improve scraper runtime

This commit is contained in:
qananasikq
2026-04-08 13:30:45 +03:00
parent 21bdf17f35
commit f96e5ca5f8
9 changed files with 207 additions and 21 deletions

View File

@@ -1,4 +1,5 @@
import logging
import random
import time
from collections.abc import Callable
from functools import wraps
@@ -8,8 +9,23 @@ from playwright.sync_api import Error, TimeoutError as PlaywrightTimeoutError
logger = logging.getLogger("iaai_scraper.retry")
# Ошибки, которые считаем временными и пригодными для повтора.
RETRYABLE_EXCEPTIONS = (
PlaywrightTimeoutError,
Error,
ConnectionError,
OSError,
TimeoutError,
)
def retryable(max_attempts: int, delay_seconds: float = 2.5) -> Callable[[Callable[..., Any]], Callable[..., Any]]:
def retryable(
max_attempts: int,
delay_seconds: float = 2.5,
backoff_multiplier: float = 2.0,
jitter_seconds: float = 0.0,
) -> Callable[[Callable[..., Any]], Callable[..., Any]]:
# Универсальный retry с backoff и небольшим случайным jitter.
def decorator(func: Callable[..., Any]) -> Callable[..., Any]:
@wraps(func)
def wrapper(*args: Any, **kwargs: Any) -> Any:
@@ -17,11 +33,15 @@ def retryable(max_attempts: int, delay_seconds: float = 2.5) -> Callable[[Callab
for attempt in range(1, max_attempts + 1):
try:
return func(*args, **kwargs)
except (PlaywrightTimeoutError, Error, ConnectionError, OSError, RuntimeError) as exc:
except RETRYABLE_EXCEPTIONS as exc:
last_error = exc
logger.warning("%s failed on attempt %s/%s: %s", func.__name__, attempt, max_attempts, exc)
if attempt < max_attempts:
time.sleep(delay_seconds * (2 ** (attempt - 1)))
sleep_for = delay_seconds * (backoff_multiplier ** (attempt - 1))
if jitter_seconds > 0:
sleep_for += random.uniform(0, jitter_seconds)
logger.debug("Retrying %s in %.2fs", func.__name__, sleep_for)
time.sleep(sleep_for)
if last_error is not None:
raise last_error
raise RuntimeError("Retry wrapper failed without a captured exception")