Spaces:
Sleeping
Sleeping
File size: 14,229 Bytes
097eaf0 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 | """
WebSocket-based watcher for v2 jobs (crawl and batch), mirroring v1 behavior.
Usage:
watcher = client.watcher(job_id, kind="crawl")
watcher.add_listener(lambda status: print(status.status))
watcher.start()
"""
import asyncio
import json
import threading
from typing import Callable, List, Optional, Literal, Union, Dict, Any
import websockets
from .types import CrawlJob, BatchScrapeJob, Document
from .utils.normalize import normalize_document_input
JobKind = Literal["crawl", "batch"]
JobType = Union[CrawlJob, BatchScrapeJob]
class Watcher:
def __init__(
self,
client: object,
job_id: str,
kind: JobKind = "crawl",
poll_interval: int = 2,
timeout: Optional[int] = None,
) -> None:
self._client = client
self._job_id = job_id
self._kind = kind
self._timeout = timeout
self._poll_interval = poll_interval
self._listeners: List[Callable[[JobType], None]] = []
self._thread: Optional[threading.Thread] = None
self._stop = threading.Event()
http_client = getattr(client, "http_client", None)
self._api_url: Optional[str] = getattr(http_client, "api_url", None)
self._api_key: Optional[str] = getattr(http_client, "api_key", None)
# v1-parity state and event handlers
self.status: str = "scraping"
self.data: List[Dict[str, Any]] = []
self._event_handlers: Dict[str, List[Callable[[Dict[str, Any]], None]]] = {
"done": [],
"error": [],
"document": [],
}
self._sent_done: bool = False
self._sent_error: bool = False
def add_listener(self, callback: Callable[[JobType], None]) -> None:
self._listeners.append(callback)
def _emit(self, status: JobType) -> None:
for cb in list(self._listeners):
try:
cb(status)
except Exception:
pass
# v1-like events API
def add_event_listener(self, event_type: str, handler: Callable[[Dict[str, Any]], None]) -> None:
if event_type in self._event_handlers:
self._event_handlers[event_type].append(handler)
def dispatch_event(self, event_type: str, detail: Dict[str, Any]) -> None:
if event_type in self._event_handlers:
for handler in self._event_handlers[event_type]:
try:
handler(detail)
except Exception:
pass
def _build_ws_url(self) -> str:
if not self._api_url:
raise ValueError("API URL is required for WebSocket watcher")
ws_base = self._api_url.replace("https://", "wss://").replace("http://", "ws://", 1)
if self._kind == "crawl":
return f"{ws_base}/v2/crawl/{self._job_id}"
return f"{ws_base}/v2/batch/scrape/{self._job_id}"
async def _run_ws(self) -> None:
uri = self._build_ws_url()
headers_list = []
if self._api_key:
headers_list.append(("Authorization", f"Bearer {self._api_key}"))
try:
async with websockets.connect(uri, max_size=None, additional_headers=headers_list) as websocket:
deadline = asyncio.get_event_loop().time() + self._timeout if self._timeout else None
while not self._stop.is_set():
# Use short recv timeouts to allow HTTP polling fallback
if deadline is not None:
remaining = max(0.0, deadline - asyncio.get_event_loop().time())
timeout = min(self._poll_interval or remaining, remaining)
else:
timeout = self._poll_interval or 5
try:
msg = await asyncio.wait_for(websocket.recv(), timeout=timeout)
except asyncio.TimeoutError:
# Quiet period: poll HTTP once to progress statuses
if await self._poll_status_once():
break
else:
continue
except asyncio.CancelledError:
break
except Exception:
# Connection error: switch to HTTP polling until terminal or timeout
while not self._stop.is_set():
if await self._poll_status_once():
return
if deadline is not None and asyncio.get_event_loop().time() >= deadline:
return
await asyncio.sleep(self._poll_interval or 2)
return
try:
body = json.loads(msg)
except Exception:
continue
# v1-style typed event handling
msg_type = body.get("type")
if msg_type == "error":
self.status = "failed"
self.dispatch_event("error", {
"status": self.status,
"data": self.data,
"error": body.get("error"),
"id": self._job_id,
})
self._sent_error = True
# Emit a final failed snapshot for listeners
if self._kind == "crawl":
job = CrawlJob(status="failed", completed=0, total=0, credits_used=0, expires_at=None, next=None, data=[])
else:
job = BatchScrapeJob(status="failed", completed=0, total=0, credits_used=0, expires_at=None, next=None, data=[])
self._emit(job)
break
elif msg_type == "catchup":
d = body.get("data", {})
self.status = d.get("status", self.status)
docs_in = d.get("data", [])
self.data.extend(docs_in)
for doc in docs_in:
self.dispatch_event("document", {"data": doc, "id": self._job_id})
elif msg_type == "document":
doc = body.get("data")
if isinstance(doc, dict):
self.data.append(doc)
self.dispatch_event("document", {"data": doc, "id": self._job_id})
elif msg_type == "done":
self.status = "completed"
# Gather any documents in the done payload
raw_payload = body.get("data", {}) or {}
docs_in = raw_payload.get("data", []) or []
if isinstance(docs_in, list) and docs_in:
for doc in docs_in:
if isinstance(doc, dict):
self.data.append(doc)
# Dispatch done event first
self.dispatch_event("done", {"status": self.status, "data": self.data, "id": self._job_id})
self._sent_done = True
# Emit a final completed snapshot for listeners and break immediately
docs: List[Document] = []
for doc in self.data:
if isinstance(doc, dict):
d = normalize_document_input(doc)
docs.append(Document(**d))
if self._kind == "crawl":
job = CrawlJob(
status="completed",
completed=raw_payload.get("completed", 0),
total=raw_payload.get("total", 0),
credits_used=raw_payload.get("creditsUsed", 0),
expires_at=raw_payload.get("expiresAt"),
next=raw_payload.get("next"),
data=docs,
)
else:
job = BatchScrapeJob(
status="completed",
completed=raw_payload.get("completed", 0),
total=raw_payload.get("total", 0),
credits_used=raw_payload.get("creditsUsed", 0),
expires_at=raw_payload.get("expiresAt"),
next=raw_payload.get("next"),
data=docs,
)
self._emit(job)
break
payload = body.get("data", body)
# Only treat messages with an explicit status as job snapshots
has_status_field = (isinstance(payload, dict) and "status" in payload) or ("status" in body)
if not has_status_field:
continue
status_str = payload.get("status", body.get("status", self.status))
if self._kind == "crawl":
docs = []
for doc in payload.get("data", []):
if isinstance(doc, dict):
d = normalize_document_input(doc)
docs.append(Document(**d))
job = CrawlJob(
status=status_str,
completed=payload.get("completed", 0),
total=payload.get("total", 0),
credits_used=payload.get("creditsUsed", 0),
expires_at=payload.get("expiresAt"),
next=payload.get("next"),
data=docs,
)
self._emit(job)
if status_str in ("completed", "failed", "cancelled"):
# Ensure done/error dispatched even if server didn't send explicit event type
if status_str == "completed" and not self._sent_done:
self.dispatch_event("done", {"status": status_str, "data": self.data, "id": self._job_id})
self._sent_done = True
if status_str == "failed" and not self._sent_error:
self.dispatch_event("error", {"status": status_str, "data": self.data, "id": self._job_id})
self._sent_error = True
break
else:
docs = []
for doc in payload.get("data", []):
if isinstance(doc, dict):
d = normalize_document_input(doc)
docs.append(Document(**d))
job = BatchScrapeJob(
status=status_str,
completed=payload.get("completed", 0),
total=payload.get("total", 0),
credits_used=payload.get("creditsUsed"),
expires_at=payload.get("expiresAt"),
next=payload.get("next"),
data=docs,
)
self._emit(job)
if status_str in ("completed", "failed", "cancelled"):
if status_str == "completed" and not self._sent_done:
self.dispatch_event("done", {"status": status_str, "data": self.data, "id": self._job_id})
self._sent_done = True
if status_str == "failed" and not self._sent_error:
self.dispatch_event("error", {"status": status_str, "data": self.data, "id": self._job_id})
self._sent_error = True
break
except Exception:
pass
finally:
# Ensure terminal event parity with v1 even on abrupt disconnects
if self.status == "completed" and not self._sent_done:
self.dispatch_event("done", {"status": self.status, "data": self.data, "id": self._job_id})
self._sent_done = True
async def _poll_status_once(self) -> bool:
"""Poll job status over HTTP once. Returns True if terminal."""
try:
if self._kind == "crawl":
job: CrawlJob = await asyncio.to_thread(self._client.get_crawl_status, self._job_id)
else:
job: BatchScrapeJob = await asyncio.to_thread(self._client.get_batch_scrape_status, self._job_id)
except Exception:
return False
self.status = job.status
self._emit(job)
if job.status in ("completed", "failed", "cancelled"):
if job.status == "completed" and not self._sent_done:
self.dispatch_event("done", {"status": job.status, "data": [d.model_dump() for d in job.data], "id": self._job_id})
self._sent_done = True
if job.status == "failed" and not self._sent_error:
self.dispatch_event("error", {"status": job.status, "data": [d.model_dump() for d in job.data], "id": self._job_id})
self._sent_error = True
return True
return False
def _loop(self) -> None:
asyncio.run(self._run_ws())
def start(self) -> None:
if self._thread and self._thread.is_alive():
return
self._stop.clear()
self._thread = threading.Thread(target=self._loop, daemon=True)
self._thread.start()
def stop(self) -> None:
self._stop.set()
if self._thread:
self._thread.join(timeout=1)
|