File size: 7,830 Bytes
6172a47 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 | """FastAPI route handlers."""
import traceback
import uuid
from fastapi import APIRouter, Depends, HTTPException, Request, Response
from fastapi.responses import StreamingResponse
from loguru import logger
from config.settings import Settings
from providers.common import get_user_facing_error_message
from providers.exceptions import InvalidRequestError, ProviderError
from .dependencies import get_provider_for_type, get_settings, require_api_key
from .models.anthropic import MessagesRequest, TokenCountRequest
from .models.responses import ModelResponse, ModelsListResponse, TokenCountResponse
from .optimization_handlers import try_optimizations
from .request_utils import get_token_count
router = APIRouter()
SUPPORTED_CLAUDE_MODELS = [
ModelResponse(
id="claude-opus-4-20250514",
display_name="Claude Opus 4",
created_at="2025-05-14T00:00:00Z",
),
ModelResponse(
id="claude-sonnet-4-20250514",
display_name="Claude Sonnet 4",
created_at="2025-05-14T00:00:00Z",
),
ModelResponse(
id="claude-haiku-4-20250514",
display_name="Claude Haiku 4",
created_at="2025-05-14T00:00:00Z",
),
ModelResponse(
id="claude-3-opus-20240229",
display_name="Claude 3 Opus",
created_at="2024-02-29T00:00:00Z",
),
ModelResponse(
id="claude-3-5-sonnet-20241022",
display_name="Claude 3.5 Sonnet",
created_at="2024-10-22T00:00:00Z",
),
ModelResponse(
id="claude-3-haiku-20240307",
display_name="Claude 3 Haiku",
created_at="2024-03-07T00:00:00Z",
),
ModelResponse(
id="claude-3-5-haiku-20241022",
display_name="Claude 3.5 Haiku",
created_at="2024-10-22T00:00:00Z",
),
]
def _probe_response(allow: str) -> Response:
"""Return an empty success response for compatibility probes."""
return Response(status_code=204, headers={"Allow": allow})
# =============================================================================
# Routes
# =============================================================================
@router.post("/v1/messages")
async def create_message(
request_data: MessagesRequest,
raw_request: Request,
settings: Settings = Depends(get_settings),
_auth=Depends(require_api_key),
):
"""Create a message (always streaming)."""
try:
if not request_data.messages:
raise InvalidRequestError("messages cannot be empty")
optimized = try_optimizations(request_data, settings)
if optimized is not None:
return optimized
logger.debug("No optimization matched, routing to provider")
# Resolve provider from the model-aware mapping
provider_type = Settings.parse_provider_type(
request_data.resolved_provider_model or settings.model
)
provider = get_provider_for_type(provider_type)
request_id = f"req_{uuid.uuid4().hex[:12]}"
logger.info(
"API_REQUEST: request_id={} model={} messages={}",
request_id,
request_data.model,
len(request_data.messages),
)
logger.debug("FULL_PAYLOAD [{}]: {}", request_id, request_data.model_dump())
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,
request_id=request_id,
),
media_type="text/event-stream",
headers={
"X-Accel-Buffering": "no",
"Cache-Control": "no-cache",
"Connection": "keep-alive",
},
)
except ProviderError:
raise
except Exception as e:
logger.error(f"Error: {e!s}\n{traceback.format_exc()}")
raise HTTPException(
status_code=getattr(e, "status_code", 500),
detail=get_user_facing_error_message(e),
) from e
@router.api_route("/v1/messages", methods=["HEAD", "OPTIONS"])
async def probe_messages(_auth=Depends(require_api_key)):
"""Respond to Claude compatibility probes for the messages endpoint."""
return _probe_response("POST, HEAD, OPTIONS")
@router.post("/v1/messages/count_tokens")
async def count_tokens(request_data: TokenCountRequest, _auth=Depends(require_api_key)):
"""Count tokens for a request."""
request_id = f"req_{uuid.uuid4().hex[:12]}"
with logger.contextualize(request_id=request_id):
try:
tokens = get_token_count(
request_data.messages, request_data.system, request_data.tools
)
logger.info(
"COUNT_TOKENS: request_id={} model={} messages={} input_tokens={}",
request_id,
getattr(request_data, "model", "unknown"),
len(request_data.messages),
tokens,
)
return TokenCountResponse(input_tokens=tokens)
except Exception as e:
logger.error(
"COUNT_TOKENS_ERROR: request_id={} error={}\n{}",
request_id,
get_user_facing_error_message(e),
traceback.format_exc(),
)
raise HTTPException(
status_code=500, detail=get_user_facing_error_message(e)
) from e
@router.api_route("/v1/messages/count_tokens", methods=["HEAD", "OPTIONS"])
async def probe_count_tokens(_auth=Depends(require_api_key)):
"""Respond to Claude compatibility probes for the token count endpoint."""
return _probe_response("POST, HEAD, OPTIONS")
@router.get("/")
async def root(
settings: Settings = Depends(get_settings), _auth=Depends(require_api_key)
):
"""Root endpoint."""
return {
"status": "ok",
"provider": settings.provider_type,
"model": settings.model,
}
@router.api_route("/", methods=["HEAD", "OPTIONS"])
async def probe_root(_auth=Depends(require_api_key)):
"""Respond to compatibility probes for the root endpoint."""
return _probe_response("GET, HEAD, OPTIONS")
@router.get("/health")
async def health():
"""Health check endpoint."""
return {"status": "healthy"}
@router.api_route("/health", methods=["HEAD", "OPTIONS"])
async def probe_health():
"""Respond to compatibility probes for the health endpoint."""
return _probe_response("GET, HEAD, OPTIONS")
@router.get("/v1/models", response_model=ModelsListResponse)
async def list_models(_auth=Depends(require_api_key)):
"""List the Claude model ids this proxy advertises for compatibility."""
return ModelsListResponse(
data=SUPPORTED_CLAUDE_MODELS,
first_id=SUPPORTED_CLAUDE_MODELS[0].id if SUPPORTED_CLAUDE_MODELS else None,
has_more=False,
last_id=SUPPORTED_CLAUDE_MODELS[-1].id if SUPPORTED_CLAUDE_MODELS else None,
)
@router.post("/stop")
async def stop_cli(request: Request, _auth=Depends(require_api_key)):
"""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()
logger.info("STOP_CLI: source=cli_manager cancelled_count=N/A")
return {"status": "stopped", "source": "cli_manager"}
raise HTTPException(status_code=503, detail="Messaging system not initialized")
count = await handler.stop_all_tasks()
logger.info("STOP_CLI: source=handler cancelled_count={}", count)
return {"status": "stopped", "cancelled_count": count}
|