UHD_Media_Tool/packages/adapters/src/sites/user-parsers/ipt.ts

/**
 * SPDX-License-Identifier: GPL-3.0-only
 * 移植自 MoviePilot(https://github.com/jxxghp/MoviePilot,GPLv3),本文件按 GPLv3 条款提供。
 * @mp-source app/modules/indexer/parser/ipt.py @ e6820a8
 *
 * IPTorrents 站点用户数据解析器(MP `IptSiteUserInfo`)。
 * 函数对照:`_parse_user_base_info` → `_parseUserBaseInfo`、`_parse_site_page`(空)、
 * `_parse_user_detail_info` → `_parseUserDetailInfo`、`_parse_user_torrent_seeding_info` → `_parseUserTorrentSeedingInfo`、
 * 其余为空实现。XPath 逐字保留;统计块下标越界时与 MP 一样抛 IndexError。
 */
import { type NextPage, SiteParserBase } from "./base.ts";
import { normalizeDatetime, parseFloatMp, parseIntMp, reSearch } from "./common.ts";
import { hasChildElements, parseHtml } from "./dom.ts";
import { SiteSchema, type SiteSchemaValue } from "./types.ts";

/** Python 列表下标(越界抛 IndexError,保持 MP 失败语义) */
function at<T>(arr: T[], i: number): T {
  const v = i < 0 ? arr[arr.length + i] : arr[i];
  if (v === undefined) throw new Error("IndexError: list index out of range");
  return v;
}

export class IptSiteUserInfo extends SiteParserBase {
  static override schema: SiteSchemaValue | null = SiteSchema.Ipt;

  protected _parseUserBaseInfo(htmlText: string): void {
    const html = parseHtml(this._prepareHtmlText(htmlText));
    if (!html) return;
    const names = html.strings('//a[contains(@href, "/u/")]//text()');
    const ids = html.strings('//a[contains(@href, "/u/")]/@href');
    if (names.length) this.username = String(at(names, -1));
    if (ids.length) {
      const m = reSearch(/\/u\/(\d+)/, ids[0]);
      if (m?.[0].trim()) {
        this.userid = m[1]!;
        this._userDetailPage = `user.php?u=${this.userid}`;
        this._torrentSeedingPage = `peers?u=${this.userid}`;
      }
    }
    const stats = html.nodes('//div[@class = "stats"]/div/div');
    if (stats.length) {
      const s = stats[0]!;
      const spans = s.strings("span/text()");
      const links = s.nodes("a");
      this.upload = this.numFilesize(String(at(spans, 1)).trim());
      this.download = this.numFilesize(String(at(spans, 2)).trim());
      this.seeding = parseIntMp(at(at(links, 2).strings("text()"), 0));
      this.leeching = parseIntMp(at(at(links, 2).strings("text()"), 1));
      this.ratio = parseFloatMp(String(at(spans, 0)).trim().replaceAll("-", "0"));
      this.bonus = parseFloatMp(at(at(links, 3).strings("text()"), 0));
    }
  }

  protected _parseSitePage(_htmlText: string): void {}

  protected _parseUserDetailInfo(htmlText: string): void {
    const html = parseHtml(htmlText);
    if (!hasChildElements(html)) return;
    const levels = html.strings('//tr/th[text()="Class"]/following-sibling::td[1]/text()');
    if (levels.length) this.userLevel = levels[0]!.trim();
    const joinAt = html.strings('//tr/th[text()="Join date"]/following-sibling::td[1]/text()');
    if (joinAt.length) this.joinAt = normalizeDatetime(joinAt[0]!.split(" (")[0]) ?? null;
  }

  protected _parseUserTorrentSeedingInfo(htmlText: string, _multiPage = false): NextPage {
    const html = parseHtml(htmlText);
    if (!hasChildElements(html)) return null;
    let seedingEndPos = 3;
    if (html.nodes('//tr/td[text() = "Leechers"]').length) {
      seedingEndPos = html.nodes('//tr/td[text() = "Leechers"]/../preceding-sibling::tr').length + 1;
      seedingEndPos = seedingEndPos - 3;
    }
    let pageSeeding = 0;
    let pageSeedingSize = 0;
    const torrents = html.strings('//tr/td[text() = "Seeders"]/../following-sibling::tr/td[position()=6]/text()');
    if (torrents.length) {
      pageSeeding = seedingEndPos;
      // Python 切片 [:n](n 为负时从尾部截)
      for (let perSize of torrents.slice(0, seedingEndPos)) {
        if (perSize.includes("(") && perSize.includes(")")) {
          perSize = perSize.split("(").at(-1)!;
          perSize = perSize.split(")")[0]!;
        }
        pageSeedingSize += this.numFilesize(perSize);
      }
    }
    this.seeding = pageSeeding;
    this.seedingSize = pageSeedingSize;
    return null;
  }

  protected _parseUserTrafficInfo(_htmlText: string): void {}

  protected _parseMessageUnreadLinks(_htmlText: string, _msgLinks: string[]): NextPage {
    return null;
  }

  protected _parseMessageContent(_htmlText: string): [string | null, string | null, string | null] {
    return [null, null, null];
  }
}