@prose-reader/metadata-fetcher 1.338.0 → 1.340.0

package/dist/index.js

@@ -1,37 +1,38 @@
 import { normalizeIsbn as e } from "@prose-reader/archive-reader";
 import { Report as t } from "@prose-reader/shared";
+import { XmlDocument as n } from "xmldoc";
 //#region src/utils/metadataAuthors.ts
-var n = (e) => {
+var r = (e) => {
 	let t = e.contributors ?? [], n = t.filter((e) => e.roles.includes("author")), r = (n.length > 0 ? n : t).map((e) => e.name.trim()).filter((e) => e.length > 0);
 	return [...new Set(r)];
-}, r = /^\d{13}$/, i = /^\d{9}[\dX]$/, a = (t) => {
+}, i = /^\d{13}$/, a = /^\d{9}[\dX]$/, o = (t) => {
 	let n = e(t);
 	if (n === void 0) return;
-	if (r.test(n)) return n;
-	if (!i.test(n)) return;
-	let a = `978${n.slice(0, 9)}`, o = 0;
-	for (let e = 0; e < a.length; e++) {
-		let t = a.charCodeAt(e) - 48;
+	if (i.test(n)) return n;
+	if (!a.test(n)) return;
+	let r = `978${n.slice(0, 9)}`, o = 0;
+	for (let e = 0; e < r.length; e++) {
+		let t = r.charCodeAt(e) - 48;
 		o += e % 2 == 0 ? t : t * 3;
 	}
-	return `${a}${(10 - o % 10) % 10}`;
-}, o = /[\u0300-\u036f]/g, s = /[^\p{L}\p{N}]+/gu, c = (e) => e.normalize("NFD").replace(o, "").toLowerCase().replace(s, " ").trim(), l = (e) => {
+	return `${r}${(10 - o % 10) % 10}`;
+}, s = /[\u0300-\u036f]/g, c = /[^\p{L}\p{N}]+/gu, l = (e) => e.normalize("NFD").replace(s, "").toLowerCase().replace(c, " ").trim(), u = (e) => {
 	let t = /* @__PURE__ */ new Map();
 	for (let n = 0; n < e.length - 1; n++) {
 		let r = e.slice(n, n + 2);
 		t.set(r, (t.get(r) ?? 0) + 1);
 	}
 	return t;
-}, u = (e, t) => {
-	let n = c(e).replaceAll(" ", ""), r = c(t).replaceAll(" ", "");
+}, d = (e, t) => {
+	let n = l(e).replaceAll(" ", ""), r = l(t).replaceAll(" ", "");
 	if (n.length === 0 || r.length === 0) return 0;
 	if (n === r) return 1;
 	if (n.length < 2 || r.length < 2) return 0;
-	let i = l(n), a = 0;
-	for (let [e, t] of l(r)) a += Math.min(t, i.get(e) ?? 0);
+	let i = u(n), a = 0;
+	for (let [e, t] of u(r)) a += Math.min(t, i.get(e) ?? 0);
 	return 2 * a / (n.length - 1 + (r.length - 1));
-}, d = /\s*[:;]\s*|\s+[-–—]\s+/, f = (e) => {
-	let [t = "", ...n] = e.split(d), r = t.trim(), i = n.join(" ").trim();
+}, f = /\s*[:;]\s*|\s+[-–—]\s+/, p = (e) => {
+	let [t = "", ...n] = e.split(f), r = t.trim(), i = n.join(" ").trim();
 	return r.length === 0 ? {
 		main: e,
 		hasSubtitle: !1
@@ -39,110 +40,192 @@ var n = (e) => {
 		main: r,
 		hasSubtitle: i.length > 0
 	};
-}, p = .9, m = (e, t) => {
-	let n = u(e, t), r = f(e), i = f(t);
-	return r.hasSubtitle === i.hasSubtitle ? n : Math.max(n, u(r.main, i.main) * p);
-}, h = (e) => c(e).split(" ").filter((e) => e.length > 0).sort().join(" "), g = (e, t) => u(h(e), h(t)), _ = {
+}, m = .9, h = [
+	"volume",
+	"part",
+	"book"
+], ee = /\b(volumes?|vols?|parts?|books?)\s+([\p{L}\p{N}]+)\b/gu, g = {
+	one: 1,
+	first: 1,
+	two: 2,
+	second: 2,
+	three: 3,
+	third: 3,
+	four: 4,
+	fourth: 4,
+	five: 5,
+	fifth: 5,
+	six: 6,
+	sixth: 6,
+	seven: 7,
+	seventh: 7,
+	eight: 8,
+	eighth: 8,
+	nine: 9,
+	ninth: 9,
+	ten: 10,
+	tenth: 10,
+	eleven: 11,
+	eleventh: 11,
+	twelve: 12,
+	twelfth: 12
+}, _ = {
+	i: 1,
+	v: 5,
+	x: 10,
+	l: 50,
+	c: 100,
+	d: 500,
+	m: 1e3
+}, v = (e) => {
+	if (!/^m{0,3}(cm|cd|d?c{0,3})(xc|xl|l?x{0,3})(ix|iv|v?i{0,3})$/i.test(e)) return;
+	let t = 0, n = 0;
+	for (let r of [...e.toLowerCase()].reverse()) {
+		let e = _[r];
+		if (e === void 0) return;
+		e < n ? t -= e : (t += e, n = e);
+	}
+	return t > 0 ? t : void 0;
+}, y = (e) => /^\d+$/.test(e) ? Number.parseInt(e, 10) : g[e] ?? v(e), b = (e) => {
+	if (e.startsWith("vol")) return "volume";
+	if (e.startsWith("part")) return "part";
+	if (e.startsWith("book")) return "book";
+}, x = (e) => {
+	let t = /* @__PURE__ */ new Map();
+	for (let n of l(e).matchAll(ee)) {
+		let e = n[1] === void 0 ? void 0 : b(n[1]), r = n[2] === void 0 ? void 0 : y(n[2]);
+		if (e === void 0 || r === void 0) continue;
+		let i = t.get(e) ?? /* @__PURE__ */ new Set();
+		i.add(r), t.set(e, i);
+	}
+	return t;
+}, S = (e, t) => {
+	let n = x(e), r = x(t);
+	for (let e of h) {
+		let t = n.get(e), i = r.get(e);
+		if (!(t === void 0 || i === void 0) && [...t].every((e) => !i.has(e))) return !0;
+	}
+	return !1;
+}, C = (e, t) => {
+	if (S(e, t)) return 0;
+	let n = d(e, t), r = p(e), i = p(t);
+	return r.hasSubtitle === i.hasSubtitle ? n : Math.max(n, d(r.main, i.main) * m);
+}, w = (e) => l(e).split(" ").filter((e) => e.length > 0).sort().join(" "), T = (e, t) => d(w(e), w(t)), E = {
 	isbn: 1,
 	gtin: 1,
 	identifiers: .6,
 	title: .8,
 	contributors: .5,
 	series: .3,
-	published: .2,
-	publisher: .15,
+	"publication.original.date": .2,
+	"publication.original.publisher": .15,
+	"publication.edition.date": .2,
+	"publication.edition.publisher": .15,
 	languages: .15,
 	numberOfPages: .15
-}, v = /* @__PURE__ */ new Set(["isbn", "gtin"]), y = (e) => {
+}, te = /* @__PURE__ */ new Set(["isbn", "gtin"]), D = (e) => {
 	let t = e?.trim();
 	return t !== void 0 && t.length > 0 ? t : void 0;
-}, b = (e) => e.value.trim().toLowerCase(), x = (e) => e.scheme === void 0 ? e.value : `${e.scheme}:${e.value}`, S = (e, t) => b(e) === b(t) ? e.scheme === void 0 || t.scheme === void 0 || e.scheme.trim().toLowerCase() === t.scheme.trim().toLowerCase() : !1, C = (e) => e.trim().toLowerCase().split("-")[0] ?? "", w = (e, t) => {
+}, ne = (e, t) => {
+	let n = D(e.value)?.toLowerCase();
+	return n !== void 0 && n === D(t.value)?.toLowerCase();
+}, O = (e) => D(e.scheme)?.toLowerCase(), re = (e) => e.scheme === void 0 ? e.value : `${e.scheme}:${e.value}`, ie = (e, t) => {
+	if (!ne(e, t)) return !1;
+	let n = O(e), r = O(t);
+	return n === void 0 || r === void 0 || n === r;
+}, ae = (e, t) => {
+	if (!ne(e, t)) return !1;
+	let n = O(e), r = O(t);
+	return n !== void 0 && n === r;
+}, k = (e) => e.trim().toLowerCase().split("-")[0] ?? "", oe = (e, t) => {
 	let n = Math.abs(e - t);
 	return n === 0 ? 1 : n === 1 ? .6 : n <= 3 ? .3 : 0;
-}, T = (e, t) => {
+}, se = (e, t) => {
 	let n = Math.max(e, t);
 	return n <= 0 ? 0 : Math.max(0, 1 - Math.abs(e - t) / n * 2);
-}, E = (e, t, n) => e.reduce((e, r) => t.reduce((e, t) => Math.max(e, n(r, t)), e), 0), D = (e, t) => {
-	let r = [], i = (e, t) => {
-		t !== void 0 && r.push({
+}, ce = (e, t, n) => e.reduce((e, r) => t.reduce((e, t) => Math.max(e, n(r, t)), e), 0), A = (e, t) => {
+	let n = [], i = (e, t) => {
+		t !== void 0 && n.push({
 			field: e,
-			weight: _[e],
+			weight: E[e],
 			score: Math.min(1, Math.max(0, t.score)),
 			query: t.query,
 			candidate: t.candidate
 		});
-	}, o = (e, t, n, r = u) => {
-		let a = y(t), o = y(n);
+	}, a = (e, t, n, r = d) => {
+		let a = D(t), o = D(n);
 		a === void 0 || o === void 0 || i(e, {
 			score: r(a, o),
 			query: a,
 			candidate: o
 		});
-	}, s = y(e.isbn), c = y(t.isbn), l = s === void 0 ? void 0 : a(s), d = c === void 0 ? void 0 : a(c);
-	s !== void 0 && c !== void 0 && l !== void 0 && d !== void 0 && i("isbn", {
-		score: +(l === d),
+	}, s = D(e.isbn), c = D(t.isbn), l = s === void 0 ? void 0 : o(s), u = c === void 0 ? void 0 : o(c);
+	s !== void 0 && c !== void 0 && l !== void 0 && u !== void 0 && i("isbn", {
+		score: +(l === u),
 		query: s,
 		candidate: c
 	});
-	let f = y(e.gtin), p = y(t.gtin);
+	let f = D(e.gtin), p = D(t.gtin);
 	f !== void 0 && p !== void 0 && i("gtin", {
 		score: +(f === p),
 		query: f,
 		candidate: p
 	});
-	let h = e.identifiers ?? [], b = t.identifiers ?? [];
-	h.length > 0 && b.length > 0 && i("identifiers", {
-		score: +!!h.some((e) => b.some((t) => S(e, t))),
-		query: h.map(x).join(", "),
-		candidate: b.map(x).join(", ")
-	}), o("title", e.title, t.title, m), o("publisher", e.publisher, t.publisher), o("series", e.belongsTo?.series?.[0]?.name, t.belongsTo?.series?.[0]?.name);
-	let D = n(t), O = n(e);
-	O.length > 0 && D.length > 0 && i("contributors", {
-		score: E(O, D, g),
-		query: O.join(", "),
-		candidate: D.join(", ")
+	let m = e.identifiers ?? [], h = t.identifiers ?? [], ee = m.some((e) => h.some((t) => ae(e, t)));
+	m.length > 0 && h.length > 0 && i("identifiers", {
+		score: +!!m.some((e) => h.some((t) => ie(e, t))),
+		query: m.map(re).join(", "),
+		candidate: h.map(re).join(", ")
+	}), a("title", e.title, t.title, C), a("publication.original.publisher", e.publication?.original?.publisher, t.publication?.original?.publisher), a("publication.edition.publisher", e.publication?.edition?.publisher, t.publication?.edition?.publisher), a("series", e.belongsTo?.series?.[0]?.name, t.belongsTo?.series?.[0]?.name);
+	let g = r(t), _ = r(e);
+	_.length > 0 && g.length > 0 && i("contributors", {
+		score: ce(_, g, T),
+		query: _.join(", "),
+		candidate: g.join(", ")
 	});
-	let k = e.published?.year, A = t.published?.year;
-	k !== void 0 && A !== void 0 && i("published", {
-		score: w(k, A),
-		query: String(k),
-		candidate: String(A)
-	});
-	let j = e.languages ?? [], M = t.languages ?? [];
-	if (j.length > 0 && M.length > 0) {
-		let e = new Set(M.map(C));
+	let v = (e, t, n) => {
+		t === void 0 || n === void 0 || i(e, {
+			score: oe(t, n),
+			query: String(t),
+			candidate: String(n)
+		});
+	};
+	v("publication.original.date", e.publication?.original?.date?.year, t.publication?.original?.date?.year), v("publication.edition.date", e.publication?.edition?.date?.year, t.publication?.edition?.date?.year);
+	let y = e.languages ?? [], b = t.languages ?? [];
+	if (y.length > 0 && b.length > 0) {
+		let e = new Set(b.map(k));
 		i("languages", {
-			score: +!!j.some((t) => e.has(C(t))),
-			query: j.join(", "),
-			candidate: M.join(", ")
+			score: +!!y.some((t) => e.has(k(t))),
+			query: y.join(", "),
+			candidate: b.join(", ")
 		});
 	}
 	e.numberOfPages !== void 0 && t.numberOfPages !== void 0 && i("numberOfPages", {
-		score: T(e.numberOfPages, t.numberOfPages),
+		score: se(e.numberOfPages, t.numberOfPages),
 		query: String(e.numberOfPages),
 		candidate: String(t.numberOfPages)
 	});
-	let N = r.filter((e) => v.has(e.field));
-	if (N.some((e) => e.score === 0)) return {
+	let x = n.filter((e) => te.has(e.field));
+	if (x.some((e) => e.score === 0)) return {
 		score: 0,
-		signals: r
+		signals: n
 	};
-	if (N.some((e) => e.score === 1)) return {
+	if (ee || x.some((e) => e.score === 1)) return {
 		score: 1,
-		signals: r
+		signals: n
 	};
-	let P = r.reduce((e, t) => e + t.weight, 0);
-	return P === 0 ? {
+	let S = n.reduce((e, t) => e + t.weight, 0);
+	return S === 0 ? {
 		score: 0,
-		signals: r
+		signals: n
 	} : {
-		score: r.reduce((e, t) => e + t.score * t.weight, 0) / P,
-		signals: r
+		score: n.reduce((e, t) => e + t.score * t.weight, 0) / S,
+		signals: n
 	};
-}, O = (e) => {
+}, j = (e) => {
 	let t = Object.entries(e).filter(([, e]) => e !== void 0);
 	return Object.fromEntries(t);
-}, k = (e) => {
+}, le = (e) => {
 	let t = /* @__PURE__ */ new Map();
 	for (let n of e) {
 		let e = `${n.scheme?.trim().toLowerCase() ?? ""}:${n.value.trim().toLowerCase()}`, r = t.get(e);
@@ -156,27 +239,35 @@ var n = (e) => {
 		});
 	}
 	return [...t.values()];
-}, A = (...e) => {
+}, M = (...e) => {
 	let t = e.filter((e) => e !== void 0), n = (e) => {
 		for (let n of t) {
 			let t = n[e];
 			if (t !== void 0) return t;
 		}
-	}, r = k(t.flatMap((e) => e.identifiers ?? [])), i = O({
+	}, r = le(t.flatMap((e) => e.identifiers ?? [])), i = j({
 		series: t.find((e) => e.belongsTo?.series !== void 0)?.belongsTo?.series,
 		collection: t.find((e) => e.belongsTo?.collection !== void 0)?.belongsTo?.collection
+	}), a = (e) => {
+		let n = j({
+			date: t.find((t) => t.publication?.[e]?.date !== void 0)?.publication?.[e]?.date,
+			publisher: t.find((t) => t.publication?.[e]?.publisher !== void 0)?.publication?.[e]?.publisher,
+			imprint: t.find((t) => t.publication?.[e]?.imprint !== void 0)?.publication?.[e]?.imprint
+		});
+		return n.date !== void 0 || n.publisher !== void 0 || n.imprint !== void 0 ? n : void 0;
+	}, o = j({
+		original: a("original"),
+		edition: a("edition")
 	});
-	return O({
+	return j({
 		title: n("title"),
 		cover: n("cover"),
 		description: n("description"),
-		publisher: n("publisher"),
-		imprint: n("imprint"),
+		publication: o.original !== void 0 || o.edition !== void 0 ? o : void 0,
 		rights: n("rights"),
 		languages: n("languages"),
 		subjects: n("subjects"),
 		contributors: n("contributors"),
-		published: n("published"),
 		readingDirection: n("readingDirection"),
 		renditionLayout: n("renditionLayout"),
 		renditionFlow: n("renditionFlow"),
@@ -191,16 +282,16 @@ var n = (e) => {
 		apple: n("apple"),
 		kobo: n("kobo")
 	});
-}, j = class extends Error {
+}, N = class extends Error {
 	status;
 	constructor(e, t) {
 		super(t ?? `Provider responded with status ${e}`), this.name = "MetadataProviderResponseError", this.status = e;
 	}
-}, M = (e) => {
+}, ue = (e) => {
 	if (typeof e != "object" || !e || !("status" in e)) return;
 	let { status: t } = e;
 	if (!(typeof t != "number" || !Number.isInteger(t))) return t >= 100 && t <= 599 ? t : void 0;
-}, N = t.namespace("@prose-reader/metadata-fetcher", void 0, { color: "#b06642" }), P = 5, F = .5, I = 1, L = ({ candidate: e, providerId: t, score: n, signals: r, accepted: i, includeRaw: a }) => O({
+}, de = t.namespace("@prose-reader/metadata-fetcher", void 0, { color: "#b06642" }), fe = 5, pe = .5, me = 2, he = ({ candidate: e, providerId: t, score: n, signals: r, accepted: i, includeRaw: a }) => j({
 	providerId: t,
 	score: n,
 	signals: r,
@@ -209,8 +300,8 @@ var n = (e) => {
 	id: e.id,
 	url: e.url,
 	raw: a ? e.raw : void 0
-}), R = async (e, t) => {
-	let n = "metadata" in e ? e.metadata : e, r = t.limit ?? P, i = t.minScore ?? F, a = t.includeRaw === !0, o = await Promise.all(t.providers.map(async (e) => {
+}), ge = async (e, t) => {
+	let n = "metadata" in e ? e.metadata : e, r = t.limit ?? fe, i = t.minScore ?? pe, a = t.includeRaw === !0, o = await Promise.all(t.providers.map(async (e) => {
 		try {
 			return {
 				provider: e,
@@ -222,12 +313,12 @@ var n = (e) => {
 			};
 		} catch (n) {
 			if (t.signal?.aborted === !0) throw n;
-			return N.error(`fetchMetadata: provider "${e.id}" failed`, n), {
+			return de.error(`fetchMetadata: provider "${e.id}" failed`, n), {
 				provider: e,
 				candidates: void 0,
-				failure: O({
+				failure: j({
 					id: e.id,
-					status: M(n)
+					status: ue(n)
 				})
 			};
 		}
@@ -235,8 +326,8 @@ var n = (e) => {
 	for (let { provider: e, candidates: t } of o) {
 		if (t === void 0) continue;
 		let o = t.map((t) => {
-			let { score: r, signals: o } = D(n, t.metadata);
-			return L({
+			let { score: r, signals: o } = A(n, t.metadata);
+			return he({
 				candidate: t,
 				providerId: e.id,
 				score: r,
@@ -245,7 +336,7 @@ var n = (e) => {
 				includeRaw: a
 			});
 		}).sort((e, t) => t.score - e.score).slice(0, r);
-		c[e.id] !== void 0 && N.warn(`fetchMetadata: several providers share the id "${e.id}"`), c[e.id] = {
+		c[e.id] !== void 0 && de.warn(`fetchMetadata: several providers share the id "${e.id}"`), c[e.id] = {
 			provider: {
 				id: e.id,
 				name: e.name
@@ -255,40 +346,72 @@ var n = (e) => {
 	}
 	let u = [...l].sort((e, t) => t.score - e.score);
 	return {
-		version: I,
-		metadata: A(...u.filter((e) => e.accepted).map((e) => e.metadata)),
+		version: me,
+		metadata: M(...u.filter((e) => e.accepted).map((e) => e.metadata)),
 		matches: u,
 		sources: c,
 		failedProviders: s
 	};
-}, z = (e) => typeof e == "object" && !!e && !Array.isArray(e), B = (e, t) => {
+}, P = "ProjectGutenberg", _e = /* @__PURE__ */ new Set(["gutenberg.org", "www.gutenberg.org"]), ve = [
+	/^\/ebooks\/(\d+)(?:[./]|$)/i,
+	/^\/files\/(\d+)(?:\/|$)/i,
+	/^\/cache\/epub\/(\d+)(?:\/|$)/i,
+	/^\/(\d+)(?:\/|$)/
+], F = (e) => {
+	let t = e.trim();
+	if (!/^\d+$/.test(t)) return;
+	let n = t.replace(/^0+(?=\d)/, "");
+	return n === "0" ? void 0 : n;
+}, ye = (e) => {
+	try {
+		let t = new URL(e.trim());
+		if (t.protocol !== "http:" && t.protocol !== "https:" || !_e.has(t.hostname.toLowerCase())) return;
+		for (let e of ve) {
+			let n = e.exec(t.pathname)?.[1];
+			if (n !== void 0) return F(n);
+		}
+	} catch {
+		return;
+	}
+}, I = (e) => {
+	for (let t of e.identifiers ?? []) {
+		let e = t.scheme?.trim().toLowerCase(), n = e === "projectgutenberg" ? F(t.value) : e === void 0 || e === "url" || e === "uri" ? ye(t.value) : void 0;
+		if (n !== void 0) return {
+			id: n,
+			identifier: {
+				value: t.value,
+				...t.scheme === void 0 ? {} : { scheme: t.scheme }
+			}
+		};
+	}
+}, L = (e) => typeof e == "object" && !!e && !Array.isArray(e), R = (e, t) => {
 	let n = e[t];
 	if (typeof n != "string") return;
 	let r = n.trim();
 	return r.length > 0 ? r : void 0;
-}, V = (e, t) => {
+}, z = (e, t) => {
 	let n = e[t];
 	return typeof n == "number" && Number.isFinite(n) ? n : void 0;
-}, H = (e, t) => {
+}, B = (e, t) => {
 	let n = e[t];
 	if (!Array.isArray(n)) return;
 	let r = n.filter((e) => typeof e == "string").map((e) => e.trim()).filter((e) => e.length > 0);
 	return r.length > 0 ? r : void 0;
-}, U = (e, t) => {
+}, be = (e, t) => {
 	let n = e[t];
-	return Array.isArray(n) ? n.filter(z) : [];
-}, W = (e) => ({
-	key: B(e, "key"),
-	title: B(e, "title"),
-	subtitle: B(e, "subtitle"),
-	author_name: H(e, "author_name"),
-	first_publish_year: V(e, "first_publish_year"),
-	publisher: H(e, "publisher"),
-	language: H(e, "language"),
-	subject: H(e, "subject"),
-	number_of_pages_median: V(e, "number_of_pages_median"),
-	cover_i: V(e, "cover_i")
-}), G = (e) => z(e) ? U(e, "docs").map(W) : [], K = {
+	return Array.isArray(n) ? n.filter(L) : [];
+}, xe = (e) => ({
+	key: R(e, "key"),
+	title: R(e, "title"),
+	subtitle: R(e, "subtitle"),
+	author_name: B(e, "author_name"),
+	first_publish_year: z(e, "first_publish_year"),
+	language: B(e, "language"),
+	subject: B(e, "subject"),
+	number_of_pages_median: z(e, "number_of_pages_median"),
+	cover_i: z(e, "cover_i"),
+	id_project_gutenberg: B(e, "id_project_gutenberg")
+}), V = (e) => L(e) ? be(e, "docs").map(xe) : [], Se = {
 	afr: "af",
 	alb: "sq",
 	amh: "am",
@@ -393,76 +516,83 @@ var n = (e) => {
 	wel: "cy",
 	yid: "yi",
 	zho: "zh"
-}, q = /* @__PURE__ */ new Set([
+}, Ce = /* @__PURE__ */ new Set([
 	"mul",
 	"und",
 	"zxx",
 	"sgn"
-]), J = (e) => {
+]), H = (e) => {
 	let t = e.trim().toLowerCase();
-	if (t.length !== 0 && !q.has(t)) return K[t] ?? t;
-}, Y = 25, ee = {
+	if (t.length !== 0 && !Ce.has(t)) return Se[t] ?? t;
+}, we = 25, Te = {
 	key: "identifiers",
 	title: "title",
 	subtitle: "title",
 	author_name: "contributors",
-	first_publish_year: "published",
-	publisher: "publisher",
+	first_publish_year: "publication.original.date",
 	language: "languages",
 	subject: "subjects",
 	number_of_pages_median: "numberOfPages",
-	cover_i: "cover"
-}, X = "OpenLibrary", Z = (e) => e.length > 0 ? e : void 0, Q = (e, t) => {
-	let n = e.title !== void 0 && e.subtitle !== void 0 ? `${e.title}: ${e.subtitle}` : e.title, r = Z([...new Set((e.language ?? []).map(J).filter((e) => e !== void 0))]), i = [...t.isbn === void 0 ? [] : [{
-		value: t.isbn,
-		scheme: "ISBN"
-	}], ...e.key === void 0 ? [] : [{
-		value: e.key,
-		scheme: X
-	}]];
-	return O({
+	cover_i: "cover",
+	id_project_gutenberg: "identifiers"
+}, U = "OpenLibrary", Ee = "ProjectGutenberg", W = (e) => e.length > 0 ? e : void 0, G = (e, t) => {
+	let n = e.title !== void 0 && e.subtitle !== void 0 ? `${e.title}: ${e.subtitle}` : e.title, r = W([...new Set((e.language ?? []).map(H).filter((e) => e !== void 0))]), i = [
+		...t.isbn === void 0 ? [] : [{
+			value: t.isbn,
+			scheme: "ISBN"
+		}],
+		...t.matchedProjectGutenbergIdentifier === void 0 ? [] : [t.matchedProjectGutenbergIdentifier],
+		...(e.id_project_gutenberg ?? []).map((e) => ({
+			value: e,
+			scheme: Ee
+		})),
+		...e.key === void 0 ? [] : [{
+			value: e.key,
+			scheme: U
+		}]
+	];
+	return j({
 		title: n,
 		cover: e.cover_i === void 0 ? void 0 : {
 			uri: `${t.coversBaseUrl}/b/id/${e.cover_i}-L.jpg`,
 			mediaType: "image/jpeg",
 			confidence: "derived"
 		},
-		publisher: e.publisher?.[0],
+		publication: e.first_publish_year === void 0 ? void 0 : { original: { date: { year: e.first_publish_year } } },
 		languages: r,
-		subjects: Z((e.subject ?? []).slice(0, 25)),
-		contributors: Z((e.author_name ?? []).map((e) => ({
+		subjects: W((e.subject ?? []).slice(0, 25)),
+		contributors: W((e.author_name ?? []).map((e) => ({
 			name: e,
 			roles: ["author"]
 		}))),
-		published: e.first_publish_year === void 0 ? void 0 : { year: e.first_publish_year },
 		numberOfPages: e.number_of_pages_median,
 		isbn: t.isbn,
-		identifiers: Z(i)
+		identifiers: W(i)
 	});
-}, $ = "openLibrary", te = "https://openlibrary.org", ne = "https://covers.openlibrary.org", re = [
+}, K = "openLibrary", De = "https://openlibrary.org", Oe = "https://covers.openlibrary.org", ke = [
 	"key",
 	"title",
 	"subtitle",
 	"author_name",
 	"first_publish_year",
-	"publisher",
 	"language",
 	"subject",
 	"number_of_pages_median",
-	"cover_i"
-].join(","), ie = (e) => {
+	"cover_i",
+	"id_project_gutenberg"
+].join(","), Ae = (e) => {
 	let t = e.title?.trim();
 	if (t === void 0 || t.length === 0) return;
-	let r = n(e)[0];
+	let n = r(e)[0];
 	return {
 		title: t,
-		...r === void 0 ? {} : { author: r }
+		...n === void 0 ? {} : { author: n }
 	};
-}, ae = (e = {}) => {
-	let t = e.baseUrl ?? te, n = e.coversBaseUrl ?? ne, r = async (n, r) => {
+}, je = (e = {}) => {
+	let t = e.baseUrl ?? De, n = e.coversBaseUrl ?? Oe, r = async (n, r) => {
 		let i = new URL("/search.json", t);
 		for (let [e, t] of Object.entries(n)) i.searchParams.set(e, t);
-		i.searchParams.set("fields", re), i.searchParams.set("limit", String(r.limit));
+		i.searchParams.set("fields", ke), i.searchParams.set("limit", String(r.limit));
 		let a = await (e.fetch ?? globalThis.fetch)(i.toString(), {
 			signal: r.signal,
 			headers: {
@@ -470,37 +600,229 @@ var n = (e) => {
 				...e.userAgent === void 0 ? {} : { "User-Agent": e.userAgent }
 			}
 		});
-		if (!a.ok) throw new j(a.status, `Open Library search failed with status ${a.status}`);
-		return G(await a.json());
-	}, i = (e, r) => e.map((e) => ({
-		metadata: Q(e, {
+		if (!a.ok) throw new N(a.status, `Open Library search failed with status ${a.status}`);
+		return V(await a.json());
+	}, i = (e, r = {}) => e.map((e) => ({
+		metadata: G(e, {
 			coversBaseUrl: n,
-			isbn: r
+			isbn: r.isbn,
+			matchedProjectGutenbergIdentifier: r.confirmedProjectGutenberg?.identifier
 		}),
 		id: e.key,
 		url: e.key === void 0 ? void 0 : `${t}${e.key}`,
 		raw: e
 	}));
 	return {
-		id: $,
+		id: K,
 		name: "Open Library",
 		search: async (e, t) => {
-			let n = e.isbn?.trim() || void 0;
+			let n = e.isbn?.trim() || void 0, a = I(e);
 			if (n !== void 0) {
 				let e = await r({ isbn: n }, t);
-				if (e.length > 0) return i(e, n);
+				if (e.length > 0) return i(e, { isbn: n });
+			}
+			if (a !== void 0) {
+				let e = await r({ q: `id_project_gutenberg:${a.id}` }, t);
+				if (e.length > 0) return i(e, { confirmedProjectGutenberg: a });
 			}
-			let a = ie(e);
-			return a === void 0 ? [] : i(await r(a, t), void 0);
+			let o = Ae(e);
+			return o === void 0 ? [] : i(await r(o, t));
+		}
+	};
+}, q = (e) => e.includes(":") ? e.slice(e.lastIndexOf(":") + 1) : e, Me = (e) => e.type === "element", J = (e, t) => e.children.filter((e) => Me(e) && q(e.name).toLowerCase() === t.toLowerCase()), Y = (e, t) => J(e, t)[0], X = (e) => {
+	let t = e?.val.trim();
+	return t !== void 0 && t.length > 0 ? t : void 0;
+}, Z = (e, t) => {
+	for (let [n, r] of Object.entries(e.attr)) {
+		if (q(n).toLowerCase() !== t.toLowerCase()) continue;
+		let e = r.trim();
+		if (e.length > 0) return e;
+	}
+}, Ne = (e, t) => J(e, t).flatMap((e) => {
+	let t = X(Y(Y(e, "Description") ?? e, "value"));
+	return t === void 0 ? [] : [t];
+}), Pe = (e) => {
+	let t = Z(e, "about"), n = t === void 0 ? void 0 : /(?:^|\/)ebooks\/(\d+)$/.exec(t)?.[1];
+	if (n === void 0) return;
+	let r = n.replace(/^0+(?=\d)/, "");
+	return r === "0" ? void 0 : r;
+}, Fe = (e) => {
+	let t = [];
+	for (let n of e.children) {
+		if (!Me(n)) continue;
+		let e = n.name.toLowerCase() === "dcterms:creator", r = n.name.toLowerCase().startsWith("marcrel:");
+		if (!e && !r) continue;
+		let i = X(Y(Y(n, "agent") ?? n, "name"));
+		i !== void 0 && t.push({
+			name: i,
+			role: e ? "aut" : q(n.name).toLowerCase()
+		});
+	}
+	return t;
+}, Ie = (e) => J(e, "subject").flatMap((e) => {
+	let t = Y(e, "Description"), n = t === void 0 ? void 0 : Z(Y(t, "memberOf") ?? t, "resource"), r = X(t === void 0 ? void 0 : Y(t, "value"));
+	return r !== void 0 && n?.endsWith("/LCSH") === !0 ? [r] : [];
+}), Le = (e) => {
+	let t = J(e, "hasFormat").flatMap((e) => {
+		let t = Y(e, "file");
+		if (t === void 0) return [];
+		let n = Z(t, "about"), r = J(t, "format").map((e) => X(Y(Y(e, "Description") ?? e, "value"))).find((e) => e?.startsWith("image/") === !0);
+		return n === void 0 || r === void 0 || !/\.cover\.(?:medium|small)\.[a-z0-9]+(?:$|[?#])/i.test(n) ? [] : [{
+			uri: n,
+			mediaType: r
+		}];
+	});
+	return t.find((e) => /\.cover\.medium\./i.test(e.uri)) ?? t[0];
+}, Q = (e) => e.length > 0 ? e : void 0, Re = (e) => {
+	let t;
+	try {
+		t = new n(e);
+	} catch (e) {
+		let t = e instanceof Error ? e.message : String(e);
+		throw Error(`Project Gutenberg RDF is malformed: ${t}`, { cause: e });
+	}
+	if (q(t.name).toLowerCase() !== "rdf") return;
+	let r = Y(t, "ebook");
+	if (r === void 0) return;
+	let i = Pe(r);
+	if (i === void 0) return;
+	let a = [...new Set(Ne(r, "language"))], o = [...new Set(Ie(r))], s = [...new Set(Ne(r, "bookshelf"))], c = Fe(r);
+	return {
+		id: i,
+		title: X(Y(r, "title")),
+		publisher: X(Y(r, "publisher")),
+		issued: X(Y(r, "issued")),
+		originalPublication: X(Y(r, "marc260")),
+		rights: X(Y(r, "rights")),
+		description: X(Y(r, "description")),
+		summary: X(Y(r, "marc520")),
+		languages: Q(a),
+		subjects: Q(o),
+		bookshelves: Q(s),
+		contributors: Q(c),
+		cover: Le(r)
+	};
+}, ze = 25, Be = {
+	id: "identifiers",
+	title: "title",
+	publisher: "publication.edition.publisher",
+	issued: "publication.edition.date",
+	originalPublication: "publication.original",
+	rights: "rights",
+	description: "description",
+	summary: "description",
+	languages: "languages",
+	subjects: "subjects",
+	bookshelves: "subjects",
+	contributors: "contributors",
+	cover: "cover"
+}, Ve = {
+	art: "artist",
+	aut: "author",
+	clr: "colorist",
+	ctb: "contributor",
+	edt: "editor",
+	ill: "illustrator",
+	nrt: "narrator",
+	trl: "translator"
+}, He = (e) => {
+	let t = /* @__PURE__ */ new Map();
+	for (let n of e.contributors ?? []) {
+		let e = n.name.trim().toLowerCase(), r = Ve[n.role.trim().toLowerCase()] ?? n.role, i = t.get(e);
+		i === void 0 ? t.set(e, {
+			name: n.name,
+			roles: [r]
+		}) : i.roles.includes(r) || i.roles.push(r);
+	}
+	return t.size > 0 ? [...t.values()] : void 0;
+}, Ue = (e) => {
+	if (e === void 0) return;
+	let t = /^(\d{4})(?:-(\d{2})(?:-(\d{2}))?)?$/.exec(e.trim());
+	if (t === null) return;
+	let n = Number(t[1]), r = t[2] === void 0 ? void 0 : Number(t[2]), i = t[3] === void 0 ? void 0 : Number(t[3]);
+	if (!(!Number.isInteger(n) || n <= 0 || r !== void 0 && (r < 1 || r > 12) || i !== void 0 && (i < 1 || i > 31))) return {
+		year: n,
+		...r === void 0 ? {} : { month: r },
+		...i === void 0 ? {} : { day: i }
+	};
+}, We = (e, t) => e.split("$").slice(1).flatMap((e) => e[0]?.toLowerCase() === t.toLowerCase() ? [e.slice(1).trim()] : []).filter((e) => e.length > 0), Ge = (e) => {
+	if (e === void 0) return;
+	let t = We(e, "b"), n = (t.length === 1 ? t[0] : void 0)?.replace(/[\s,;:/]+$/u, "").trim(), r = We(e, "c"), i = [...new Set(r.flatMap((e) => [...e.matchAll(/\d{4}/g)].flatMap((e) => e[0] === void 0 ? [] : [Number(e[0])])))], a = j({
+		date: i.length === 1 ? { year: i[0] } : void 0,
+		publisher: n !== void 0 && n.length > 0 && !/^\[?s\.n\.?\]?$/i.test(n) ? n : void 0
+	});
+	return a.date !== void 0 || a.publisher !== void 0 ? a : void 0;
+}, Ke = (e, t) => {
+	try {
+		let n = new URL(e, t);
+		return n.protocol === "http:" || n.protocol === "https:" ? n.toString() : void 0;
+	} catch {
+		return;
+	}
+}, qe = (e) => e.length > 0 ? e : void 0, Je = (e, t) => {
+	let n = [t.matchedIdentifier, {
+		value: e.id,
+		scheme: P
+	}].filter((e, t, n) => n.findIndex((t) => t.value.trim().toLowerCase() === e.value.trim().toLowerCase() && t.scheme?.trim().toLowerCase() === e.scheme?.trim().toLowerCase()) === t), r = [.../* @__PURE__ */ new Set([...e.subjects ?? [], ...e.bookshelves ?? []])].slice(0, 25), i = e.cover === void 0 ? void 0 : Ke(e.cover.uri, t.baseUrl), a = i === void 0 ? void 0 : {
+		uri: i,
+		mediaType: e.cover?.mediaType,
+		confidence: "derived"
+	}, o = j({
+		date: Ue(e.issued),
+		publisher: e.publisher
+	}), s = j({
+		original: Ge(e.originalPublication),
+		edition: o.date !== void 0 || o.publisher !== void 0 ? o : void 0
+	});
+	return j({
+		title: e.title,
+		publication: s.original !== void 0 || s.edition !== void 0 ? s : void 0,
+		description: e.summary ?? e.description,
+		rights: e.rights,
+		languages: qe([...new Set(e.languages ?? [])]),
+		subjects: qe(r),
+		contributors: He(e),
+		identifiers: n,
+		cover: a
+	});
+}, $ = "projectGutenberg", Ye = "https://www.gutenberg.org", Xe = (e = {}) => {
+	let t = e.baseUrl ?? Ye;
+	return {
+		id: $,
+		name: "Project Gutenberg",
+		search: async (n, r) => {
+			let i = I(n);
+			if (i === void 0) return [];
+			let a = new URL(`/cache/epub/${i.id}/pg${i.id}.rdf`, t), o = await (e.fetch ?? globalThis.fetch)(a.toString(), {
+				signal: r.signal,
+				headers: {
+					Accept: "application/rdf+xml, application/xml;q=0.9",
+					...e.userAgent === void 0 ? {} : { "User-Agent": e.userAgent }
+				}
+			});
+			if (o.status === 404) return [];
+			if (!o.ok) throw new N(o.status, `Project Gutenberg metadata lookup failed with status ${o.status}`);
+			let s = Re(await o.text());
+			if (s === void 0 || s.id !== i.id) throw Error(`Project Gutenberg RDF did not contain the requested eBook ${i.id}`);
+			let c = new URL(`/ebooks/${i.id}`, t).toString();
+			return [{
+				id: s.id,
+				url: c,
+				raw: s,
+				metadata: Je(s, {
+					baseUrl: t,
+					matchedIdentifier: i.identifier
+				})
+			}];
 		}
 	};
-}, oe = (e) => [
+}, Ze = (e) => [
 	e.title,
 	e.isbn,
 	e.gtin,
 	e.belongsTo?.series?.[0]?.name
-].some((e) => e !== void 0 && e.trim().length > 0) || (e.identifiers ?? []).length > 0 || n(e).length > 0;
+].some((e) => e !== void 0 && e.trim().length > 0) || (e.identifiers ?? []).length > 0 || r(e).length > 0;
 //#endregion
-export { _ as METADATA_MATCH_WEIGHTS, j as MetadataProviderResponseError, X as OPEN_LIBRARY_IDENTIFIER_SCHEME, Y as OPEN_LIBRARY_MAX_SUBJECTS, $ as OPEN_LIBRARY_PROVIDER_ID, ae as createOpenLibraryProvider, R as fetchMetadata, oe as hasSearchTerms, J as marcLanguageToBcp47, A as mergeResolvedMetadata, n as metadataAuthors, c as normalizeForComparison, ee as openLibraryMetadataHomes, G as parseOpenLibrarySearchResponse, g as personNameSimilarity, Q as resolveOpenLibraryDoc, M as responseErrorStatus, D as scoreMetadataCandidate, u as textSimilarity, m as titleSimilarity, a as toIsbn13 };
+export { E as METADATA_MATCH_WEIGHTS, N as MetadataProviderResponseError, U as OPEN_LIBRARY_IDENTIFIER_SCHEME, we as OPEN_LIBRARY_MAX_SUBJECTS, K as OPEN_LIBRARY_PROVIDER_ID, P as PROJECT_GUTENBERG_IDENTIFIER_SCHEME, ze as PROJECT_GUTENBERG_MAX_SUBJECTS, $ as PROJECT_GUTENBERG_PROVIDER_ID, je as createOpenLibraryProvider, Xe as createProjectGutenbergProvider, ge as fetchMetadata, Ze as hasSearchTerms, H as marcLanguageToBcp47, M as mergeResolvedMetadata, r as metadataAuthors, l as normalizeForComparison, Te as openLibraryMetadataHomes, V as parseOpenLibrarySearchResponse, Re as parseProjectGutenbergRdf, T as personNameSimilarity, I as projectGutenbergLookupFromMetadata, Be as projectGutenbergMetadataHomes, G as resolveOpenLibraryDoc, Je as resolveProjectGutenbergRecord, ue as responseErrorStatus, A as scoreMetadataCandidate, d as textSimilarity, C as titleSimilarity, o as toIsbn13 };
 
 //# sourceMappingURL=index.js.map