                "Invalid value for '--api': 'invalid' is not one of 'anthropic', 'gemini', 'ollama', 'openai', 'offline', "
+                "'azure'.",
         ),
         # (
         #     "--badge-color",
         #     "invalid",
         #     "Invalid value for '-bc' / '--badge-color'",
         # ),
         (
-            "--badge-style",
-            "invalid",
-            "Invalid value for '-bs' / '--badge-style'",
+                "--badge-style",
+                "invalid",
+                "Invalid value for '-bs' / '--badge-style'",
         ),
         (
-            "--context-window",
-            "invalid",
-            "Invalid value for '-cw' / '--context-window'",
+                "--context-window",
+                "invalid",
+                "Invalid value for '-cw' / '--context-window'",
         ),
         (
-            "--header-style",
-            "invalid",
-            "Invalid value for '-hs' / '--header-style'",
+                "--header-style",
+                "invalid",
+                "Invalid value for '-hs' / '--header-style'",
         ),
         ("--logo", "invalid", "Invalid value for '-l' / '--logo'"),
         # ("--model", "invalid", "Invalid value for '-m' / '--model'"),
         (
-            "--rate-limit",
-            "invalid",
-            "Invalid value for '-rl' / '--rate-limit'",
+                "--rate-limit",
+                "invalid",
+                "Invalid value for '-rl' / '--rate-limit'",
         ),
         (
-            "--temperature",
-            "invalid",
-            "Invalid value for '-t' / '--temperature'",
+                "--temperature",
+                "invalid",
+                "Invalid value for '-t' / '--temperature'",
         ),
         (
-            "--navigation-style",
-            "invalid",
-            "Invalid value for '-ns' / '--navigation-style'",
+                "--navigation-style",
+                "invalid",
+                "Invalid value for '-ns' / '--navigation-style'",
         ),
         (
-            "--top-p",
-            "invalid",
-            "Invalid value for '-tp' / '--top-p'",
+                "--top-p",
+                "invalid",
+                "Invalid value for '-tp' / '--top-p'",
         ),
         (
-            "--tree-max-depth",
-            "invalid",
-            "Invalid value for '-td' / '--tree-max-depth'",
+                "--tree-max-depth",
+                "invalid",
+                "Invalid value for '-td' / '--tree-max-depth'",
         ),
     ],
 )
 def test_invalid_option_values(
-    temp_dir: LocalPath, cli_runner: CliRunner, option, value, expected
+        temp_dir: LocalPath, cli_runner: CliRunner, option, value, expected
 ):
     result = cli_runner.invoke(main, ["--repository", str(temp_dir), option, value])
     assert result.exit_code != 0

tests/models/conftest.py

@@ -4,12 +4,14 @@
 
 import aiohttp
 import pytest
+
 from readmeai.config.settings import ConfigLoader
 from readmeai.extractors.models import RepositoryContext
 from readmeai.models.anthropic import ANTHROPIC_AVAILABLE, AnthropicHandler
+from readmeai.models.azure import AzureOpenAIHandler
 from readmeai.models.enums import GeminiModels
 from readmeai.models.gemini import GeminiHandler
-from readmeai.models.openai import OpenAIHandler
+from readmeai.models.openai_ import OpenAIHandler
 
 
 @pytest.fixture
@@ -46,9 +48,7 @@ def mock_aiohttp_session():
 
 
 @pytest.fixture
-def anthropic_handler(
-    mock_config_loader: ConfigLoader, mock_repository_context: RepositoryContext
-):
+def anthropic_handler(mock_config_loader: ConfigLoader, mock_repository_context: RepositoryContext):
     if not ANTHROPIC_AVAILABLE:
         pytest.skip("Anthropic library is not available")
     context = mock_repository_context
@@ -60,20 +60,18 @@ def anthropic_handler(
 
 @pytest.fixture
 def anthropic_handler_with_mock_session(
-    anthropic_handler: AnthropicHandler, monkeypatch: pytest.MonkeyPatch
+        anthropic_handler: AnthropicHandler, monkeypatch: pytest.MonkeyPatch
 ):
     monkeypatch.setenv("ANTHROPIC_API_KEY", "test_api_key")
-    mock_create = AsyncMock(
-        return_value=MagicMock(content=[MagicMock(text="test_response")])
-    )
+    mock_create = AsyncMock(return_value=MagicMock(content=[MagicMock(text="test_response")]))
     anthropic_handler.client.messages.create = mock_create
     return anthropic_handler
 
 
 @pytest.fixture
 def gemini_handler(
-    mock_config_loader: ConfigLoader,
-    mock_repository_context: RepositoryContext,
+        mock_config_loader: ConfigLoader,
+        mock_repository_context: RepositoryContext,
 ):
     """Fixture to provide a GeminiHandler instance."""
     mock_config_loader.config.llm.model = GeminiModels.GEMINI_FLASH.value
@@ -86,9 +84,9 @@ def gemini_handler(
 
 @pytest.fixture
 def openai_handler(
-    mock_config_loader: ConfigLoader,
-    mock_repository_context: RepositoryContext,
-    monkeypatch: pytest.MonkeyPatch,
+        mock_config_loader: ConfigLoader,
+        mock_repository_context: RepositoryContext,
+        monkeypatch: pytest.MonkeyPatch,
 ):
     """Fixture to provide an OpenAIHandler instance with a mocked API key."""
     monkeypatch.setenv("OPENAI_API_KEY", "test_api_key")
@@ -100,13 +98,39 @@ def openai_handler(
 
 @pytest.fixture
 def openai_handler_with_mock_session(
-    openai_handler: OpenAIHandler, mock_aiohttp_session: MagicMock
+        openai_handler: OpenAIHandler, mock_aiohttp_session: MagicMock
 ):
     """Fixture to provide an OpenAIHandler with a mocked session."""
     openai_handler._session = mock_aiohttp_session
     return openai_handler
 
 
+@pytest.fixture
+def azure_openai_handler(
+        mock_config_loader: ConfigLoader,
+        mock_repository_context: RepositoryContext,
+):
+    """Fixture to provide an AzureOpenAIHandler instance."""
+    with patch.dict("os.environ", {
+        "AZURE_ENDPOINT": "https://test.azure.com",
+        "AZURE_API_KEY": "test_azure_api_key",
+        "AZURE_API_VERSION": "2023-05-15",
+    }):
+        return AzureOpenAIHandler(
+            config_loader=mock_config_loader,
+            context=mock_repository_context,
+        )
+
+
+@pytest.fixture
+def azure_openai_handler_with_mock_session(
+        azure_openai_handler: AzureOpenAIHandler, mock_aiohttp_session: MagicMock
+):
+    """Fixture to provide an AzureOpenAIHandler with a mocked session."""
+    azure_openai_handler._session = mock_aiohttp_session
+    return azure_openai_handler
+
+
 @pytest.fixture
 def ollama_localhost():
     """Fixture to provide a localhost URL for Ollama."""