Summary by GPT-4

In this commit, the README.md file has been updated to provide clearer instructions on how to install and set up the CLI tool. The _llama_index.py file has been updated to support both Azure OpenAI and OpenAI API types. The _openai.py file has been updated to use the term "model" instead of "engine" for consistency. The context.py file has been updated to load context from environment variables or context files, with precedence given to config files if available. The test files have also been updated accordingly.

Overall, these changes improve the clarity of instructions and provide better support for different API types.

Suggestions

Here are some suggestions for improving the changes in this PR:

1. In the README.md file, change the title "How to install CLI" to "How to Install CLI" for consistency in capitalization.

2. In the _llama_index.py file, add a comment explaining the difference between using Azure OpenAI API and OpenAI API directly in the _load_service_context() function.

3. In the context.py file, update the docstring of _load_azure_openai_context() function to mention that it also checks for OPENAI_API_KEY environment variable if Azure OpenAI API is not available.

4. In the test_gpt_cli.py file, update the import statement to follow PEP8 guidelines:

   from gpt_review._gpt_cli import cli
   import gpt_review.constants as C

5. In the test_openai.py file, update function names and test names to reflect changes from "engine" to "model":

   • Rename get_engine_test() to get_model_test()
   • Rename test_get_engine() to test_get_model()
   • Rename test_int_get_engine() to test_int_get_model()

6. Update comments and docstrings throughout the codebase where "engine" is mentioned but should now be "model".