feat(scrape): 覆盖方式加「更新」并设为缺省,季集缺元数据不写占位,作品类型判定统一

- 每次写入在 scrape_record.files[].hash 记 sha256;update 模式只覆盖本工具写过且没被改过的文件,
  被改过的跳过(scrape_user_modified),内容没变不重写(scrape_unchanged),旧记录无 hash 按修改时间兜底;
  迁移 0070 给 scrape_record.mode 加 update,设置缺省改为 update(旧值照旧有效)
- 季 / 集单元缺元数据时按 TMDB id 现取季详情补齐;仍缺跳过(scrape_metadata_missing)并记 partial,不写占位 NFO
- scrapeKindOf 统一作品类型判定,合集 / 其他类作品 NFO 根元素为 <movie>;TMDB 映射按作品类型过滤 entity_type
- 兜底异常记 scrape_internal_error 并保留已写文件结果
- 新增 scheduleEntityRescrape(刮削过的作品延后排 update 重刮,排队记录合批防抖)

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Cxwgnn2pbzQ7rXv3rq7fei
This commit is contained in:
rootandClaude Opus 5.5 committed 2026-10-03 21:04:57 +08:00
1 parent 00045a24cb
commit dbbf39eda4
13 files changed
+607 -54

No files matched your search

@@ -0,0 +1,23 @@
import type { Kysely } from "kysely";
import { execDdl, oneOf } from "../src/migration-helpers.ts";
// 刮削覆盖策略加「更新」(update):只覆盖本工具上次写入且之后没被改过的文件(按 scrape_record.files 里记的 hash 判断),
// 被人改过的跳过(scrape_user_modified)。scrape_record.mode 的 CHECK 增加 update;逐文件 hash 放在 files jsonb 里,不加列。
const OLD = ["missing", "overwrite"];
const NEW = [...OLD, "update"];
const swap = (values: readonly string[]) => `
alter table scrape_record drop constraint if exists scrape_record_mode_check;
alter table scrape_record add constraint scrape_record_mode_check ${oneOf("mode", values)};
`;
// biome-ignore lint/suspicious/noExplicitAny: 迁移不依赖应用表类型
export async function up(db: Kysely<any>): Promise<void> {
await execDdl(db, swap(NEW));
}
// biome-ignore lint/suspicious/noExplicitAny: 同上
export async function down(db: Kysely<any>): Promise<void> {
await execDdl(db, `update scrape_record set mode = 'missing' where mode = 'update'; ${swap(OLD)}`);
}
+1 -1
View File
@@ -233,7 +233,7 @@ export type ScheduledJobRunTrigger = "schedule" | "manual" | "skip";
export type ScheduledJobRunStatus = "ok" | "failed" | "skipped" | "running";
export type ScrapeRecordStatus = "queued" | "running" | "succeeded" | "partial" | "failed";
export type ScrapeRecordTrigger = "auto" | "manual";
export type ScrapeRecordMode = "missing" | "overwrite";
export type ScrapeRecordMode = "missing" | "overwrite" | "update";
export type SecretPurpose =
| "site_cookie"
| "site_passkey"