Clover/models/api_client.py

"""
API client for Clover - A terminal assistant for AI-powered project management
Handles communication with the OpenAI-compatible LLM server
"""

import json
import os
import time
from typing import Any, Dict, Optional

import requests

from config.settings import load_config


class APIClient:
    """
    API client for communicating with the LLM server
    """

    def __init__(self):
        """Initialize the API client with configuration"""
        self.config = load_config()
        self.base_url = self.config.get("base_url", "http://192.168.8.223:11434")
        self.api_key = self.config.get("api_key")

    def _make_request(
        self, endpoint: str, method: str = "GET", data: Optional[Dict] = None
    ) -> Dict[str, Any]:
        """
        Make a request to the LLM API server with retry and exponential backoff.

        Retries up to 3 times with exponential backoff (2s base delay) for
        transient failures (5xx errors, connection errors, timeouts).

        Args:
            endpoint (str): API endpoint
            method (str): HTTP method (GET, POST)
            data (dict): Request data

        Returns:
            dict: Response from the API
        """
        max_retries = self.config.get("max_retries", 3)
        base_delay = self.config.get("retry_base_delay", 2)

        last_exception = None

        for attempt in range(max_retries + 1):
            try:
                # Ensure base_url doesn't have trailing slash and endpoint has leading slash
                base = self.base_url.rstrip("/")
                endpoint = endpoint if endpoint.startswith("/") else f"/{endpoint}"
                url = f"{base}{endpoint}"

                headers = {
                    "Content-Type": "application/json",
                    "User-Agent": "clover-cli/1.0",
                    "Accept": "application/json",
                }

                # Add API key if available
                if self.api_key:
                    headers["Authorization"] = f"Bearer {self.api_key}"

                if method == "GET":
                    response = requests.get(
                        url,
                        headers=headers,
                        timeout=self.config.get("timeout", 300),
                    )
                elif method == "POST":
                    response = requests.post(
                        url,
                        headers=headers,
                        json=data,
                        timeout=self.config.get("timeout", 300),
                    )
                else:
                    raise ValueError(f"Unsupported HTTP method: {method}")

                # Retry on 5xx server errors
                if response.status_code >= 500 and attempt < max_retries:
                    delay = base_delay * (2**attempt)
                    print(
                        f"Server error {response.status_code}, "
                        f"retrying in {delay}s (attempt {attempt + 1}/{max_retries})"
                    )
                    time.sleep(delay)
                    continue

                response.raise_for_status()
                return {"success": True, "data": response.json()}

            except requests.exceptions.Timeout as e:
                last_exception = e
                if attempt < max_retries:
                    delay = base_delay * (2**attempt)
                    print(
                        f"Timeout, retrying in {delay}s "
                        f"(attempt {attempt + 1}/{max_retries})"
                    )
                    time.sleep(delay)
                    continue

            except requests.exceptions.ConnectionError as e:
                last_exception = e
                if attempt < max_retries:
                    delay = base_delay * (2**attempt)
                    print(
                        f"Connection error, retrying in {delay}s "
                        f"(attempt {attempt + 1}/{max_retries})"
                    )
                    time.sleep(delay)
                    continue

            except requests.exceptions.RequestException as e:
                return {"success": False, "error": f"HTTP Request failed: {str(e)}"}
            except json.JSONDecodeError as e:
                return {"success": False, "error": f"Invalid JSON response: {str(e)}"}
            except Exception as e:
                return {"success": False, "error": f"Unexpected error: {str(e)}"}

        # All retries exhausted
        if last_exception:
            return {
                "success": False,
                "error": f"Request failed after {max_retries} retries: {str(last_exception)}",
            }
        return {"success": False, "error": "Request failed after maximum retries"}

    def list_models(self) -> Dict[str, Any]:
        """
        Get list of available models from the LLM server

        Returns:
            dict: Available models information
        """
        # Use Ollama standard endpoint for model listing
        result = self._make_request("/api/tags", "GET")
        if not result["success"]:
            return {"error": result["error"], "models": []}

        try:
            data = result["data"]
            # Handle Ollama format correctly - the response has a "models" key with array
            if isinstance(data, dict) and "models" in data:
                models_list = data["models"]
            else:
                # If it's already just a list of models
                models_list = data if isinstance(data, list) else []

            return {"models": models_list}
        except Exception as e:
            return {"error": f"Failed to parse models response: {str(e)}", "models": []}

    def chat_completion(
        self, messages: list, model: str = None, **kwargs
    ) -> Dict[str, Any]:
        """
        Get a completion from the LLM using Ollama chat endpoint

        Args:
            messages (list): List of message dictionaries (roles and content)
            model (str): Model to use
            **kwargs: Additional parameters for the API

        Returns:
            dict: Response from the LLM
        """
        if model is None:
            # Reload config to get latest model setting
            from config.settings import load_config

            current_config = load_config()
            model = current_config.get("model", "qwen3-coder:30b")

        # Convert messages to prompt format expected by Ollama generate endpoint
        prompt_text = ""
        for message in messages:
            role = message.get("role", "user")
            content = message.get("content", "")

            # Format messages properly for the model
            if role == "system":
                prompt_text += f"System: {content}\n\n"
            elif role == "assistant":
                prompt_text += f"Assistant: {content}\n\n"
            else:  # user
                prompt_text += f"User: {content}\n\n"

        # Add instruction for assistant response
        prompt_text += "Assistant:"

        # Prepare the data for Ollama generate endpoint
        data = {"model": model, "prompt": prompt_text, "stream": False, **kwargs}

        result = self._make_request("/api/generate", "POST", data)
        if not result["success"]:
            return {"error": result["error"]}

        # Extract the response from Ollama's generate format
        try:
            response_data = result["data"]
            # In Ollama generate responses, the actual text is in the "response" field
            if "response" in response_data:
                return {
                    "choices": [{"message": {"content": response_data["response"]}}]
                }
            else:
                # If we get a different format, return what we found
                return response_data
        except Exception as e:
            return {"error": f"Failed to process chat completion result: {str(e)}"}

    def generate_text(self, prompt: str, model: str = None, **kwargs) -> Dict[str, Any]:
        """
        Generate text using the LLM

        Args:
            prompt (str): Prompt to send to the LLM
            model (str): Model to use
            **kwargs: Additional parameters for the API

        Returns:
            dict: Generated response
        """
        if model is None:
            model = self.config.get("model", "qwen3-coder:30b")

        # Format as Ollama generate request
        data = {"model": model, "prompt": prompt, "stream": False, **kwargs}

        result = self._make_request("/api/generate", "POST", data)
        if not result["success"]:
            return {"error": result["error"]}

        try:
            response_data = result["data"]
            # Extract the actual response text for generate endpoint
            if "response" in response_data:
                return {
                    "choices": [{"message": {"content": response_data["response"]}}]
                }
            else:
                # If we got back something else, return it as-is
                return response_data
        except Exception as e:
            return {"error": f"Failed to process generate result: {str(e)}"}

    def get_model_info(self, model_name: str) -> Dict[str, Any]:
        """
        Get information about a specific model

        Args:
            model_name (str): Name of the model

        Returns:
            dict: Model information
        """
        # Use Ollama's show endpoint
        result = self._make_request(f"/api/show/{model_name}", "POST")
        if not result["success"]:
            return {"error": result["error"]}

        return result["data"]