Update app.py

app.py

@@ -1,13 +1,13 @@
 """
 ╔══════════════════════════════════════════════════════════════════╗
-║         Universal AI Gateway  –  Production v6.0                ║
-║                                                                  ║
-║  • كل نموذج مربوط بـ provider يعمل فعلاً (لا auto عشوائي)       ║
-║  • عائلة Cohere كاملة عبر HuggingSpace (بدون auth)              ║
-║  • Kimi عبر Jmuz                                                 ║
-║  • GPT-4 عبر Jmuz / Liaobots / PollinationsAI                   ║
-║  • Streaming حقيقي بدون MissingAuthError                         ║
-║  • Fallback chain مرتب لكل نموذج                                ║
+║         Universal AI Gateway  –  Production v6.0               ║
+║                                                                ║
+║  • كل نموذج مربوط بـ provider يعمل فعلاً (لا auto عشوائي)     ║
+║  • عائلة Cohere كاملة عبر HuggingSpace (بدون auth)            ║
+║  • Kimi عبر Jmuz                                               ║
+║  • GPT-4 عبر Jmuz / Liaobots / PollinationsAI                 ║
+║  • Streaming حقيقي بدون MissingAuthError                      ║
+║  • Fallback chain مرتب لكل نموذج                              ║
 ╚══════════════════════════════════════════════════════════════════╝
 """
 
@@ -18,44 +18,49 @@ import json
 import logging
 import time
 import uuid
+
 from contextlib import asynccontextmanager
 from typing import Any, Dict, List, Optional
 
 import g4f
+
 from fastapi import FastAPI, HTTPException, Request
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import JSONResponse, StreamingResponse
+
 from g4f import Provider
 from g4f.client import Client
+
 from pydantic import BaseModel
 
 # ──────────────────────────────────────────────────────────────────
 # LOGGING
 # ──────────────────────────────────────────────────────────────────
+
 logging.basicConfig(
     level=logging.INFO,
     format="%(asctime)s | %(levelname)-7s | %(message)s",
     datefmt="%H:%M:%S",
 )
+
 log = logging.getLogger("gw")
 
 # ──────────────────────────────────────────────────────────────────
 # CONFIG
 # ──────────────────────────────────────────────────────────────────
-API_KEY       = "sk-your-secret-key"
-CALL_TIMEOUT  = 60   # seconds – non-streaming
-STREAM_TIMEOUT = 90  # seconds – streaming
 
+API_KEY = "sk-your-secret-key"
+
+CALL_TIMEOUT = 60
+STREAM_TIMEOUT = 90
 
 # ══════════════════════════════════════════════════════════════════
-#  MODEL → PROVIDER ROUTING TABLE
-#
-#  قائمة مرتبة: أول provider ينجح يُستخدم، الباقي fallback.
-#  لا يوجد أي provider يحتاج API key هنا.
+# MODEL → PROVIDER ROUTING TABLE
 # ══════════════════════════════════════════════════════════════════
+
 ROUTING: Dict[str, List[Any]] = {
 
-    # ── Cohere family (HuggingSpace يستضيف spaces رسمية لـ Cohere) ─
+    # ── Cohere ────────────────────────────────────────────────
     "command-r":            [Provider.HuggingSpace, Provider.Jmuz],
     "command-r-plus":       [Provider.HuggingSpace, Provider.Jmuz],
     "command-r7b":          [Provider.HuggingSpace],
@@ -67,56 +72,56 @@ ROUTING: Dict[str, List[Any]] = {
     "c4ai-aya-expanse-8b":  [Provider.HuggingSpace],
     "c4ai-aya-expanse-32b": [Provider.HuggingSpace],
 
-    # ── Kimi / Moonshot ───────────────────────────────────────────
+    # ── Kimi / Moonshot ──────────────────────────────────────
     "kimi":             [Provider.Jmuz],
     "moonshot-v1-8k":   [Provider.Jmuz],
     "moonshot-v1-32k":  [Provider.Jmuz],
     "moonshot-v1-128k": [Provider.Jmuz],
 
-    # ── GPT-4 family ──────────────────────────────────────────────
+    # ── GPT ──────────────────────────────────────────────────
     "gpt-4":         [Provider.Jmuz, Provider.Liaobots, Provider.PollinationsAI],
     "gpt-4-turbo":   [Provider.Jmuz, Provider.Liaobots, Provider.PollinationsAI],
     "gpt-4o":        [Provider.PollinationsAI, Provider.Jmuz, Provider.Liaobots],
     "gpt-4o-mini":   [Provider.PollinationsAI, Provider.DDG, Provider.Jmuz],
     "gpt-3.5-turbo": [Provider.DDG, Provider.Jmuz, Provider.PollinationsAI],
 
-    # ── DeepSeek ──────────────────────────────────────────────────
+    # ── DeepSeek ─────────────────────────────────────────────
     "deepseek-chat": [Provider.PollinationsAI, Provider.Jmuz],
     "deepseek-r1":   [Provider.PollinationsAI, Provider.Jmuz, Provider.DDG],
     "deepseek-v3":   [Provider.PollinationsAI, Provider.Jmuz],
 
-    # ── Llama ─────────────────────────────────────────────────────
-    "llama-3.1-8b":    [Provider.PollinationsAI, Provider.Jmuz, Provider.DDG],
-    "llama-3.1-70b":   [Provider.PollinationsAI, Provider.Jmuz],
-    "llama-3.3-70b":   [Provider.PollinationsAI, Provider.Jmuz],
-    "llama-3.2-11b":   [Provider.PollinationsAI],
-    "llama-4-scout":   [Provider.PollinationsAI],
-    "llama-4-maverick":[Provider.PollinationsAI],
+    # ── Llama ────────────────────────────────────────────────
+    "llama-3.1-8b":     [Provider.PollinationsAI, Provider.Jmuz, Provider.DDG],
+    "llama-3.1-70b":    [Provider.PollinationsAI, Provider.Jmuz],
+    "llama-3.3-70b":    [Provider.PollinationsAI, Provider.Jmuz],
+    "llama-3.2-11b":    [Provider.PollinationsAI],
+    "llama-4-scout":    [Provider.PollinationsAI],
+    "llama-4-maverick": [Provider.PollinationsAI],
 
-    # ── Mistral ───────────────────────────────────────────────────
+    # ── Mistral ──────────────────────────────────────────────
     "mistral-7b":    [Provider.PollinationsAI, Provider.Jmuz],
     "mixtral-8x7b":  [Provider.PollinationsAI, Provider.Jmuz],
     "mistral-small": [Provider.PollinationsAI],
     "mistral-large": [Provider.PollinationsAI, Provider.Jmuz],
 
-    # ── Gemini ────────────────────────────────────────────────────
+    # ── Gemini ───────────────────────────────────────────────
     "gemini-2.0-flash": [Provider.PollinationsAI, Provider.Jmuz],
     "gemini-1.5-flash": [Provider.PollinationsAI, Provider.Jmuz],
     "gemini-1.5-pro":   [Provider.PollinationsAI, Provider.Jmuz],
     "gemini-pro":       [Provider.PollinationsAI, Provider.Jmuz],
 
-    # ── Qwen ──────────────────────────────────────────────────────
+    # ── Qwen ─────────────────────────────────────────────────
     "qwen-2.5-72b":       [Provider.PollinationsAI, Provider.Jmuz],
     "qwen-2.5-coder-32b": [Provider.PollinationsAI],
     "qwq-32b":            [Provider.PollinationsAI, Provider.Jmuz],
 
-    # ── Claude (عبر Jmuz proxy — لا حاجة لـ Anthropic key) ───────
-    "claude-3-haiku":   [Provider.Jmuz, Provider.Liaobots],
-    "claude-3-sonnet":  [Provider.Jmuz, Provider.Liaobots],
-    "claude-3-opus":    [Provider.Jmuz, Provider.Liaobots],
-    "claude-3.5-sonnet":[Provider.Jmuz, Provider.Liaobots],
+    # ── Claude ───────────────────────────────────────────────
+    "claude-3-haiku":    [Provider.Jmuz, Provider.Liaobots],
+    "claude-3-sonnet":   [Provider.Jmuz, Provider.Liaobots],
+    "claude-3-opus":     [Provider.Jmuz, Provider.Liaobots],
+    "claude-3.5-sonnet": [Provider.Jmuz, Provider.Liaobots],
 
-    # ── Other ─────────────────────────────────────────────────────
+    # ─��� Other ────────────────────────────────────────────────
     "phi-4":     [Provider.PollinationsAI],
     "sonar-pro": [Provider.PollinationsAI, Provider.Jmuz],
     "sonar":     [Provider.PollinationsAI, Provider.Jmuz],
@@ -124,194 +129,419 @@ ROUTING: Dict[str, List[Any]] = {
 
 ALL_MODELS: List[str] = sorted(ROUTING.keys())
 
-
 # ══════════════════════════════════════════════════════════════════
-#  CORE CALL  (sync – runs in thread)
+# CORE CALL
 # ══════════════════════════════════════════════════════════════════
+
 def _call(provider_cls: Any, model: str, messages: list, stream: bool) -> Any:
+
     client = Client(provider=provider_cls)
+
     return client.chat.completions.create(
         model=model,
         messages=messages,
         stream=stream,
     )
 
-
 # ══════════════════════════════════════════════════════════════════
-#  SMART COMPLETION  (async wrapper with fallback chain)
+# SMART COMPLETION
 # ══════════════════════════════════════════════════════════════════
-async def smart_completion(model: str, messages: list, stream: bool = False) -> Any:
+
+async def smart_completion(
+    model: str,
+    messages: list,
+    stream: bool = False
+) -> Any:
+
     chain = ROUTING.get(model)
+
     if not chain:
-        log.warning(f"Unknown model '{model}' – using generic fallback chain")
-        chain = [Provider.PollinationsAI, Provider.Jmuz, Provider.DDG]
+        log.warning(
+            f"Unknown model '{model}' – using generic fallback chain"
+        )
+        chain = [
+            Provider.PollinationsAI,
+            Provider.Jmuz,
+            Provider.DDG,
+        ]
 
     timeout = STREAM_TIMEOUT if stream else CALL_TIMEOUT
+
     errors: List[str] = []
 
     for provider_cls in chain:
-        pname = getattr(provider_cls, "__name__", str(provider_cls))
+
+        pname = getattr(
+            provider_cls,
+            "__name__",
+            str(provider_cls)
+        )
+
         try:
-            log.info(f"  -> {pname} | model={model} | stream={stream}")
+
+            log.info(
+                f"  -> {pname} | model={model} | stream={stream}"
+            )
+
             resp = await asyncio.wait_for(
-                asyncio.to_thread(_call, provider_cls, model, messages, stream),
+                asyncio.to_thread(
+                    _call,
+                    provider_cls,
+                    model,
+                    messages,
+                    stream,
+                ),
                 timeout=timeout,
             )
+
             log.info(f"  OK {pname}")
+
             return resp
+
         except asyncio.TimeoutError:
+
             msg = f"{pname}: timeout after {timeout}s"
+
         except Exception as exc:
+
             msg = f"{pname}: {type(exc).__name__}: {exc}"
+
         log.warning(f"  FAIL {msg}")
-        errors.append(msg)
 
-    raise RuntimeError(f"All providers failed for '{model}':\n" + "\n".join(errors))
+        errors.append(msg)
 
+    raise RuntimeError(
+        f"All providers failed for '{model}':\n"
+        + "\n".join(errors)
+    )
 
 # ══════════════════════════════════════════════════════════════════
-#  STREAMING GENERATOR
+# STREAMING GENERATOR
 # ══════════════════════════════════════════════════════════════════
+
 async def sse_generator(model: str, messages: list):
-    cid     = f"chatcmpl-{uuid.uuid4().hex}"
+
+    cid = f"chatcmpl-{uuid.uuid4().hex}"
     created = int(time.time())
-    sent    = False
+
+    sent = False
 
     try:
-        response = await smart_completion(model, messages, stream=True)
+
+        response = await smart_completion(
+            model,
+            messages,
+            stream=True,
+        )
+
         for chunk in response:
+
             try:
-                content = chunk.choices[0].delta.content or ""
+
+                content = (
+                    chunk.choices[0].delta.content or ""
+                )
+
             except Exception:
+
                 content = ""
+
             if not content:
                 continue
+
             sent = True
-            yield "data: " + json.dumps({
-                "id": cid, "object": "chat.completion.chunk",
-                "created": created, "model": model,
-                "choices": [{"index": 0, "delta": {"content": content}, "finish_reason": None}],
-            }, ensure_ascii=False) + "\n\n"
+
+            yield (
+                "data: "
+                + json.dumps(
+                    {
+                        "id": cid,
+                        "object": "chat.completion.chunk",
+                        "created": created,
+                        "model": model,
+                        "choices": [
+                            {
+                                "index": 0,
+                                "delta": {
+                                    "content": content
+                                },
+                                "finish_reason": None,
+                            }
+                        ],
+                    },
+                    ensure_ascii=False,
+                )
+                + "\n\n"
+            )
+
             await asyncio.sleep(0)
 
         if not sent:
-            yield "data: " + json.dumps(
-                {"error": {"message": "Provider returned empty stream", "type": "empty_stream"}}
-            ) + "\n\n"
+
+            yield (
+                "data: "
+                + json.dumps(
+                    {
+                        "error": {
+                            "message": "Provider returned empty stream",
+                            "type": "empty_stream",
+                        }
+                    }
+                )
+                + "\n\n"
+            )
 
         # stop chunk
-        yield "data: " + json.dumps({
-            "id": cid, "object": "chat.completion.chunk",
-            "created": created, "model": model,
-            "choices": [{"index": 0, "delta": {}, "finish_reason": "stop"}],
-        }) + "\n\n"
+        yield (
+            "data: "
+            + json.dumps(
+                {
+                    "id": cid,
+                    "object": "chat.completion.chunk",
+                    "created": created,
+                    "model": model,
+                    "choices": [
+                        {
+                            "index": 0,
+                            "delta": {},
+                            "finish_reason": "stop",
+                        }
+                    ],
+                }
+            )
+            + "\n\n"
+        )
+
         yield "data: [DONE]\n\n"
 
     except Exception as exc:
+
         log.error(f"Stream error: {exc}")
-        yield "data: " + json.dumps(
-            {"error": {"message": str(exc), "type": "server_error"}}
-        ) + "\n\n"
-        yield "data: [DONE]\n\n"
 
+        yield (
+            "data: "
+            + json.dumps(
+                {
+                    "error": {
+                        "message": str(exc),
+                        "type": "server_error",
+                    }
+                }
+            )
+            + "\n\n"
+        )
+
+        yield "data: [DONE]\n\n"
 
 # ══════════════════════════════════════════════════════════════════
-#  FASTAPI
+# FASTAPI
 # ══════════════════════════════════════════════════════════════════
+
 @asynccontextmanager
 async def lifespan(app: FastAPI):
-    log.info(f"Gateway ready — {len(ALL_MODELS)} models | providers pinned (no auto)")
+
+    log.info(
+        f"Gateway ready — {len(ALL_MODELS)} models | providers pinned (no auto)"
+    )
+
     yield
 
 
-app = FastAPI(title="Universal AI Gateway", version="6.0.0", lifespan=lifespan)
+app = FastAPI(
+    title="Universal AI Gateway",
+    version="6.0.0",
+    lifespan=lifespan,
+)
+
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"], allow_credentials=True,
-    allow_methods=["*"], allow_headers=["*"],
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
 )
 
+# ══════════════════════════════════════════════════════════════════
+# Pydantic Models
+# ══════════════════════════════════════════════════════════════════
 
 class Message(BaseModel):
     role: str
     content: str
 
+
 class ChatRequest(BaseModel):
     model: str
     messages: List[Message]
     stream: bool = False
     temperature: Optional[float] = 0.7
-    max_tokens:  Optional[int]   = 4096
+    max_tokens: Optional[int] = 4096
 
+# ══════════════════════════════════════════════════════════════════
+# AUTH
+# ════════════════════════���═════════════════════════════════════════
 
 def _auth(req: Request):
+
     auth = req.headers.get("Authorization", "")
+
     if not auth:
         return
+
     if not auth.startswith("Bearer "):
-        raise HTTPException(401, "Invalid Authorization format")
+        raise HTTPException(
+            401,
+            "Invalid Authorization format"
+        )
+
     if auth.removeprefix("Bearer ").strip() != API_KEY:
-        raise HTTPException(403, "Invalid API key")
+        raise HTTPException(
+            403,
+            "Invalid API key"
+        )
 
+# ══════════════════════════════════════════════════════════════════
+# ROOT
+# ══════════════════════════════════════════════════════════════════
 
 @app.get("/")
 async def root():
+
     return {
         "service": "Universal AI Gateway",
         "version": "6.0.0",
-        "models":  len(ALL_MODELS),
-        "docs":    "/docs",
+        "models": len(ALL_MODELS),
+        "docs": "/docs",
     }
 
+# ══════════════════════════════════════════════════════════════════
+# MODELS ENDPOINT
+# ══════════════════════════════════════════════════════════════════
 
 @app.get("/v1/models")
 async def get_models(req: Request):
+
     _auth(req)
+
     now = int(time.time())
+
     return {
         "object": "list",
         "data": [
             {
-                "id": m, "object": "model", "created": now, "owned_by": "g4f",
-                "providers": [getattr(p, "__name__", str(p)) for p in ROUTING.get(m, [])],
+                "id": m,
+                "object": "model",
+                "created": now,
+                "owned_by": "g4f",
+                "providers": [
+                    getattr(p, "__name__", str(p))
+                    for p in ROUTING.get(m, [])
+                ],
             }
             for m in ALL_MODELS
         ],
     }
 
+# ══════════════════════════════════════════════════════════════════
+# CHAT COMPLETIONS
+# ══════════════════════════════════════════════════════════════════
 
 @app.post("/v1/chat/completions")
-async def chat_completions(req: Request, body: ChatRequest):
+async def chat_completions(
+    req: Request,
+    body: ChatRequest
+):
+
     _auth(req)
-    messages = [{"role": m.role, "content": m.content} for m in body.messages]
-    log.info(f"Request model={body.model} stream={body.stream}")
 
+    messages = [
+        {
+            "role": m.role,
+            "content": m.content,
+        }
+        for m in body.messages
+    ]
+
+    log.info(
+        f"Request model={body.model} stream={body.stream}"
+    )
+
+    # STREAMING
     if body.stream:
+
         return StreamingResponse(
             sse_generator(body.model, messages),
             media_type="text/event-stream",
-            headers={"Cache-Control": "no-cache", "Connection": "keep-alive", "X-Accel-Buffering": "no"},
+            headers={
+                "Cache-Control": "no-cache",
+                "Connection": "keep-alive",
+                "X-Accel-Buffering": "no",
+            },
         )
 
+    # NORMAL RESPONSE
     try:
-        response = await smart_completion(body.model, messages, stream=False)
+
+        response = await smart_completion(
+            body.model,
+            messages,
+            stream=False,
+        )
+
         try:
-            content = response.choices[0].message.content
+
+            content = (
+                response.choices[0].message.content
+            )
+
         except Exception:
+
             content = str(response)
-        return JSONResponse({
-            "id": f"chatcmpl-{uuid.uuid4().hex}",
-            "object": "chat.completion",
-            "created": int(time.time()),
-            "model": body.model,
-            "choices": [{"index": 0, "message": {"role": "assistant", "content": content}, "finish_reason": "stop"}],
-            "usage": {"prompt_tokens": 0, "completion_tokens": 0, "total_tokens": 0},
-        })
+
+        return JSONResponse(
+            {
+                "id": f"chatcmpl-{uuid.uuid4().hex}",
+                "object": "chat.completion",
+                "created": int(time.time()),
+                "model": body.model,
+                "choices": [
+                    {
+                        "index": 0,
+                        "message": {
+                            "role": "assistant",
+                            "content": content,
+                        },
+                        "finish_reason": "stop",
+                    }
+                ],
+                "usage": {
+                    "prompt_tokens": 0,
+                    "completion_tokens": 0,
+                    "total_tokens": 0,
+                },
+            }
+        )
+
     except Exception as exc:
+
         log.error(f"Error: {exc}")
-        raise HTTPException(500, str(exc))
 
+        raise HTTPException(
+            500,
+            str(exc)
+        )
+
+# ══════════════════════════════════════════════════════════════════
+# RUN
+# ══════════════════════════════════════════════════════════════════
 
 if __name__ == "__main__":
+
     import uvicorn
-    uvicorn.run(app, host="0.0.0.0", port=7860, log_level="info")
+
+    uvicorn.run(
+        app,
+        host="0.0.0.0",
+        port=7860,
+        log_level="info",
+    )