feat(adapters): RSS and PT site adapters ported from MoviePilot v3 + mocks (W1-3b)
- rss: port of MP RssHelper._parse_impl/get_rss_link semantics with safe XML (entities off, DTD/ENTITY rejected, size/item caps), ETag/Last-Modified, backoff + Retry-After, circuit breaker with manual reset, passkey redaction, overlap-window GUID+digest dedupe - sites/spider: TS port of MP SiteSpider (indexer schema, search URL builder, field DSL, filters, Jinja templates via nunjucks, login-page detection); audiences definition hand-written as JSON; raw category id preserved; shared per-site token bucket - sites/mteam: port of MP MTorrentSpider (params, result mapping, promo factors, genDlToken flow) without embedding the API key in enclosures; business code/message checks - testkit: mock RSS server, mock NexusPHP site, mock M-Team API - docs: W1-3b migration mapping appended to porting ledger Co-Authored-By: Claude Fable 5.1 <[email protected]>
This commit is contained in:
rootandClaude Fable 5.1 committed
1 parent
ef2e504a94
commit
a6d049db81
47 files changed
+5414
No files matched your search
@@ -0,0 +1,249 @@
|
||||
/**
|
||||
* 通用 spider 站点适配器(NexusPHP 等 HTML 站)。
|
||||
*
|
||||
* 对应 MoviePilot e6820a8:
|
||||
* - `app/modules/indexer/__init__.py IndexerModule.search_torrents / refresh_torrents` → `search` / `list`
|
||||
* (`__search_check` 的英文站不搜中文 → search 前检查;`__clear_search_text` → `clearSearchText`;
|
||||
* `SitesHelper.check(domain)` 流控 → 站点令牌桶)
|
||||
* - `SiteSpider.get_torrents`(RequestUtils GET + cookie + ua)→ `fetchPage`
|
||||
* - `app/chain/download/submission.py download_torrent`(带 cookie GET enclosure,磁力原样返回)→ `downloadTarget`
|
||||
*
|
||||
* 本项目调整:登录页不再"返回空列表",而是抛 `auth_expired`(REQ §4.7.1 不展示空列表冒充无内容);
|
||||
* 3xx 不自动跟随(避免把 cookie 带到站外,并把跳转 login 识别为失效);凭证经 SecretResolver 注入。
|
||||
*/
|
||||
import { redactText } from "../../rss/redact";
|
||||
import { buildHttpContext, type SiteAdapterDeps } from "../deps";
|
||||
import { readBody, readText, siteFetch } from "../http";
|
||||
import {
|
||||
type DownloadTarget,
|
||||
type SiteAdapter,
|
||||
type SiteCallOptions,
|
||||
SiteError,
|
||||
type SiteHealth,
|
||||
type SiteListFilters,
|
||||
type SitePage,
|
||||
type SiteReleaseDetail,
|
||||
} from "../types";
|
||||
import { inspectTorrent } from "../util";
|
||||
import { parseDetailPage } from "./detail";
|
||||
import { spiderTorrentToRelease } from "./map";
|
||||
import type { IndexerDefinition } from "./schema";
|
||||
import { loadHtml, normalizeSelector, SiteSpider } from "./spider";
|
||||
|
||||
/** MP `text_tools.remove_punctuation(text, replacement=" ", allow_space=True)` */
|
||||
export function clearSearchText(text: string): string {
|
||||
return text
|
||||
.replace(/[-]/g, "")
|
||||
.replace(/[、.。,，·:：;；!！?？'’"“”()（）[\]【】「」\-—―+|\\_/&#～~]/g, " ")
|
||||
.replace(/\s+/g, " ")
|
||||
.trim();
|
||||
}
|
||||
|
||||
function containsChinese(s: string): boolean {
|
||||
return /[一-鿿]/.test(s);
|
||||
}
|
||||
|
||||
export function healthFromError(siteId: string, e: unknown, checkedAt: string): SiteHealth {
|
||||
if (e instanceof SiteError) {
|
||||
const status: SiteHealth["status"] =
|
||||
e.kind === "network"
|
||||
? "unreachable"
|
||||
: e.kind === "not_found" || e.kind === "unsupported"
|
||||
? "parse_error"
|
||||
: e.kind;
|
||||
return { siteId, status, ok: false, checkedAt, latencyMs: null, message: e.message };
|
||||
}
|
||||
return { siteId, status: "unreachable", ok: false, checkedAt, latencyMs: null, message: redactText(String(e)) };
|
||||
}
|
||||
|
||||
export function createSpiderSiteAdapter(input: IndexerDefinition, deps: SiteAdapterDeps): SiteAdapter {
|
||||
const def: IndexerDefinition = deps.baseUrl ? { ...input, domain: deps.baseUrl } : input;
|
||||
const ext = def.umt;
|
||||
const http = buildHttpContext(def.id, deps, ext.rateLimit);
|
||||
const base = def.domain.endsWith("/") ? def.domain : `${def.domain}/`;
|
||||
|
||||
async function cookie(): Promise<string> {
|
||||
const c = await deps.secrets.resolve({ siteId: def.id, kind: "cookie" });
|
||||
if (!c) throw new SiteError(def.id, "auth_missing", "未配置站点 Cookie");
|
||||
return c;
|
||||
}
|
||||
|
||||
async function fetchPage(url: string, what: string, opts: SiteCallOptions): Promise<string> {
|
||||
const res = await siteFetch(http, url, {
|
||||
method: "GET",
|
||||
headers: { cookie: await cookie(), accept: "text/html,application/xhtml+xml", referer: base },
|
||||
redirect: "manual",
|
||||
...(opts.signal ? { signal: opts.signal } : {}),
|
||||
});
|
||||
if (res.status >= 300 && res.status < 400) {
|
||||
const loc = res.headers.get("location") ?? "";
|
||||
await res.body?.cancel().catch(() => {});
|
||||
if (/login/i.test(loc)) throw new SiteError(def.id, "auth_expired", `${what}:跳转登录页,Cookie 已失效`);
|
||||
throw new SiteError(def.id, "parse_error", redactText(`${what}:意外重定向到 ${loc}`), { httpStatus: res.status });
|
||||
}
|
||||
if (res.status === 401 || res.status === 403) {
|
||||
await res.body?.cancel().catch(() => {});
|
||||
throw new SiteError(def.id, "auth_invalid", `${what}:HTTP ${res.status}`, { httpStatus: res.status });
|
||||
}
|
||||
if (!res.ok) {
|
||||
await res.body?.cancel().catch(() => {});
|
||||
throw new SiteError(def.id, res.status === 404 ? "not_found" : "network", `${what}:HTTP ${res.status}`, {
|
||||
httpStatus: res.status,
|
||||
});
|
||||
}
|
||||
return readText(http, res);
|
||||
}
|
||||
|
||||
async function run(
|
||||
page: number,
|
||||
filters: SiteListFilters,
|
||||
keyword: string | undefined,
|
||||
opts: SiteCallOptions,
|
||||
): Promise<SitePage> {
|
||||
const what = keyword ? "搜索" : `列表第 ${page} 页`;
|
||||
const spider = new SiteSpider(def, {
|
||||
...(keyword ? { keyword } : {}),
|
||||
...(filters.mtype ? { mtype: filters.mtype } : {}),
|
||||
...(filters.categoryIds?.length ? { cat: filters.categoryIds.join(",") } : {}),
|
||||
page: Math.max(1, page) - 1, // 对外 1 起,MP 内部 0 起
|
||||
});
|
||||
if (!spider.canSearch()) throw new SiteError(def.id, "unsupported", `${what}:站点定义不支持该媒体类型`);
|
||||
const html = await fetchPage(spider.getSearchUrl(), what, opts);
|
||||
const parsed = spider.parse(html);
|
||||
if (parsed.loginPage) throw new SiteError(def.id, "auth_expired", `${what}:返回登录或权限提示页`);
|
||||
if (parsed.isError || !parsed.$) throw new SiteError(def.id, "parse_error", `${what}:响应为空`);
|
||||
const $ = parsed.$;
|
||||
if (parsed.torrents.length === 0) {
|
||||
const text = $("body").text();
|
||||
const hasList =
|
||||
$(normalizeSelector(spider.def.torrents.list.selector)).length > 0 || $("table.torrents").length > 0;
|
||||
const empty = (ext.emptyMarkers ?? []).some((m) => text.includes(m));
|
||||
if (!hasList && !empty) throw new SiteError(def.id, "parse_error", `${what}:页面结构无法识别(可能改版)`);
|
||||
if (ext.loginSuccess && $(ext.loginSuccess).length === 0 && !empty) {
|
||||
throw new SiteError(def.id, "auth_expired", `${what}:未找到登录标志`);
|
||||
}
|
||||
}
|
||||
const observedAt = http.now().toISOString();
|
||||
let items = parsed.torrents
|
||||
.map((t) => spiderTorrentToRelease(t, def, observedAt))
|
||||
.filter((r): r is NonNullable<typeof r> => r !== null);
|
||||
if (filters.freeOnly) items = items.filter((r) => r.promotion.downloadFactor === 0);
|
||||
if (!keyword && filters.categoryIds?.length) {
|
||||
// MP 浏览模式不带分类参数;本地按原始分类过滤
|
||||
const set = new Set(filters.categoryIds);
|
||||
items = items.filter((r) => r.category.id !== null && set.has(r.category.id));
|
||||
}
|
||||
let maxRaw: number | null = null;
|
||||
if (ext.pagination) {
|
||||
$(ext.pagination).each((_, el) => {
|
||||
const p = Number(/[?&]page=(\d+)/.exec($(el).attr("href") ?? "")?.[1]);
|
||||
if (Number.isInteger(p) && (maxRaw === null || p > maxRaw)) maxRaw = p;
|
||||
});
|
||||
}
|
||||
const start = keyword ? 0 : Number(def.browse?.start ?? 0);
|
||||
return {
|
||||
items,
|
||||
page: Math.max(1, page),
|
||||
totalPages: maxRaw === null ? null : Math.max(page, (maxRaw as number) - start + 1),
|
||||
totalPagesEstimated: true,
|
||||
fetchedAt: observedAt,
|
||||
};
|
||||
}
|
||||
|
||||
return {
|
||||
id: def.id,
|
||||
kind: ext.kind,
|
||||
displayName: ext.displayName ?? def.name,
|
||||
capabilities: ext.capabilities,
|
||||
definitionVersion: ext.version,
|
||||
|
||||
list(page, filters = {}, opts = {}) {
|
||||
return run(page, filters, undefined, opts);
|
||||
},
|
||||
|
||||
async search(query, filters = {}, opts = {}) {
|
||||
const kw = /^tt\d+$/.test(query.trim()) ? query.trim() : clearSearchText(query);
|
||||
if (kw && def.language === "en" && containsChinese(kw)) {
|
||||
throw new SiteError(def.id, "unsupported", "该站点不支持中文搜索");
|
||||
}
|
||||
return run(filters.page ?? 1, filters, kw, opts);
|
||||
},
|
||||
|
||||
async detail(siteReleaseId, opts = {}): Promise<SiteReleaseDetail> {
|
||||
if (!/^\d+$/.test(siteReleaseId)) throw new SiteError(def.id, "not_found", "资源 ID 非法");
|
||||
if (!ext.detail) throw new SiteError(def.id, "unsupported", "站点定义未配置详情页解析");
|
||||
const url = new URL(`${ext.detail.path}?id=${siteReleaseId}&hit=1`, base).toString();
|
||||
const html = await fetchPage(url, `详情 ${siteReleaseId}`, opts);
|
||||
const r = parseDetailPage(html, def, siteReleaseId, http.now().toISOString());
|
||||
if (r.kind === "login") throw new SiteError(def.id, "auth_expired", `详情 ${siteReleaseId}:返回登录页`);
|
||||
if (r.kind === "not_found") throw new SiteError(def.id, "not_found", `详情 ${siteReleaseId}:资源不存在`);
|
||||
if (!r.detail) throw new SiteError(def.id, "parse_error", `详情 ${siteReleaseId}:页面结构无法识别`);
|
||||
return r.detail;
|
||||
},
|
||||
|
||||
async downloadTarget(siteReleaseId, opts = {}): Promise<DownloadTarget> {
|
||||
if (!/^\d+$/.test(siteReleaseId)) throw new SiteError(def.id, "not_found", "资源 ID 非法");
|
||||
const what = `下载 ${siteReleaseId}`;
|
||||
const url = new URL(`${ext.downloadPath}?id=${siteReleaseId}`, base).toString();
|
||||
const res = await siteFetch(http, url, {
|
||||
method: "GET",
|
||||
headers: { cookie: await cookie(), referer: base },
|
||||
redirect: "manual",
|
||||
...(opts.signal ? { signal: opts.signal } : {}),
|
||||
});
|
||||
const verifiedAt = http.now().toISOString();
|
||||
if (res.status >= 300 && res.status < 400) {
|
||||
const loc = res.headers.get("location") ?? "";
|
||||
await res.body?.cancel().catch(() => {});
|
||||
// MP download_torrent:磁力链原样返回
|
||||
if (loc.startsWith("magnet:")) return { kind: "magnet", uri: loc, verifiedAt };
|
||||
if (/login/i.test(loc)) throw new SiteError(def.id, "auth_expired", `${what}:跳转登录页,Cookie 已失效`);
|
||||
throw new SiteError(def.id, "parse_error", `${what}:意外重定向`, { httpStatus: res.status });
|
||||
}
|
||||
if (!res.ok) {
|
||||
await res.body?.cancel().catch(() => {});
|
||||
throw new SiteError(def.id, res.status === 404 ? "not_found" : "network", `${what}:HTTP ${res.status}`, {
|
||||
httpStatus: res.status,
|
||||
});
|
||||
}
|
||||
const bytes = await readBody(http, res);
|
||||
const info = inspectTorrent(bytes);
|
||||
if (!info) {
|
||||
// 站点在 cookie 失效时常返回 200 + 登录页 HTML
|
||||
const $ = loadHtml(new TextDecoder().decode(bytes.subarray(0, 200_000)));
|
||||
if (SiteSpider.isLoginOrPermissionPage($)) throw new SiteError(def.id, "auth_expired", `${what}:返回登录页`);
|
||||
const text = $("body").text();
|
||||
if (ext.detail?.notFoundMarkers.some((m) => text.includes(m))) {
|
||||
throw new SiteError(def.id, "not_found", `${what}:资源不存在`);
|
||||
}
|
||||
throw new SiteError(def.id, "parse_error", `${what}:返回内容不是 .torrent`);
|
||||
}
|
||||
const disposition = res.headers.get("content-disposition") ?? "";
|
||||
const fileName =
|
||||
/filename\*=UTF-8''([^;]+)/i.exec(disposition)?.[1] ?? /filename="?([^";]+)"?/i.exec(disposition)?.[1] ?? null;
|
||||
return {
|
||||
kind: "torrent",
|
||||
bytes,
|
||||
infoHash: info.infoHash,
|
||||
name: info.name,
|
||||
fileName: fileName ? decodeURIComponent(fileName) : null,
|
||||
verifiedAt,
|
||||
};
|
||||
},
|
||||
|
||||
async healthCheck(opts = {}): Promise<SiteHealth> {
|
||||
const started = Date.now();
|
||||
const checkedAt = http.now().toISOString();
|
||||
try {
|
||||
const html = await fetchPage(new URL("index.php", base).toString(), "健康检查", opts);
|
||||
const $ = loadHtml(html);
|
||||
if (SiteSpider.isLoginOrPermissionPage($) || (ext.loginSuccess && $(ext.loginSuccess).length === 0)) {
|
||||
throw new SiteError(def.id, "auth_expired", "健康检查:未登录");
|
||||
}
|
||||
return { siteId: def.id, status: "ok", ok: true, checkedAt, latencyMs: Date.now() - started };
|
||||
} catch (e) {
|
||||
return healthFromError(def.id, e, checkedAt);
|
||||
}
|
||||
},
|
||||
};
|
||||
}
|
||||
Reference in new issue
Block a user