Add File

src/landppt/api/openai_compat.py

@@ -0,0 +1,180 @@
+"""
+OpenAI-compatible API endpoints
+"""
+
+from fastapi import APIRouter, HTTPException, Request
+from fastapi.responses import StreamingResponse
+import json
+import asyncio
+from typing import AsyncGenerator
+
+from .models import (
+    ChatCompletionRequest, ChatCompletionResponse, ChatCompletionChoice,
+    CompletionRequest, CompletionResponse, CompletionChoice,
+    ChatMessage, Usage
+)
+from ..services.ai_service import AIService
+from ..core.config import ai_config
+
+router = APIRouter()
+ai_service = AIService()
+
+@router.post("/chat/completions", response_model=ChatCompletionResponse)
+async def create_chat_completion(request: ChatCompletionRequest):
+    """
+    Create a chat completion (OpenAI compatible)
+    """
+    try:
+        # Check if this is a PPT-related request
+        last_message = request.messages[-1].content if request.messages else ""
+        
+        if ai_service.is_ppt_request(last_message):
+            # Handle PPT generation request
+            response_content = await ai_service.handle_ppt_chat_request(request)
+        else:
+            # Handle general chat request
+            response_content = await ai_service.handle_general_chat_request(request)
+        
+        # Calculate token usage (simplified)
+        prompt_tokens = sum(len(msg.content.split()) for msg in request.messages)
+        completion_tokens = len(response_content.split())
+        
+        choice = ChatCompletionChoice(
+            index=0,
+            message=ChatMessage(role="assistant", content=response_content),
+            finish_reason="stop"
+        )
+        
+        return ChatCompletionResponse(
+            model=request.model,
+            choices=[choice],
+            usage=Usage(
+                prompt_tokens=prompt_tokens,
+                completion_tokens=completion_tokens,
+                total_tokens=prompt_tokens + completion_tokens
+            )
+        )
+        
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error generating completion: {str(e)}")
+
+@router.post("/completions", response_model=CompletionResponse)
+async def create_completion(request: CompletionRequest):
+    """
+    Create a text completion (OpenAI compatible)
+    """
+    try:
+        prompt = request.prompt if isinstance(request.prompt, str) else request.prompt[0]
+        
+        if ai_service.is_ppt_request(prompt):
+            # Handle PPT generation request
+            response_text = await ai_service.handle_ppt_completion_request(request)
+        else:
+            # Handle general completion request
+            response_text = await ai_service.handle_general_completion_request(request)
+        
+        # Calculate token usage (simplified)
+        prompt_tokens = len(prompt.split())
+        completion_tokens = len(response_text.split())
+        
+        choice = CompletionChoice(
+            text=response_text,
+            index=0,
+            finish_reason="stop"
+        )
+        
+        return CompletionResponse(
+            model=request.model,
+            choices=[choice],
+            usage=Usage(
+                prompt_tokens=prompt_tokens,
+                completion_tokens=completion_tokens,
+                total_tokens=prompt_tokens + completion_tokens
+            )
+        )
+        
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error generating completion: {str(e)}")
+
+@router.get("/models")
+async def list_models():
+    """
+    List available models (OpenAI compatible)
+    """
+    return {
+        "object": "list",
+        "data": [
+            {
+                "id": "landppt-v1",
+                "object": "model",
+                "created": 1677610602,
+                "owned_by": "landppt",
+                "permission": [],
+                "root": "landppt-v1",
+                "parent": None
+            },
+            {
+                "id": "landppt-ppt-generator",
+                "object": "model", 
+                "created": 1677610602,
+                "owned_by": "landppt",
+                "permission": [],
+                "root": "landppt-ppt-generator",
+                "parent": None
+            }
+        ]
+    }
+
+async def stream_chat_completion(request: ChatCompletionRequest) -> AsyncGenerator[str, None]:
+    """
+    Stream chat completion responses
+    """
+    try:
+        # Simulate streaming response
+        last_message = request.messages[-1].content if request.messages else ""
+        
+        if ai_service.is_ppt_request(last_message):
+            response_content = await ai_service.handle_ppt_chat_request(request)
+        else:
+            response_content = await ai_service.handle_general_chat_request(request)
+        
+        # Split response into chunks for streaming
+        words = response_content.split()
+        for i, word in enumerate(words):
+            chunk_data = {
+                "id": f"chatcmpl-{i}",
+                "object": "chat.completion.chunk",
+                "created": 1677610602,
+                "model": request.model,
+                "choices": [{
+                    "index": 0,
+                    "delta": {"content": word + " "},
+                    "finish_reason": None
+                }]
+            }
+            yield f"data: {json.dumps(chunk_data)}\n\n"
+            await asyncio.sleep(0.05)  # Simulate processing delay
+        
+        # Send final chunk
+        final_chunk = {
+            "id": f"chatcmpl-final",
+            "object": "chat.completion.chunk", 
+            "created": 1677610602,
+            "model": request.model,
+            "choices": [{
+                "index": 0,
+                "delta": {},
+                "finish_reason": "stop"
+            }]
+        }
+        yield f"data: {json.dumps(final_chunk)}\n\n"
+        yield "data: [DONE]\n\n"
+        
+    except Exception as e:
+        error_chunk = {
+            "error": {
+                "message": str(e),
+                "type": "server_error"
+            }
+        }
+        yield f"data: {json.dumps(error_chunk)}\n\n"