Peach: Local-first video metadata scraper and media manager.
"""一次补完库里女优的资料:逐条跑补女优资料后继(ADR-0067),不等处理任务按名额排。
处理任务每轮只给存量 `STOCK_SHARE` 个名额,九百多位女优要一个多月才轮遍;这个脚本把
同一条后继在当前进程里挨个跑掉。取页、限流、冷却、双向核对、落库与判词 CSV 全是后继
自己那一套,这里只负责挑人和记进度:
- 挑人用后继的 `stock`:有路进得去、记号未到期的女优,作品多的在前。跑过又没变的由记号
跳过,所以中断之后原样重跑就是接着跑。
- 写入批次是 `auto:performer-profile@manual-<时间>`,
`scripts/revert_auto_landing.py --source auto:performer-profile` 整批撤回。
- 连续 `STOP_AFTER_UNFETCHED` 位都「未取得」多半是站点冷却或断网,停下来报告,不把剩下
的女优全刷成「未取得」。
用法:
python scripts/run_performer_profiles.py # dry-run,只报待跑人数
python scripts/run_performer_profiles.py --apply --backup <路径> [--limit N]
"""
from __future__ import annotations
import argparse
import sys
import time
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src"))
from peach import performer_profile_followup as profiles # noqa: E402
from peach.followups import Attempts, attempts_root # noqa: E402
from peach.migrations import sqlite_backup # noqa: E402
from peach.scripting import (BACKUP_REQUIRED, add_ledger_write_args, open_readonly, # noqa: E402
verify_after_write)
#: 连续几位「未取得」就停。站点冷却期内每一位都会立刻落成「未取得」,再跑只是刷判词。
STOP_AFTER_UNFETCHED = 5
class Handle:
"""后继要的任务句柄只用到批次号与进度,这里把进度打到终端。"""
def __init__(self, run_id: str):
self.run_id = run_id
def progress(self, **_kwargs) -> None:
return None
def pending(connection, generated: Path, limit: int | None) -> list:
attempts = Attempts(attempts_root(generated))
return profiles.stock(connection, attempts, limit=limit or 10**9)
def sweep(contract, items, run_id: str, *, out=sys.stdout,
stop_after: int = STOP_AFTER_UNFETCHED) -> dict:
"""挨个跑,返回各结论的计数;连续 `stop_after` 位未取得就停。"""
handle, tally, streak = Handle(run_id), {}, 0
for index, item in enumerate(items, 1):
try:
summary = profiles.run(contract, item.key, handle)
except Exception as error: # 一位出错只算她自己,接着跑下一位。
summary = {"outcome": f"出错:{type(error).__name__}: {error}"}
outcome = str(summary.get("outcome", ""))
head = outcome.split(";")[0] or "无结论"
tally[head] = tally.get(head, 0) + 1
print(f"{index}/{len(items)} {summary.get('name', item.label)}:{outcome}", file=out, flush=True)
streak = streak + 1 if outcome == profiles.UNFETCHED else 0
if streak >= stop_after:
print(f"连续 {streak} 位未取得,停下:多半是站点冷却或网络断了,稍后原样重跑即可接着来",
file=out, flush=True)
tally["停在"] = index
break
return tally
def main(argv=None) -> int:
# 原地改编码:`sweep` 的默认输出在定义时就绑着这个对象。中文控制台是 GBK,名字里的生僻字会让打印抛错。
sys.stdout.reconfigure(encoding="utf-8", errors="replace")
parser = argparse.ArgumentParser(description=__doc__.splitlines()[0])
add_ledger_write_args(parser)
parser.add_argument("--limit", type=int, help="这一趟最多跑几位")
args = parser.parse_args(argv)
if args.apply and not args.backup:
raise SystemExit(BACKUP_REQUIRED)
from peach.web_state import WebContract
contract = WebContract(args.db)
with open_readonly(args.db) as connection:
items = pending(connection, contract.candidate_root, args.limit)
print(f"待补资料的女优:{len(items)} 位")
if not args.apply:
for item in items[:10]:
print(f" {item.label}")
return 0
sqlite_backup(Path(args.db), Path(args.backup))
print(f"备份:{args.backup}")
run_id = f"manual-{time.strftime('%Y%m%dT%H%M%S')}"
tally = sweep(contract, items, run_id)
print("结论:" + ",".join(f"{key} {value}" for key, value in tally.items()))
print(f"批次:{profiles.SOURCE}@{run_id};判词:{contract.candidate_root / profiles.REVIEW_FILE}")
with open_readonly(args.db) as connection:
integrity, violations = verify_after_write(connection)
print(f"integrity_check:{integrity};外键违规:{violations}")
return 0 if integrity == "ok" and not violations else 1
if __name__ == "__main__":
raise SystemExit(main())