The configuration system in Codegate is managed through the Config class in config.py. It supports multiple configuration sources with a clear priority order.

1. CLI arguments
2. Environment variables
3. Config file (YAML)
4. Default values (including default prompts from prompts/default.yaml)

• Port: 8989
• Host: "localhost"
• Log Level: "INFO"
• Log Format: "JSON"
• Prompts: Default prompts from prompts/default.yaml
• Provider URLs:
  • vLLM: "http://localhost:8000"
  • OpenAI: "https://api.openai.com/v1"
  • Anthropic: "https://api.anthropic.com/v1"

Load configuration from a YAML file:

config = Config.from_file("config.yaml")

Example config.yaml:

port: 8989
host: localhost
log_level: INFO
log_format: JSON
provider_urls:
  vllm: "https://vllm.example.com"
  openai: "https://api.openai.com/v1"
  anthropic: "https://api.anthropic.com/v1"

Environment variables are automatically loaded with these mappings:

• CODEGATE_APP_PORT: Server port
• CODEGATE_APP_HOST: Server host
• CODEGATE_APP_LOG_LEVEL: Logging level
• CODEGATE_LOG_FORMAT: Log format
• CODEGATE_PROMPTS_FILE: Path to prompts YAML file
• CODEGATE_PROVIDER_VLLM_URL: vLLM provider URL
• CODEGATE_PROVIDER_OPENAI_URL: OpenAI provider URL
• CODEGATE_PROVIDER_ANTHROPIC_URL: Anthropic provider URL

config = Config.from_env()

Provider URLs can be configured in several ways:

1. In Configuration File:

   provider_urls:
     vllm: "https://vllm.example.com"  # /v1 path is added automatically
     openai: "https://api.openai.com/v1"
     anthropic: "https://api.anthropic.com/v1"

2. Via Environment Variables:

   export CODEGATE_PROVIDER_VLLM_URL=https://vllm.example.com
   export CODEGATE_PROVIDER_OPENAI_URL=https://api.openai.com/v1
   export CODEGATE_PROVIDER_ANTHROPIC_URL=https://api.anthropic.com/v1

3. Via CLI Flags:

   codegate serve --vllm-url https://vllm.example.com

Note: For the vLLM provider, the /v1 path is automatically appended to the base URL if not present.

Available log levels (case-insensitive):

• ERROR
• WARNING
• INFO
• DEBUG

Available log formats (case-insensitive):

• JSON
• TEXT

Prompts can be configured in several ways:

1. Default Prompts:

   • Located in prompts/default.yaml
   • Loaded automatically if no other prompts are specified

2. In Configuration File:

   # Option 1: Direct prompts definition
   prompts:
     my_prompt: "Custom prompt text"
     another_prompt: "Another prompt text"
   
   # Option 2: Reference to prompts file
   prompts: "path/to/prompts.yaml"

3. Via Environment Variable:

   export CODEGATE_PROMPTS_FILE=path/to/prompts.yaml

4. Via CLI Flag:

   codegate serve --prompts path/to/prompts.yaml

Prompts files should be in YAML format with string values:

prompt_name: "Prompt text content"
another_prompt: "More prompt text"

Access prompts in code:

config = Config.load()
prompt = config.prompts.prompt_name

The configuration system uses a custom ConfigurationError exception for handling configuration-related errors, such as:

• Invalid port numbers (must be between 1 and 65535)
• Invalid log levels
• Invalid log formats
• YAML parsing errors
• File reading errors
• Invalid prompt values (must be strings)
• Missing or invalid prompts files