@zonuexe/techbook-mcp 0.2.3 → 0.2.4

package/src/adapters/publishers/manatee.ts

@@ -1,113 +0,0 @@
-import type { PublisherAdapter, PublisherDeps } from "../../domain/publisher.js";
-import type { BookRecord, SearchQuery } from "../../domain/book.js";
-import type { HtmlDocument } from "../../ports/html-parser.js";
-import { fetchText, parseJapanesePrice, resolveUrl } from "./base.js";
-
-const BASE_URL = "https://book.mynavi.jp/manatee";
-const BOOKS_URL = `${BASE_URL}/books/`;
-
-/**
- * `.attribute li` 内の著者リンクを配列にする。
- * 各 <a> のテキストが著者名（役割は括弧内テキストで付記されているが名前は <a> 内）。
- */
-function parseAuthors(doc: HtmlDocument): string[] {
-  return doc
-    .select(".attribute li a")
-    .map(el => el.text().trim())
-    .filter(Boolean);
-}
-
-export const manateeAdapter: PublisherAdapter = {
-  id: "manatee",
-  name: "マナティ",
-  baseUrl: BASE_URL,
-
-  async search(query: SearchQuery, deps: PublisherDeps): Promise<BookRecord[]> {
-    const word = [query.title, query.author].filter(Boolean).join(" ");
-    if (!word) return [];
-
-    const url = `${BOOKS_URL}?topics_keyword=${encodeURIComponent(word)}`;
-    const html = await fetchText(url, deps);
-    const doc = deps.parser.parse(html);
-
-    const results: BookRecord[] = [];
-    const limit = query.limit ?? 10;
-
-    // 書籍アイテムは <!-- item --> コメントで区切られた <li> 内
-    for (const li of doc.select("ul.category_list li")) {
-      const titleEl = li.find("dl.detail dt a")[0];
-      if (!titleEl) continue;
-
-      const title = titleEl.text().trim();
-      if (!title) continue;
-
-      const href = titleEl.attr("href");
-      if (!href) continue;
-      const bookUrl = href.startsWith("http") ? href : resolveUrl(BASE_URL, href);
-
-      const priceText = li.find(".detail__price")[0]?.text().trim();
-      const price = priceText ? parseJapanesePrice(priceText) : undefined;
-
-      const imgEl = li.find(".image img")[0];
-      const imgSrc = imgEl?.attr("src");
-      const coverImageUrl = imgSrc ? resolveUrl(BASE_URL, imgSrc) : undefined;
-
-      results.push({
-        title,
-        authors: [],
-        publisher: "マナティ",
-        url: bookUrl,
-        price,
-        coverImageUrl,
-        ebookStores: [{ name: "マナティ", url: bookUrl, drm: "social" }],
-      });
-
-      if (results.length >= limit) break;
-    }
-
-    return results;
-  },
-
-  async getDetail(url: string, deps: PublisherDeps): Promise<BookRecord> {
-    const html = await fetchText(url, deps);
-    const doc = deps.parser.parse(html);
-
-    const title = doc.selectOne("h1.title")?.text().trim() ?? "";
-
-    // `.intro` の最初の <p> が出版社名
-    const publisher = doc.selectOne(".intro p")?.text().trim() ?? "マナティ";
-
-    const authors = parseAuthors(doc);
-
-    // 購入形態テーブルの最初の行から価格取得
-    const priceText = doc.selectOne("#item_selectlist table .price")?.text().trim();
-    const price = priceText ? parseJapanesePrice(priceText) : undefined;
-
-    // 表紙画像の alt 属性が "ISBN.jpg" 形式（例: "9784839984274.jpg"）
-    const imgEl = doc.selectOne("div.item_meta_image img");
-    const imgSrc = imgEl?.attr("src");
-    const coverImageUrl = imgSrc ? resolveUrl(BASE_URL, imgSrc) : undefined;
-    const imgAlt = imgEl?.attr("alt") ?? "";
-    const isbn = imgAlt.replace(/\.jpg$/i, "").replace(/-/g, "") || undefined;
-
-    // 発売日: "発売日：2024-03-22"
-    const dateText = doc.selectOne("p.date")?.text().trim();
-    const publishedAt = dateText?.match(/(\d{4}-\d{2}-\d{2})/)?.[1] ?? undefined;
-
-    const descEl = doc.selectOne(".item_desc p:not(.date):not(.pages)");
-    const description = descEl?.text().trim() || undefined;
-
-    return {
-      title,
-      authors,
-      publisher,
-      url,
-      isbn,
-      price,
-      publishedAt,
-      description,
-      coverImageUrl,
-      ebookStores: [{ name: "マナティ", url, drm: "social" }],
-    };
-  },
-};

package/src/adapters/publishers/maruzen-publishing.ts

@@ -1,129 +0,0 @@
-import type { PublisherAdapter, PublisherDeps } from "../../domain/publisher.js";
-import type { BookRecord, SearchQuery } from "../../domain/book.js";
-import { fetchText, resolveUrl, extractEbookStoresFromDoc, parseJapaneseDateToISO, stripAuthorRole } from "./base.js";
-
-const BASE_URL = "https://www.maruzen-publishing.co.jp";
-const SEARCH_URL = `${BASE_URL}/search/`;
-
-/** 403回避のために自サイトをRefererとして付与する必要がある */
-const EXTRA_HEADERS = { Referer: `${BASE_URL}/` };
-
-/**
- * "ボリス・チェルニー 著　折山文哉 訳" → ["ボリス・チェルニー", "折山文哉"]
- * 役割語（著・訳・編・監訳・監修など）を除去する。
- */
-function parseAuthorsFromText(text: string): string[] {
-  return text.split(/[　\s]+(?=\S)/).map(stripAuthorRole).filter(Boolean);
-}
-
-/**
- * div.author 内の各リンクから役割語を除去して著者名のみ返す。
- */
-function parseAuthorLinks(authors: string[]): string[] {
-  return authors.map(stripAuthorRole).filter(Boolean);
-}
-
-/**
- * "2020/03/31" → "2020-03-31"
- * "2020年3月31日" → "2020-03-31"
- */
-function parseDate(text: string): string | undefined {
-  const m1 = text.match(/(\d{4})\/(\d{1,2})\/(\d{1,2})/);
-  if (m1) {
-    return `${m1[1]}-${m1[2].padStart(2, "0")}-${m1[3].padStart(2, "0")}`;
-  }
-  return parseJapaneseDateToISO(text);
-}
-
-export const maruzenPublishingAdapter: PublisherAdapter = {
-  id: "maruzen-publishing",
-  name: "丸善出版",
-  baseUrl: BASE_URL,
-
-  async search(query: SearchQuery, deps: PublisherDeps): Promise<BookRecord[]> {
-    const word = [query.title, query.author].filter(Boolean).join(" ");
-    if (!word) return [];
-
-    const url = `${SEARCH_URL}?search_keyword=${encodeURIComponent(word)}&format=1`;
-    const html = await fetchText(url, deps, EXTRA_HEADERS);
-    const doc = deps.parser.parse(html);
-
-    const results: BookRecord[] = [];
-    const limit = query.limit ?? 10;
-
-    for (const item of doc.select("div.booklist div.item")) {
-      const linkEl = item.find("div.ttl a")[0];
-      if (!linkEl) continue;
-
-      const title = linkEl.text().trim();
-      const href = linkEl.attr("href");
-      if (!title || !href) continue;
-
-      const bookUrl = href.startsWith("http") ? href : `${BASE_URL}${href}`;
-
-      const authorLinks = item.find("div.author a").map(el => el.text().trim());
-      const authors = parseAuthorLinks(authorLinks);
-
-      const imgEl = item.find("div.image img")[0];
-      const imgSrc = imgEl?.attr("src");
-      const coverImageUrl = imgSrc ? resolveUrl(BASE_URL, imgSrc) : undefined;
-
-      results.push({
-        title,
-        authors,
-        publisher: "丸善出版",
-        url: bookUrl,
-        coverImageUrl,
-      });
-
-      if (results.length >= limit) break;
-    }
-
-    return results;
-  },
-
-  async getDetail(url: string, deps: PublisherDeps): Promise<BookRecord> {
-    const html = await fetchText(url, deps, EXTRA_HEADERS);
-    const doc = deps.parser.parse(html);
-
-    // タイトルは <title> タグ、またはh1など（サイト構造に依存）
-    // div#bookData table の th/td からメタデータを取得
-    const table = doc.selectOne("div#bookData table");
-    let authors: string[] = [];
-    let publisher = "丸善出版";
-    let publishedAt: string | undefined;
-
-    if (table) {
-      for (const row of table.find("tr")) {
-        const th = row.find("th")[0]?.text().trim() ?? "";
-        const tdText = row.find("td")[0]?.text().trim() ?? "";
-
-        if (th === "著者") {
-          authors = parseAuthorsFromText(tdText);
-        } else if (th === "発行元") {
-          publisher = tdText || "丸善出版";
-        } else if (th === "出版年月日") {
-          publishedAt = parseDate(tdText);
-        }
-      }
-    }
-
-    // ページタイトルから書籍タイトルを取得（" | 丸善出版" を除去）
-    const pageTitle = doc.selectOne("title")?.text().trim() ?? "";
-    const title = pageTitle.replace(/\s*[|｜]\s*丸善出版.*$/, "").trim();
-
-    // 電子書籍ストア（kw.maruzen.co.jp は機関向けなので除外）
-    const ebookStores = extractEbookStoresFromDoc(doc).filter(
-      store => !store.url.includes("kw.maruzen.co.jp"),
-    );
-
-    return {
-      title,
-      authors,
-      publisher,
-      url,
-      publishedAt,
-      ebookStores: ebookStores.length > 0 ? ebookStores : undefined,
-    };
-  },
-};

package/src/adapters/publishers/optronics.ts

@@ -1,113 +0,0 @@
-import type { PublisherAdapter, PublisherDeps } from "../../domain/publisher.js";
-import type { BookRecord, SearchQuery } from "../../domain/book.js";
-import { fetchText, parseJapanesePrice, resolveUrl } from "./base.js";
-
-const BASE_URL = "https://optronics-ebook.com";
-const LIST_URL = `${BASE_URL}/products/list.php`;
-
-/** 電子書籍カテゴリID */
-const EBOOK_CATEGORY_ID = "1";
-
-/**
- * listcomment / main_comment のテキストから各フィールドを抽出する。
- * フォーマット例:
- *   "著者:波多腰 玄一\n発行:㈱オプトロニクス社\n頁数:384頁\n..."
- */
-function parseComment(text: string): { authors: string[]; publisher?: string } {
-  const authorMatch = text.match(/著者[：:]\s*([^\n<]+)/);
-  const authors = authorMatch
-    ? authorMatch[1].split(/[、,]/).map(s => s.trim()).filter(Boolean)
-    : [];
-
-  const publisherMatch = text.match(/発行[：:]\s*([^\n<]+)/);
-  const publisher = publisherMatch
-    ? publisherMatch[1].replace(/[㈱㈲]/g, "").trim() || undefined
-    : undefined;
-
-  return { authors, publisher };
-}
-
-export const optronicsAdapter: PublisherAdapter = {
-  id: "optronics",
-  name: "オプトロニクス社",
-  baseUrl: BASE_URL,
-
-  async search(query: SearchQuery, deps: PublisherDeps): Promise<BookRecord[]> {
-    const word = [query.title, query.author].filter(Boolean).join(" ");
-    if (!word) return [];
-
-    const url = `${LIST_URL}?name=${encodeURIComponent(word)}&category_id=${EBOOK_CATEGORY_ID}`;
-    const html = await fetchText(url, deps);
-    const doc = deps.parser.parse(html);
-
-    const results: BookRecord[] = [];
-    const limit = query.limit ?? 10;
-
-    for (const block of doc.select("div.list_area")) {
-      const titleEl = block.find("div.listrightbloc h3 a")[0];
-      if (!titleEl) continue;
-
-      const title = titleEl.text().trim();
-      const href = titleEl.attr("href");
-      if (!title || !href) continue;
-
-      const bookUrl = href.startsWith("http") ? href : `${BASE_URL}${href}`;
-
-      const priceText = block.find("div.pricebox .price")[0]?.text().trim();
-      const price = priceText ? parseJapanesePrice(priceText) : undefined;
-
-      const commentText = block.find("div.listcomment")[0]?.text().trim() ?? "";
-      const { authors, publisher } = parseComment(commentText);
-
-      const imgEl = block.find("div.listphoto img.picture")[0];
-      const imgSrc = imgEl?.attr("src");
-      const coverImageUrl = imgSrc ? resolveUrl(BASE_URL, imgSrc) : undefined;
-
-      results.push({
-        title,
-        authors,
-        publisher: publisher ?? "オプトロニクス社",
-        url: bookUrl,
-        price,
-        coverImageUrl,
-        ebookStores: [{ name: "オプトロニクス社", url: bookUrl, drm: "free" }],
-      });
-
-      if (results.length >= limit) break;
-    }
-
-    return results;
-  },
-
-  async getDetail(url: string, deps: PublisherDeps): Promise<BookRecord> {
-    const html = await fetchText(url, deps);
-    const doc = deps.parser.parse(html);
-
-    // タイトル: <title>OPTRONICS eBOOK / {title}</title>
-    const rawTitle = doc.selectOne("title")?.text().trim() ?? "";
-    const title = rawTitle.replace(/^OPTRONICS\s+eBOOK\s*[/／]\s*/, "").trim();
-
-    // 価格: #price02_default
-    const priceText = doc.selectOne("#price02_default")?.text().trim();
-    const price = priceText ? parseJapanesePrice(priceText) : undefined;
-
-    // 著者・発行元: div.main_comment テキスト
-    const commentText = doc.selectOne("div.main_comment")?.text().trim() ?? "";
-    const { authors, publisher } = parseComment(commentText);
-
-    // カバー画像: img.picture
-    const imgEl = doc.selectOne("img.picture");
-    const imgSrc = imgEl?.attr("src");
-    const coverImageUrl = imgSrc ? resolveUrl(BASE_URL, imgSrc) : undefined;
-
-    return {
-      title,
-      authors,
-      publisher: publisher ?? "オプトロニクス社",
-      url,
-      price,
-      coverImageUrl,
-      ebookStores: [{ name: "オプトロニクス社", url, drm: "free" }],
-    };
-  },
-};

package/src/adapters/publishers/oreilly-japan.ts

@@ -1,133 +0,0 @@
-import type { PublisherAdapter, PublisherDeps } from "../../domain/publisher.js";
-import type { BookRecord, SearchQuery } from "../../domain/book.js";
-import { fetchText, parseJapanesePrice, resolveUrl, parseJapaneseDateToISO, stripAuthorRole } from "./base.js";
-
-const BASE_URL = "https://www.oreilly.co.jp";
-const EBOOK_LIST_URL = `${BASE_URL}/ebook/`;
-
-/**
- * "2026年04月03日" → "2026-04-03"
- * "2025-04-08" (content属性) はそのまま返す
- */
-function parseOreillyDate(text: string): string | undefined {
-  const isoMatch = text.match(/\d{4}-\d{2}-\d{2}/);
-  if (isoMatch) return isoMatch[0];
-  return parseJapaneseDateToISO(text);
-}
-
-/**
- * 著者文字列をパースして配列に変換する。
- * 例: "Dan Vanderkam　著、今村 謙士　訳" → ["Dan Vanderkam", "今村 謙士"]
- */
-function parseAuthors(text: string): string[] {
-  return text.split(/[、,]/).map(stripAuthorRole).filter(Boolean);
-}
-
-export const oreillyJapanAdapter: PublisherAdapter = {
-  id: "oreilly-japan",
-  name: "オライリー・ジャパン",
-  baseUrl: BASE_URL,
-
-  async search(query: SearchQuery, deps: PublisherDeps): Promise<BookRecord[]> {
-    // 検索APIがないためタイトルでローカルフィルタリングする
-    // 著者のみの検索は各書籍詳細ページを全取得しないと不可能なため非対応
-    if (!query.title) return [];
-
-    const keyword = query.title.toLowerCase();
-    const limit = query.limit ?? 10;
-
-    const html = await fetchText(EBOOK_LIST_URL, deps);
-    const doc = deps.parser.parse(html);
-
-    const results: BookRecord[] = [];
-
-    for (const row of doc.select(".ebookCatalog tbody tr")) {
-      const titleEl = row.find("td.title a")[0];
-      if (!titleEl) continue;
-
-      const title = titleEl.text().trim();
-      if (!title.toLowerCase().includes(keyword)) continue;
-
-      const href = titleEl.attr("href");
-      if (!href) continue;
-      const url = resolveUrl(EBOOK_LIST_URL, href);
-
-      const isbnRaw = row.find("td.isbn")[0]?.text().trim();
-      const isbn = isbnRaw?.replace(/-/g, "");
-
-      const priceText = row.find("td.price")[0]?.text().trim();
-      const price = priceText ? parseJapanesePrice(priceText) : undefined;
-
-      const dateText = row.find("td")[3]?.text().trim();
-      const publishedAt = dateText ? parseOreillyDate(dateText) : undefined;
-
-      const coverImageUrl = isbn
-        ? `${BASE_URL}/books/images/picture_large${isbnRaw}.jpeg`
-        : undefined;
-
-      results.push({
-        title,
-        authors: [],
-        publisher: "オライリー・ジャパン",
-        url,
-        isbn,
-        price,
-        publishedAt,
-        coverImageUrl,
-        ebookStores: [{ name: "オライリー・ジャパン", url, drm: "free" }],
-      });
-
-      if (results.length >= limit) break;
-    }
-
-    return results;
-  },
-
-  async getDetail(url: string, deps: PublisherDeps): Promise<BookRecord> {
-    const html = await fetchText(url, deps);
-    const doc = deps.parser.parse(html);
-
-    const titleMain = doc.selectOne("h1[itemprop='name']")?.text().trim() ?? "";
-    const subTitle = doc.selectOne("p.sub_title")?.text()
-      .replace(/^[\s\u3000―-]+/, "").trim();
-    const title = subTitle ? `${titleMain} ―${subTitle}` : titleMain;
-
-    const authorText = doc.selectOne("span[itemprop='author']")?.text().trim() ?? "";
-    const authors = parseAuthors(authorText);
-
-    const isbnRaw = doc.selectOne("dd[itemprop='isbn']")?.text().trim();
-    const isbn = isbnRaw?.replace(/-/g, "");
-
-    const publishedAt = doc.selectOne("dd[itemprop='datePublished']")?.attr("content")
-      ?? undefined;
-
-    const coverImageUrl = doc.selectOne("img.cover-photo")?.attr("src") ?? undefined;
-
-    const description = doc.selectOne("p[itemprop='description']")?.text().trim()
-      || undefined;
-
-    // Ebook価格を取得: "Ebook" という option-name の次の div
-    let price: number | undefined;
-    for (const item of doc.select(".option-item")) {
-      const name = item.find(".option-name")[0]?.text().trim();
-      if (name === "Ebook") {
-        const priceText = item.find("div")[1]?.text().trim();
-        if (priceText) price = parseJapanesePrice(priceText);
-        break;
-      }
-    }
-
-    return {
-      title,
-      authors,
-      publisher: "オライリー・ジャパン",
-      url,
-      isbn,
-      price,
-      publishedAt,
-      description,
-      coverImageUrl,
-      ebookStores: [{ name: "オライリー・ジャパン", url, drm: "free" }],
-    };
-  },
-};

package/src/adapters/publishers/peaks.ts

@@ -1,98 +0,0 @@
-import type { PublisherAdapter, PublisherDeps } from "../../domain/publisher.js";
-import type { BookRecord, SearchQuery } from "../../domain/book.js";
-import { fetchText, parseJapanesePrice } from "./base.js";
-
-const BASE_URL = "https://peaks.cc";
-
-export const peaksAdapter: PublisherAdapter = {
-  id: "peaks",
-  name: "PEAKS",
-  baseUrl: BASE_URL,
-
-  async search(query: SearchQuery, deps: PublisherDeps): Promise<BookRecord[]> {
-    // 検索APIなし。タイトルでトップページの全書籍をローカルフィルタリング。
-    // 著者のみの検索は各詳細ページを全取得しないと不可能なため非対応。
-    if (!query.title) return [];
-
-    const keyword = query.title.toLowerCase();
-    const html = await fetchText(BASE_URL, deps);
-    const doc = deps.parser.parse(html);
-    const limit = query.limit ?? 10;
-
-    const results: BookRecord[] = [];
-
-    // トップページ掲載の全書籍: article.p-project__unit
-    for (const article of doc.select("article.p-project__unit")) {
-      const imgEl = article.find("div.p-project__unit_image img")[0];
-      const title = imgEl?.attr("alt")?.trim();
-      if (!title) continue;
-
-      if (!title.toLowerCase().includes(keyword)) continue;
-
-      const linkEl = article.find("div.p-project__unit_image a")[0];
-      const href = linkEl?.attr("href");
-      if (!href) continue;
-
-      const bookUrl = `${BASE_URL}${href}`;
-      const coverImageUrl = imgEl.attr("src") ?? undefined;
-
-      results.push({
-        title,
-        authors: [],
-        publisher: "PEAKS",
-        url: bookUrl,
-        coverImageUrl,
-        ebookStores: [{ name: "PEAKS", url: bookUrl, drm: "free" }],
-      });
-
-      if (results.length >= limit) break;
-    }
-
-    return results;
-  },
-
-  async getDetail(url: string, deps: PublisherDeps): Promise<BookRecord> {
-    const html = await fetchText(url, deps);
-    const doc = deps.parser.parse(html);
-
-    const getMeta = (name: string, attr = "content"): string | undefined =>
-      doc.selectOne(`meta[name="${name}"]`)?.attr(attr)?.trim() ||
-      doc.selectOne(`meta[property="${name}"]`)?.attr("content")?.trim() ||
-      undefined;
-
-    // タイトル: twitter:title（『』なし）が最もクリーン
-    const title =
-      getMeta("twitter:title") ||
-      doc.selectOne("h2.p-project__intro_title")?.text()
-        .trim().replace(/^[『「]|[』」]$/g, "") ||
-      "";
-
-    // 著者: p-project__intro_author 内のリンクテキスト（末尾カンマを除去）
-    const authors = doc.select("ul.p-project__intro_author a")
-      .map(el => el.text().trim().replace(/,\s*$/, ""))
-      .filter(Boolean);
-
-    // 価格: 電子版 option の span テキスト
-    const priceEl = doc.selectOne("option[kind='electronic'] span") ??
-                    doc.selectOne("select#editions option");
-    const price = priceEl
-      ? parseJapanesePrice(priceEl.text())
-      : undefined;
-
-    // カバー画像: twitter:image または og:image
-    const coverImageUrl =
-      getMeta("twitter:image") ||
-      getMeta("og:image") ||
-      undefined;
-
-    return {
-      title,
-      authors,
-      publisher: "PEAKS",
-      url,
-      price,
-      coverImageUrl,
-      ebookStores: [{ name: "PEAKS", url, drm: "free" }],
-    };
-  },
-};