mirror of
https://github.com/Alishahryar1/free-claude-code.git
synced 2026-04-28 19:40:54 +00:00
110 lines
3.4 KiB
Python
110 lines
3.4 KiB
Python
"""FastAPI route handlers."""
|
|
|
|
import logging
|
|
import uuid
|
|
|
|
from fastapi import APIRouter, Request, Depends, HTTPException
|
|
from fastapi.responses import StreamingResponse
|
|
|
|
from .models.anthropic import MessagesRequest, TokenCountRequest
|
|
from .models.responses import MessagesResponse, TokenCountResponse
|
|
from .dependencies import get_provider, get_settings
|
|
from .request_utils import get_token_count
|
|
from .optimization_handlers import try_optimizations
|
|
from config.settings import Settings
|
|
from providers.base import BaseProvider
|
|
from providers.exceptions import ProviderError
|
|
from providers.logging_utils import log_request_compact
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
router = APIRouter()
|
|
|
|
|
|
# =============================================================================
|
|
# Routes
|
|
# =============================================================================
|
|
|
|
|
|
@router.post("/v1/messages")
|
|
async def create_message(
|
|
request_data: MessagesRequest,
|
|
raw_request: Request,
|
|
provider: BaseProvider = Depends(get_provider),
|
|
settings: Settings = Depends(get_settings),
|
|
):
|
|
"""Create a message (always streaming)."""
|
|
|
|
try:
|
|
optimized = try_optimizations(request_data, settings)
|
|
if optimized is not None:
|
|
return optimized
|
|
|
|
request_id = f"req_{uuid.uuid4().hex[:12]}"
|
|
log_request_compact(logger, request_id, request_data)
|
|
|
|
input_tokens = get_token_count(
|
|
request_data.messages, request_data.system, request_data.tools
|
|
)
|
|
return StreamingResponse(
|
|
provider.stream_response(request_data, input_tokens=input_tokens),
|
|
media_type="text/event-stream",
|
|
headers={
|
|
"X-Accel-Buffering": "no",
|
|
"Cache-Control": "no-cache",
|
|
"Connection": "keep-alive",
|
|
},
|
|
)
|
|
|
|
except ProviderError:
|
|
raise
|
|
except Exception as e:
|
|
import traceback
|
|
|
|
logger.error(f"Error: {str(e)}\n{traceback.format_exc()}")
|
|
raise HTTPException(status_code=getattr(e, "status_code", 500), detail=str(e))
|
|
|
|
|
|
@router.post("/v1/messages/count_tokens")
|
|
async def count_tokens(request_data: TokenCountRequest):
|
|
"""Count tokens for a request."""
|
|
try:
|
|
return TokenCountResponse(
|
|
input_tokens=get_token_count(
|
|
request_data.messages, request_data.system, request_data.tools
|
|
)
|
|
)
|
|
except Exception as e:
|
|
raise HTTPException(status_code=500, detail=str(e))
|
|
|
|
|
|
@router.get("/")
|
|
async def root(settings: Settings = Depends(get_settings)):
|
|
"""Root endpoint."""
|
|
return {
|
|
"status": "ok",
|
|
"provider": "nvidia_nim",
|
|
"model": settings.model,
|
|
}
|
|
|
|
|
|
@router.get("/health")
|
|
async def health():
|
|
"""Health check endpoint."""
|
|
return {"status": "healthy"}
|
|
|
|
|
|
@router.post("/stop")
|
|
async def stop_cli(request: Request):
|
|
"""Stop all CLI sessions and pending tasks."""
|
|
handler = getattr(request.app.state, "message_handler", None)
|
|
if not handler:
|
|
# Fallback if messaging not initialized
|
|
cli_manager = getattr(request.app.state, "cli_manager", None)
|
|
if cli_manager:
|
|
await cli_manager.stop_all()
|
|
return {"status": "stopped", "source": "cli_manager"}
|
|
raise HTTPException(status_code=503, detail="Messaging system not initialized")
|
|
|
|
count = await handler.stop_all_tasks()
|
|
return {"status": "stopped", "cancelled_count": count}
|