feat: KOC资源页新增一键爬取更新全部账号ID

lib/account-id-refresh-service.ts

@@ -0,0 +1,134 @@
+import { runInBackground } from "./background";
+import {
+  resolveProfileDetailsFromMcp,
+  type CollectionMcpConfig,
+} from "./mcp-collection-client";
+import type { DatabaseClient } from "./database";
+
+export type PublicAccountIdRefreshStatus = {
+  status: "running" | "done";
+  total: number;
+  processed: number;
+  updated: number;
+  failed: number;
+  skipped: number;
+  startedAt: string;
+  finishedAt: string | null;
+};
+
+type RefreshTargetRow = {
+  id: string;
+  platform: string;
+  platform_uid: string;
+  profile_url: string;
+};
+
+export type ProfileScraper = (
+  profileUrl: string,
+  platform: "小红书" | "抖音",
+) => Promise<{ redId: string | null }>;
+
+const defaultScraper =
+  (config: CollectionMcpConfig): ProfileScraper =>
+  (profileUrl, platform) =>
+    resolveProfileDetailsFromMcp(profileUrl, platform, config);
+
+let currentJob: PublicAccountIdRefreshStatus | null = null;
+
+export function getPublicAccountIdRefreshStatus(): PublicAccountIdRefreshStatus | null {
+  return currentJob ? { ...currentJob } : null;
+}
+
+export function startPublicAccountIdRefresh(
+  db: DatabaseClient,
+  config: CollectionMcpConfig,
+): PublicAccountIdRefreshStatus {
+  if (currentJob?.status === "running") {
+    return { ...currentJob };
+  }
+  const job: PublicAccountIdRefreshStatus = {
+    status: "running",
+    total: 0,
+    processed: 0,
+    updated: 0,
+    failed: 0,
+    skipped: 0,
+    startedAt: new Date().toISOString(),
+    finishedAt: null,
+  };
+  currentJob = job;
+  runInBackground(
+    executePublicAccountIdRefresh(job, db, defaultScraper(config)),
+    "public account id refresh",
+  );
+  return { ...job };
+}
+
+async function executePublicAccountIdRefresh(
+  job: PublicAccountIdRefreshStatus,
+  db: DatabaseClient,
+  scraper: ProfileScraper,
+) {
+  const rows = await db
+    .prepare(
+      `SELECT id, platform, platform_uid, profile_url
+       FROM accounts
+       WHERE profile_url IS NOT NULL AND profile_url != ''
+         AND platform IN ('小红书', '抖音')`,
+    )
+    .all<RefreshTargetRow>();
+  const targets = rows.results.filter(
+    (row) => !row.platform_uid.startsWith("xhs-"),
+  );
+  job.total = targets.length;
+
+  let cursor = 0;
+  const worker = async () => {
+    while (cursor < targets.length) {
+      const row = targets[cursor];
+      cursor += 1;
+      try {
+        const details = await scraper(
+          row.profile_url,
+          row.platform === "抖音" ? "抖音" : "小红书",
+        );
+        const redId = details.redId?.trim();
+        if (redId) {
+          await db
+            .prepare(`UPDATE accounts SET public_account_id = ? WHERE id = ?`)
+            .bind(redId, row.id)
+            .run();
+          job.updated += 1;
+        } else {
+          job.skipped += 1;
+        }
+      } catch {
+        job.failed += 1;
+      }
+      job.processed += 1;
+    }
+  };
+  await Promise.all(
+    Array.from({ length: Math.min(4, targets.length) }, worker),
+  );
+  job.status = "done";
+  job.finishedAt = new Date().toISOString();
+}
+
+export async function runPublicAccountIdRefresh(
+  db: DatabaseClient,
+  scraper: ProfileScraper,
+): Promise<PublicAccountIdRefreshStatus> {
+  const job: PublicAccountIdRefreshStatus = {
+    status: "running",
+    total: 0,
+    processed: 0,
+    updated: 0,
+    failed: 0,
+    skipped: 0,
+    startedAt: new Date().toISOString(),
+    finishedAt: null,
+  };
+  await executePublicAccountIdRefresh(job, db, scraper);
+  return job;
+}