test(porting): differential tests against MoviePilot Rust-accelerated parsers

Snapshot differential tests (known deviations registered, current state passes):
- title/path recognition: 216 samples vs MP MetaInfo/MetaInfoPath
- filter rules: 42 rules x 28 torrents vs MP FilterModule
- RSS/indexer parsing: 15 feeds, 72 indexer samples
- zhconv: 69 samples
Generator: scripts/mp-sync/gen-differential.py; report: docs/porting/mp-rust-logic-parity.md

Co-Authored-By: Claude Opus 5.5 <[email protected]>

packages/adapters/test/differential/indexer.test.ts

@@ -0,0 +1,123 @@
+/**
+ * 站点索引解析差分:MP `SiteSpider.parse`(PyQuery 回退)vs 本项目 `SiteSpider.parse`。
+ * 同一 indexer 配置(本项目 definitions/*.json 或内联配置)+ 同一 HTML;期望由 scripts/mp-sync/gen-differential.py rss|indexer 生成(indexer.json)。
+ */
+import { describe, expect, it } from "vitest";
+import { BUILTIN_SITE_DEFINITIONS, type IndexerDefinition, SiteSpider } from "../../src/sites/index.ts";
+import {
+  diffAgainstRegistry,
+  dumpMismatches,
+  FieldCollector,
+  formatMismatches,
+  loadDoc,
+  readInput,
+} from "./diff-util.ts";
+
+interface IndexerSample {
+  id: string;
+  definition:
+    | { source: "definition"; id: string }
+    | {
+        source: "inline";
+        config: unknown;
+      };
+  input: { file?: string; inline?: string };
+  query: { keyword: string | null; search_type: "torrents" | "subtitles" };
+  mp: {
+    is_error?: boolean;
+    torrents?: Array<Record<string, unknown>>;
+    status?: string;
+    error?: string;
+  };
+}
+
+const doc = loadDoc<IndexerSample>("indexer.json");
+
+function definitionOf(s: IndexerSample): IndexerDefinition {
+  if (s.definition.source === "definition") {
+    return BUILTIN_SITE_DEFINITIONS[s.definition.id] as IndexerDefinition;
+  }
+  return s.definition.config as IndexerDefinition;
+}
+
+/**
+ * 行对齐:条数相同按下标;条数不同(MP 整行丢弃等)且双方都有 page_url 时按 page_url 配对,
+ * 避免一处丢行导致后续所有行错位。行号 i 取 MP 侧下标。
+ */
+function alignRows(
+  mp: Array<Record<string, unknown>>,
+  ours: Array<Record<string, unknown>>,
+): Array<[number, Record<string, unknown>, Record<string, unknown>]> {
+  const byUrl = mp.length !== ours.length && [...mp, ...ours].every((r) => typeof r.page_url === "string");
+  if (!byUrl) {
+    return mp.slice(0, Math.min(mp.length, ours.length)).map((m, i) => [i, m, ours[i]!]);
+  }
+  const index = new Map(ours.map((o) => [o.page_url as string, o]));
+  const out: Array<[number, Record<string, unknown>, Record<string, unknown>]> = [];
+  mp.forEach((m, i) => {
+    const o = index.get(m.page_url as string);
+    if (o) out.push([i, m, o]);
+  });
+  return out;
+}
+
+describe("站点索引差分(MP SiteSpider Python 回退)", () => {
+  const c = new FieldCollector();
+  for (const s of doc.samples) {
+    if (s.mp.status === "error") {
+      throw new Error(`${s.id}: MP 运行失败 ${s.mp.error}`);
+    }
+    if (s.query.search_type === "subtitles") {
+      // 本项目 SiteSpider 未移植字幕分支(parse_indexer_subtitles)
+      c.mismatches.push({
+        key: `${s.id}.status`,
+        mp: `ok:${s.mp.torrents?.length ?? 0}`,
+        ours: "unsupported",
+      });
+      continue;
+    }
+    const html = s.input.file ? readInput(s.input.file) : (s.input.inline ?? "");
+    const spider = new SiteSpider(definitionOf(s), s.query.keyword ? { keyword: s.query.keyword } : {});
+    const ours = spider.parse(html);
+    const mpRows = s.mp.torrents ?? [];
+    if (Boolean(s.mp.is_error) !== ours.isError) {
+      c.mismatches.push({
+        key: `${s.id}.status`,
+        mp: s.mp.is_error,
+        ours: ours.isError,
+      });
+    }
+    if (mpRows.length !== ours.torrents.length) {
+      c.mismatches.push({
+        key: `${s.id}.count`,
+        mp: mpRows.length,
+        ours: ours.torrents.length,
+      });
+    }
+    for (const [i, m, o] of alignRows(mpRows, ours.torrents as Array<Record<string, unknown>>)) {
+      for (const f of doc.fields) c.check(`${s.id}#${i}.${f}`, m[f], o[f]);
+    }
+  }
+
+  it("有样本且记录了生成来源", () => {
+    expect(doc.samples.length).toBeGreaterThan(10);
+    expect(doc.generatedFrom.commit).toMatch(/^[0-9a-f]{40}$/);
+  });
+
+  it("实际不一致集合 == 登记的 deviations", () => {
+    dumpMismatches("indexer", c.mismatches, {
+      compared: c.compared,
+      equal: c.equal,
+    });
+    const { unregistered, stale } = diffAgainstRegistry(c.mismatches, doc.deviations);
+    expect(formatMismatches(unregistered), "未登记的不一致").toBe("");
+    expect(stale.join("\n"), "登记了但已不再出现的偏差(请从 indexer.json deviations 删除)").toBe("");
+  });
+
+  it("登记项分类合法", () => {
+    for (const [key, d] of Object.entries(doc.deviations)) {
+      expect(["intentional", "missing", "our_bug", "mp_bug"], key).toContain(d.kind);
+      expect(d.note.length, key).toBeGreaterThan(0);
+    }
+  });
+});