add api for post browser-use tasks

This commit is contained in:
Кобылкевич Фёдор 2026-04-07 22:01:53 +03:00
parent 890d492de0
commit d277955a9a
11 changed files with 320 additions and 8 deletions

View file

@ -14,3 +14,9 @@ TELEGRAM_HOME_CHANNEL=
BROWSER_URL=http://browser:9222
BROWSER_VIEW_URL=
BROWSER_API_HOST=0.0.0.0
BROWSER_API_PORT=8088
BROWSER_USE_RPC_URL=http://browser:8787/run
BROWSER_USE_RPC_TIMEOUT=900
BROWSER_API_MAX_CONCURRENCY=2

13
api/Dockerfile Normal file
View file

@ -0,0 +1,13 @@
FROM python:3.12-slim
ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
WORKDIR /app
COPY requirements.txt /app/requirements.txt
RUN pip install --no-cache-dir -r /app/requirements.txt
COPY . /app/api
CMD ["uvicorn", "api.main:app", "--host", "0.0.0.0", "--port", "8088"]

17
api/README.md Normal file
View file

@ -0,0 +1,17 @@
# Browser REST API
REST-обертка над `browser-use` RPC (`http://browser:8787/run`).
## Endpoints
- `GET /health`
- `POST /api/browser/tasks`
- `GET /api/browser/tasks/{task_id}`
- `GET /api/browser/tasks/{task_id}/result`
## Пример
```bash
curl -sS -X POST http://localhost:8088/api/browser/tasks \
-H "Content-Type: application/json" \
-d '{"task":"Открой example.com и верни заголовок страницы","timeout":300}'

0
api/__init__.py Normal file
View file

31
api/browser_rpc_client.py Normal file
View file

@ -0,0 +1,31 @@
from typing import Any
import httpx
class BrowserRpcError(RuntimeError): ...
async def run_browser_task(rpc_url: str, task: str, timeout_sec: float) -> dict[str, Any]:
payload = {"task": task}
timeout = httpx.Timeout(timeout_sec)
async with httpx.AsyncClient(timeout=timeout) as client:
try:
response = await client.post(rpc_url, json=payload)
except httpx.HTTPError as exc:
raise BrowserRpcError(f"Transport error: {exc}")
if response.status_code >= 400:
body = response.text
raise BrowserRpcError(f"RPC HTTP: {response.status_code}: {body}")
try:
data = response.json()
except ValueError as exc:
raise BrowserRpcError("RPC returned non-JSON response")
if not isinstance(data, dict):
raise BrowserRpcError("RPC returned invalid payload type")
return data

16
api/config.py Normal file
View file

@ -0,0 +1,16 @@
import os
from dataclasses import dataclass
@dataclass(frozen=True)
class Settings:
app_host: str = os.getenv("BROWSER_API_HOST", "0.0.0.0")
app_port: int = os.getenv("BROWSER_API_PORT", "8080")
browser_rpc_url: str = os.getenv("BROWSER_USE_RPC_URL", "http://browser:8787/run")
browser_rpc_timeout: float = float(os.getenv("BROWSER_USE_RPC_TIMEOUT", "900"))
max_concurrency = int(os.getenv("BROWSER_API_MAX_CONCURRENCY", "2"))
settings = Settings()

112
api/main.py Normal file
View file

@ -0,0 +1,112 @@
import asyncio
from contextlib import asynccontextmanager
from fastapi import FastAPI, HTTPException
from fastapi.responses import JSONResponse
from api.browser_rpc_client import BrowserRpcError
from api.config import settings
from api.schemas import BrowserTaskRequest, BrowserTaskAcceptedResponse, BrowserTaskStatusResponse, \
BrowserTaskResultResponse, TaskStatus
from api.task_store import TaskStore
from api.browser_rpc_client import run_browser_task
store = TaskStore()
_semaphore = asyncio.Semaphore(settings.max_concurrency)
@asynccontextmanager
async def lifespan(app: FastAPI):
yield
app = FastAPI(title="Browser API", version="1.0.0", lifespan=lifespan)
async def _worker(task_id: str) -> None:
rec = await store.set_running(task_id)
if rec is None:
return None
async with _semaphore:
try:
raw = await asyncio.wait_for(
run_browser_task(
rpc_url=settings.browser_rpc_url,
task=rec.task,
timeout_sec=float(rec.timeout),
),
timeout=float(rec.timeout) + 5,
)
success = bool(raw.get("success"))
await store.set_done(task_id=task_id, success=success, raw_response=raw, error=None)
except asyncio.TimeoutError:
await store.set_done(
task_id=task_id,
success=False,
raw_response=None,
error="Timeout exceeded",
)
except BrowserRpcError as exc:
await store.set_done(
task_id=task_id,
success=False,
raw_response=None,
error=str(exc),
)
except Exception as exc:
await store.set_done(
task_id=task_id,
success=False,
raw_response=None,
error=f"Internal error: {exc}",
)
@app.get("/health")
async def health() -> dict:
return {"ok": True}
@app.post("/api/browser/tasks", response_model=BrowserTaskAcceptedResponse, status_code=202)
async def create_task(payload: BrowserTaskRequest) -> BrowserTaskAcceptedResponse:
rec = await store.create(task=payload.task.strip(), timeout=payload.timeout, metadata=payload.metadata)
asyncio.create_task(_worker(rec.task_id))
return BrowserTaskAcceptedResponse(task_id=rec.task_id, status=rec.status)
@app.get("/api/browser/tasks/{task_id}", response_model=BrowserTaskStatusResponse)
async def get_task_status(task_id: str) -> BrowserTaskStatusResponse:
rec = await store.get(task_id=task_id)
if rec is None:
raise HTTPException(status_code=404, detail="Task not found")
return BrowserTaskStatusResponse(task_id=rec.task_id, status=rec.status, create_at=rec.create_at,
started_at=rec.started_at, finished_at=rec.finished_at, error=rec.error)
@app.get("/api/browser/tasks/{task_id}/result", response_model=BrowserTaskResultResponse)
async def get_task_result(task_id: str) -> BrowserTaskResultResponse:
rec = await store.get(task_id=task_id)
if rec is None:
raise HTTPException(status_code=404, detail="Task not found")
if rec.status in (TaskStatus.queued, TaskStatus.running):
return JSONResponse(status_code=202, content={
"task_id": rec.task_id,
"status": rec.status,
"success": False,
"execution_time": rec.execution_time,
"result": None,
"error": None,
"raw_response": None,
})
return BrowserTaskResultResponse(
task_id=rec.task_id,
status=rec.status,
success=(rec.status == "succeeded"),
execution_time=rec.execution_time,
result=rec.result,
error=rec.error,
raw_response=rec.raw_response,
)

4
api/requirements.txt Normal file
View file

@ -0,0 +1,4 @@
fastapi==0.135.3
uvicorn[standard]==0.44.0
httpx==0.28.1
pydantic==2.12.5

View file

@ -1,15 +1,41 @@
from typing import Optional
from enum import Enum
from typing import Optional, Any
from pydantic import BaseModel, Field
class TaskStatus(str, Enum):
queued = "queued"
running = "running"
succeeded = "succeeded"
failed = "failed"
class BrowserTaskRequest(BaseModel):
task: str = Field(..., description="Задача для браузера")
timeout: int = Field(300, description="Максимальное время выполнения задачи в секундах")
metadata: dict[str, Any] | None = Field(default=None, description="Метаданные клиента")
class BrowserTaskResponse(BaseModel):
class BrowserTaskAcceptedResponse(BaseModel):
task_id: str
status: TaskStatus
class BrowserTaskStatusResponse(BaseModel):
task_id: str
status: TaskStatus
create_at: float
started_at: float | None = None
finished_at: float | None = None
error: str | None = None
class BrowserTaskResultResponse(BaseModel):
task_id: str
status: TaskStatus
success: bool
result: Optional[str] = None
error: Optional[str] = None
execution_time: float
result: str | None = None
error: str | None = None
raw_response: dict[str, Any] | None = None

69
api/task_store.py Normal file
View file

@ -0,0 +1,69 @@
import time
import uuid
from asyncio import Lock
from dataclasses import dataclass, field
from enum import Enum
from typing import Any
from api.schemas import TaskStatus
@dataclass
class TaskRecord:
task_id: str
task: str
timeout: int
metadata: dict[str, Any] | None
status: TaskStatus = TaskStatus.queued
create_at: float = field(default_factory=time.time)
started_at: float | None = None
finished_at: float | None = None
result: str | None = None
error: str | None = None
raw_response: dict[str, Any] | None = None
@property
def execution_time(self) -> float:
if self.started_at is None:
return 0
end = self.finished_at if self.finished_at is not None else time.time()
return max(0, end - self.started_at)
class TaskStore:
def __init__(self) -> None:
self._lock = Lock()
self._tasks: dict[str, TaskRecord] = {}
async def create(self, task: str, timeout: int, metadata: dict[str, Any] | None) -> TaskRecord:
task_id = uuid.uuid4().hex
rec = TaskRecord(task_id=task_id, task=task, timeout=timeout, metadata=metadata)
async with self._lock:
self._tasks[task_id] = rec
return rec
async def get(self, task_id: str) -> TaskRecord | None:
async with self._lock:
return self._tasks.get(task_id)
async def set_running(self, task_id: str) -> TaskRecord | None:
async with self._lock:
rec = self._tasks.get(task_id)
if rec is None:
return None
rec.status = TaskStatus.running
rec.started_at = time.time()
return rec
async def set_done(self, task_id: str, success: bool, raw_response: dict[str, Any] | None,
error: str | None) -> TaskRecord | None:
async with self._lock:
rec = self._tasks.get(task_id)
if rec is None:
return None
rec.finished_at = time.time()
rec.raw_response = raw_response
rec.error = error or (raw_response.get("error") if isinstance(raw_response, dict) else None)
rec.status = TaskStatus.succeeded if success else TaskStatus.failed
return rec

View file

@ -35,7 +35,6 @@ services:
fi;
exec python -m gateway.run
"
browser:
build:
context: ./browser_env
@ -55,12 +54,31 @@ services:
- browser_profiles:/src/browser_data
restart: always
healthcheck:
test: ["CMD-SHELL", "curl -fsS http://127.0.0.1:9222/json/version >/dev/null && curl -fsS http://127.0.0.1:8787/health >/dev/null || exit 1"]
test: [ "CMD-SHELL", "curl -fsS http://127.0.0.1:9222/json/version >/dev/null && curl -fsS http://127.0.0.1:8787/health >/dev/null || exit 1" ]
interval: 10s
timeout: 3s
retries: 12
start_period: 20s
browser-api:
build:
context: ./api
dockerfile: Dockerfile
container_name: hermes-browser-api
environment:
- BROWSER_USE_RPC_URL=http://browser:8787/run
- BROWSER_API_HOST=0.0.0.0
- BROWSER_API_PORT=8088
- BROWSER_API_MAX_CONCURRENCY=2
depends_on:
browser:
condition: service_healthy
ports:
- "8088:8088"
restart: always
networks:
- hermes-net
tunnel:
image: cloudflare/cloudflared:latest
profiles: