feat(identity): web metadata enrichment for titles missing on TMDB (META-1)

- metadata.enrich (propose, low, replay safe): normalized-title cache (kv_cache region enrich, 7 days) -> daily budget -> web.search (title + 短剧 / 电视剧平台, second query limited to whitelist via site:) -> web.fetch of up to 3 relevant whitelisted pages (registered site domains never read, redirects out of the whitelist dropped) -> upstream model extraction with rule fallback -> server-side evidence check (rule_confirmed / multi_evidence / needs_review). Suggestion only. - POST /identity-pool/{id}/enrich (suggest) and /enrich/apply (console session): provisional entity or same-title merge, season/episode units, field-level metadata with source web/agent + evidence, aliases, release assignment. Poster proxy GET /img/enrich/{id} (og:image from whitelisted pages only). - Auto-enrich once per pool entry when subscription/rule lookups find nothing on TMDB (identity.autoEnrich, off by default); identity.enrichDailyBudget, identity.enrichDomains settings; web.fetch now returns og:image. - Pool sheet "元数据补全" section; entity detail shows evidence links. - Migration 0049_enrichment (metadata_enrichment, metadata_field.evidence, source web). Requires image rebuild / api + worker restart to take effect. Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 15:57:48 +08:00
1 parent 18a5cf1de7
commit f4666a4671
@@ -0,0 +1,228 @@
+import type { EnrichFieldName, IdentityPoolItem, MetadataEnrichOutput } from "@umt/contracts";
+import { ExternalLink, LoaderCircle, Sparkles } from "lucide-react";
+import { useId, useState } from "react";
+import { Link } from "react-router";
+import { toast } from "sonner";
+import { enrichPosterUrl, useApplyEnrich, useEnrichPool } from "~/api/identity.ts";
+import { DetailSection } from "~/components/console/detail-sheet.tsx";
+import { consolePath } from "~/components/console/nav.ts";
+import { Button } from "~/components/ui/button.tsx";
+import { Label } from "~/components/ui/label.tsx";
+import { Switch } from "~/components/ui/switch.tsx";
+import { errorMessage } from "~/lib/api.ts";
+import { ENRICH_FIELD_LABEL, ENRICH_SKIPPED, EVIDENCE_LEVEL } from "./labels.ts";
+import { Chip } from "./parts.tsx";
+
+// META-1 待识别资源 → 元数据补全:先看建议(网页搜索 + 白名单网页 + Agent 抽取,不改作品),再应用(建本地临时作品并关联)。
+
+type Suggestion = NonNullable<MetadataEnrichOutput["suggestion"]>;
+type Value = { value: unknown; sourceUrl: string | null; evidenceLevel: string; verbatim: boolean };
+
+const hostOf = (u: string | null) => {
+  if (!u) return null;
+  try {
+    return new URL(u).hostname.replace(/^www\./, "");
+  } catch {
+    return null;
+  }
+};
+
+function SourceLink({ url }: { url: string | null }) {
+  const host = hostOf(url);
+  if (!url || !host) return <span className="text-meta text-muted-foreground">无出处</span>;
+  return (
+    <a
+      href={url}
+      target="_blank"
+      rel="noreferrer noopener"
+      className="inline-flex items-center gap-1 text-meta break-all text-muted-foreground underline-offset-4 hover:text-foreground hover:underline"
+    >
+      {host}
+      <ExternalLink className="h-3 w-3 shrink-0" />
+    </a>
+  );
+}
+
+function valuesOf(s: Suggestion, k: EnrichFieldName): Value[] {
+  const v = s.fields[k] as Value | Value[] | null;
+  return Array.isArray(v) ? v : v ? [v] : [];
+}
+
+function FieldRow({ k, s, enrichmentId }: { k: EnrichFieldName; s: Suggestion; enrichmentId: string }) {
+  const vals = valuesOf(s, k);
+  if (!vals.length) return null;
+  return (
+    <li className="flex flex-wrap items-start gap-x-3 gap-y-1 py-2 first:pt-0 last:pb-0">
+      <div className="w-16 shrink-0 pt-0.5 text-meta text-muted-foreground">{ENRICH_FIELD_LABEL[k] ?? k}</div>
+      <div className="min-w-0 flex-1 basis-48 space-y-1">
+        {k === "posterUrl" ? (
+          <img
+            src={enrichPosterUrl(enrichmentId)}
+            alt="海报"
+            className="h-32 w-auto rounded-md border border-border object-cover"
+            loading="lazy"
+          />
+        ) : null}
+        {vals.map((v, i) => (
+          <div key={i} className="min-w-0">
+            {k !== "posterUrl" ? <p className="text-sm break-words text-foreground">{String(v.value)}</p> : null}
+            <div className="flex flex-wrap items-center gap-1.5">
+              <Chip>{EVIDENCE_LEVEL[v.evidenceLevel] ?? v.evidenceLevel}</Chip>
+              {v.verbatim ? null : <Chip>Agent 归纳</Chip>}
+              <SourceLink url={v.sourceUrl} />
+            </div>
+          </div>
+        ))}
+      </div>
+    </li>
+  );
+}
+
+const FIELDS: EnrichFieldName[] = [
+  "title",
+  "aliases",
+  "year",
+  "platform",
+  "episodeCount",
+  "genre",
+  "synopsis",
+  "posterUrl",
+];
+
+export function EnrichSection({ item }: { item: IdentityPoolItem }) {
+  const uid = useId();
+  const enrich = useEnrichPool();
+  const apply = useApplyEnrich();
+  const [out, setOut] = useState<MetadataEnrichOutput | null>(null);
+  const [lock, setLock] = useState(false);
+  const s = out?.suggestion ?? null;
+
+  const run = async (fresh: boolean) => {
+    try {
+      const r = await enrich.mutateAsync({ id: item.id, ...(fresh ? { fresh: true } : {}) });
+      setOut(r);
+      if (r.status === "failed") toast.error(r.error ?? "没有找到可用的信息");
+    } catch (err) {
+      toast.error(errorMessage(err, "补全失败"));
+    }
+  };
+  const doApply = async (mediaEntityId?: string) => {
+    if (!out) return;
+    try {
+      const r = await apply.mutateAsync({
+        id: item.id,
+        enrichmentId: out.enrichmentId,
+        lock,
+        ...(mediaEntityId ? { mediaEntityId } : {}),
+      });
+      toast.success(r.created ? "已建临时作品并关联" : "已并入已有作品并关联");
+      setOut({ ...out, status: "applied", appliedMediaEntityId: r.mediaEntityId });
+    } catch (err) {
+      toast.error(errorMessage(err, "应用失败"));
+    }
+  };
+
+  return (
+    <DetailSection
+      title="元数据补全"
+      actions={
+        <Button variant="outline" size="sm" disabled={enrich.isPending} onClick={() => run(false)}>
+          {enrich.isPending ? <LoaderCircle className="animate-spin" /> : <Sparkles />}
+          {out ? "重新载入" : "补全"}
+        </Button>
+      }
+    >
+      {!out ? (
+        <p className="text-caption text-muted-foreground">
+          TMDB 查不到时，按标题搜索网页，只读平台和百科网站，抽取字段并附出处。先看建议，确认后才建作品。
+        </p>
+      ) : (
+        <div className="space-y-3">
+          <p className="text-meta text-muted-foreground">
+            {out.notice}
+            {" · "}
+            {out.cached ? "7 天内的缓存结果" : `搜索 ${out.usage.searchCalls} 次，读取网页 ${out.usage.fetchCalls} 次`}
+            {s ? ` · ${s.extractor === "model" ? `Agent 抽取${s.model ? `（${s.model}）` : ""}` : "规则抽取"}` : ""}
+            {` · 今日 ${out.budget.usedToday}/${out.budget.dailyLimit}`}
+          </p>
+          {out.error ? <p className="text-caption text-warning">{out.error}</p> : null}
+          {s ? (
+            <ul className="divide-y divide-border/60">
+              {FIELDS.map((k) => (
+                <FieldRow key={k} k={k} s={s} enrichmentId={out.enrichmentId} />
+              ))}
+            </ul>
+          ) : null}
+          {s?.sources.length ? (
+            <details className="text-meta text-muted-foreground">
+              <summary className="cursor-pointer">搜索到的链接（{s.sources.length}）</summary>
+              <ul className="mt-1.5 space-y-1">
+                {s.sources.map((x) => (
+                  <li key={x.url} className="flex flex-wrap items-center gap-1.5">
+                    <SourceLink url={x.url} />
+                    <span>
+                      {x.fetched ? "已读取" : x.skipped ? (ENRICH_SKIPPED[x.skipped] ?? x.skipped) : "未读取"}
+                    </span>
+                  </li>
+                ))}
+              </ul>
+            </details>
+          ) : null}
+          {out.appliedMediaEntityId ? (
+            <Button asChild variant="outline" size="sm">
+              <Link to={consolePath("library", out.appliedMediaEntityId)}>查看作品</Link>
+            </Button>
+          ) : out.status === "suggested" && s ? (
+            <div className="space-y-2">
+              <div className="flex items-center gap-2">
+                <Switch id={`${uid}-lock`} checked={lock} onCheckedChange={setLock} />
+                <Label htmlFor={`${uid}-lock`} className="text-caption">
+                  应用的字段同时锁定
+                </Label>
+              </div>
+              {out.existing.length ? (
+                <div className="space-y-1.5">
+                  <p className="text-caption text-muted-foreground">
+                    已有同名作品。点「应用」时年份不冲突的自动并入，也可以指定一个：
+                  </p>
+                  {out.existing.map((e) => (
+                    <div key={e.mediaEntityId} className="flex flex-wrap items-center gap-2">
+                      <Link
+                        to={consolePath("library", e.mediaEntityId)}
+                        className="text-sm underline-offset-4 hover:underline"
+                      >
+                        {e.title}
+                        {e.year ? ` (${e.year})` : ""}
+                      </Link>
+                      <Button
+                        variant="outline"
+                        size="sm"
+                        disabled={apply.isPending}
+                        onClick={() => doApply(e.mediaEntityId)}
+                      >
+                        并入
+                      </Button>
+                    </div>
+                  ))}
+                </div>
+              ) : null}
+              <div className="flex flex-wrap gap-2">
+                <Button size="sm" disabled={apply.isPending} onClick={() => doApply()}>
+                  {apply.isPending ? <LoaderCircle className="animate-spin" /> : null}
+                  应用
+                </Button>
+                <Button variant="ghost" size="sm" disabled={enrich.isPending} onClick={() => run(true)}>
+                  重新搜索
+                </Button>
+              </div>
+            </div>
+          ) : (
+            <Button variant="ghost" size="sm" disabled={enrich.isPending} onClick={() => run(true)}>
+              重新搜索
+            </Button>
+          )}
+        </div>
+      )}
+    </DetailSection>
+  );
+}