@prose-reader/metadata-fetcher 1.339.0 → 1.341.0

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
@@ -39,11 +40,11 @@ var n = (e) => {
 		main: r,
 		hasSubtitle: i.length > 0
 	};
-}, p = .9, m = [
+}, m = .9, h = [
 	"volume",
 	"part",
 	"book"
-], h = /\b(volumes?|vols?|parts?|books?)\s+([\p{L}\p{N}]+)\b/gu, g = {
+], g = /\b(volumes?|vols?|parts?|books?)\s+([\p{L}\p{N}]+)\b/gu, _ = {
 	one: 1,
 	first: 1,
 	two: 2,
@@ -68,7 +69,7 @@ var n = (e) => {
 	eleventh: 11,
 	twelve: 12,
 	twelfth: 12
-}, _ = {
+}, v = {
 	i: 1,
 	v: 5,
 	x: 10,
@@ -76,78 +77,91 @@ var n = (e) => {
 	c: 100,
 	d: 500,
 	m: 1e3
-}, v = (e) => {
+}, y = (e) => {
 	if (!/^m{0,3}(cm|cd|d?c{0,3})(xc|xl|l?x{0,3})(ix|iv|v?i{0,3})$/i.test(e)) return;
 	let t = 0, n = 0;
 	for (let r of [...e.toLowerCase()].reverse()) {
-		let e = _[r];
+		let e = v[r];
 		if (e === void 0) return;
 		e < n ? t -= e : (t += e, n = e);
 	}
 	return t > 0 ? t : void 0;
-}, y = (e) => /^\d+$/.test(e) ? Number.parseInt(e, 10) : g[e] ?? v(e), b = (e) => {
+}, b = (e) => /^\d+$/.test(e) ? Number.parseInt(e, 10) : _[e] ?? y(e), x = (e) => {
 	if (e.startsWith("vol")) return "volume";
 	if (e.startsWith("part")) return "part";
 	if (e.startsWith("book")) return "book";
-}, x = (e) => {
+}, S = (e) => {
 	let t = /* @__PURE__ */ new Map();
-	for (let n of c(e).matchAll(h)) {
-		let e = n[1] === void 0 ? void 0 : b(n[1]), r = n[2] === void 0 ? void 0 : y(n[2]);
+	for (let n of l(e).matchAll(g)) {
+		let e = n[1] === void 0 ? void 0 : x(n[1]), r = n[2] === void 0 ? void 0 : b(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 m) {
+}, C = (e, t) => {
+	let n = S(e), r = S(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 = u(e, t), r = f(e), i = f(t);
-	return r.hasSubtitle === i.hasSubtitle ? n : Math.max(n, u(r.main, i.main) * p);
-}, w = (e) => c(e).split(" ").filter((e) => e.length > 0).sort().join(" "), T = (e, t) => u(w(e), w(t)), E = {
+}, w = (e, t) => {
+	if (C(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);
+}, T = (e) => l(e).split(" ").filter((e) => e.length > 0).sort().join(" "), E = (e, t) => d(T(e), T(t)), D = {
 	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
-}, D = /* @__PURE__ */ new Set(["isbn", "gtin"]), O = (e) => {
+}, ee = /* @__PURE__ */ new Set(["isbn", "gtin"]), O = (e) => {
 	let t = e?.trim();
 	return t !== void 0 && t.length > 0 ? t : void 0;
-}, k = (e) => e.value.trim().toLowerCase(), A = (e) => e.scheme === void 0 ? e.value : `${e.scheme}:${e.value}`, j = (e, t) => k(e) === k(t) ? e.scheme === void 0 || t.scheme === void 0 || e.scheme.trim().toLowerCase() === t.scheme.trim().toLowerCase() : !1, M = (e) => e.trim().toLowerCase().split("-")[0] ?? "", N = (e, t) => {
+}, k = (e, t) => {
+	let n = O(e.value)?.toLowerCase();
+	return n !== void 0 && n === O(t.value)?.toLowerCase();
+}, A = (e) => O(e.scheme)?.toLowerCase(), te = (e) => e.scheme === void 0 ? e.value : `${e.scheme}:${e.value}`, ne = (e, t) => {
+	if (!k(e, t)) return !1;
+	let n = A(e), r = A(t);
+	return n === void 0 || r === void 0 || n === r;
+}, re = (e, t) => {
+	if (!k(e, t)) return !1;
+	let n = A(e), r = A(t);
+	return n !== void 0 && n === r;
+}, ie = (e) => e.trim().toLowerCase().split("-")[0] ?? "", ae = (e, t) => {
 	let n = Math.abs(e - t);
 	return n === 0 ? 1 : n === 1 ? .6 : n <= 3 ? .3 : 0;
-}, ee = (e, t) => {
+}, oe = (e, t) => {
 	let n = Math.max(e, t);
 	return n <= 0 ? 0 : Math.max(0, 1 - Math.abs(e - t) / n * 2);
-}, P = (e, t, n) => e.reduce((e, r) => t.reduce((e, t) => Math.max(e, n(r, t)), e), 0), F = (e, t) => {
-	let r = [], i = (e, t) => {
-		t !== void 0 && r.push({
+}, se = (e, t, n) => e.reduce((e, r) => t.reduce((e, t) => Math.max(e, n(r, t)), e), 0), ce = (e, t) => {
+	let n = [], i = (e, t) => {
+		t !== void 0 && n.push({
 			field: e,
-			weight: E[e],
+			weight: D[e],
 			score: Math.min(1, Math.max(0, t.score)),
 			query: t.query,
 			candidate: t.candidate
 		});
-	}, o = (e, t, n, r = u) => {
+	}, a = (e, t, n, r = d) => {
 		let a = O(t), o = O(n);
 		a === void 0 || o === void 0 || i(e, {
 			score: r(a, o),
 			query: a,
 			candidate: o
 		});
-	}, s = O(e.isbn), c = O(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 = O(e.isbn), c = O(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
 	});
@@ -157,117 +171,70 @@ var n = (e) => {
 		query: f,
 		candidate: p
 	});
-	let m = e.identifiers ?? [], h = t.identifiers ?? [];
+	let m = e.identifiers ?? [], h = t.identifiers ?? [], g = m.some((e) => h.some((t) => re(e, t)));
 	m.length > 0 && h.length > 0 && i("identifiers", {
-		score: +!!m.some((e) => h.some((t) => j(e, t))),
-		query: m.map(A).join(", "),
-		candidate: h.map(A).join(", ")
-	}), o("title", e.title, t.title, C), o("publisher", e.publisher, t.publisher), o("series", e.belongsTo?.series?.[0]?.name, t.belongsTo?.series?.[0]?.name);
-	let g = n(t), _ = n(e);
-	_.length > 0 && g.length > 0 && i("contributors", {
-		score: P(_, g, T),
-		query: _.join(", "),
-		candidate: g.join(", ")
-	});
-	let v = e.published?.year, y = t.published?.year;
-	v !== void 0 && y !== void 0 && i("published", {
-		score: N(v, y),
-		query: String(v),
-		candidate: String(y)
+		score: +!!m.some((e) => h.some((t) => ne(e, t))),
+		query: m.map(te).join(", "),
+		candidate: h.map(te).join(", ")
+	}), a("title", e.title, t.title, w), 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 _ = r(t), v = r(e);
+	v.length > 0 && _.length > 0 && i("contributors", {
+		score: se(v, _, E),
+		query: v.join(", "),
+		candidate: _.join(", ")
 	});
+	let y = (e, t, n) => {
+		t === void 0 || n === void 0 || i(e, {
+			score: ae(t, n),
+			query: String(t),
+			candidate: String(n)
+		});
+	};
+	y("publication.original.date", e.publication?.original?.date?.year, t.publication?.original?.date?.year), y("publication.edition.date", e.publication?.edition?.date?.year, t.publication?.edition?.date?.year);
 	let b = e.languages ?? [], x = t.languages ?? [];
 	if (b.length > 0 && x.length > 0) {
-		let e = new Set(x.map(M));
+		let e = new Set(x.map(ie));
 		i("languages", {
-			score: +!!b.some((t) => e.has(M(t))),
+			score: +!!b.some((t) => e.has(ie(t))),
 			query: b.join(", "),
 			candidate: x.join(", ")
 		});
 	}
 	e.numberOfPages !== void 0 && t.numberOfPages !== void 0 && i("numberOfPages", {
-		score: ee(e.numberOfPages, t.numberOfPages),
+		score: oe(e.numberOfPages, t.numberOfPages),
 		query: String(e.numberOfPages),
 		candidate: String(t.numberOfPages)
 	});
-	let S = r.filter((e) => D.has(e.field));
+	let S = n.filter((e) => ee.has(e.field));
 	if (S.some((e) => e.score === 0)) return {
 		score: 0,
-		signals: r
+		signals: n
 	};
-	if (S.some((e) => e.score === 1)) return {
+	if (g || S.some((e) => e.score === 1)) return {
 		score: 1,
-		signals: r
+		signals: n
 	};
-	let w = r.reduce((e, t) => e + t.weight, 0);
-	return w === 0 ? {
+	let C = n.reduce((e, t) => e + t.weight, 0);
+	return C === 0 ? {
 		score: 0,
-		signals: r
+		signals: n
 	} : {
-		score: r.reduce((e, t) => e + t.score * t.weight, 0) / w,
-		signals: r
+		score: n.reduce((e, t) => e + t.score * t.weight, 0) / C,
+		signals: n
 	};
-}, I = (e) => {
-	let t = Object.entries(e).filter(([, e]) => e !== void 0);
-	return Object.fromEntries(t);
-}, te = (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);
-		if (r === void 0) {
-			t.set(e, n);
-			continue;
-		}
-		n.unique === !0 && r.unique !== !0 && t.set(e, {
-			...r,
-			unique: !0
-		});
-	}
-	return [...t.values()];
-}, L = (...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 = te(t.flatMap((e) => e.identifiers ?? [])), i = I({
-		series: t.find((e) => e.belongsTo?.series !== void 0)?.belongsTo?.series,
-		collection: t.find((e) => e.belongsTo?.collection !== void 0)?.belongsTo?.collection
-	});
-	return I({
-		title: n("title"),
-		cover: n("cover"),
-		description: n("description"),
-		publisher: n("publisher"),
-		imprint: n("imprint"),
-		rights: n("rights"),
-		languages: n("languages"),
-		subjects: n("subjects"),
-		contributors: n("contributors"),
-		published: n("published"),
-		readingDirection: n("readingDirection"),
-		renditionLayout: n("renditionLayout"),
-		renditionFlow: n("renditionFlow"),
-		renditionSpread: n("renditionSpread"),
-		numberOfPages: n("numberOfPages"),
-		gtin: n("gtin"),
-		isbn: n("isbn"),
-		identifiers: r.length > 0 ? r : void 0,
-		belongsTo: Object.keys(i).length > 0 ? i : void 0,
-		properties: n("properties"),
-		comic: n("comic"),
-		apple: n("apple"),
-		kobo: n("kobo")
-	});
-}, R = class extends Error {
+}, j = class extends Error {
 	status;
 	constructor(e, t) {
 		super(t ?? `Provider responded with status ${e}`), this.name = "MetadataProviderResponseError", this.status = e;
 	}
-}, z = (e) => {
+}, M = (e) => {
 	if (typeof e != "object" || !e || !("status" in e)) return;
 	let { status: t } = e;
 	if (!(typeof t != "number" || !Number.isInteger(t))) return t >= 100 && t <= 599 ? t : void 0;
-}, B = t.namespace("@prose-reader/metadata-fetcher", void 0, { color: "#b06642" }), ne = 5, re = .5, ie = 1, ae = ({ candidate: e, providerId: t, score: n, signals: r, accepted: i, includeRaw: a }) => I({
+}, N = t.namespace("@prose-reader/metadata-fetcher", void 0, { color: "#b06642" }), P = (e) => {
+	let t = Object.entries(e).filter(([, e]) => e !== void 0);
+	return Object.fromEntries(t);
+}, le = 5, ue = .5, de = 3, fe = ({ candidate: e, providerId: t, score: n, signals: r, accepted: i, includeRaw: a }) => P({
 	providerId: t,
 	score: n,
 	signals: r,
@@ -276,8 +243,8 @@ var n = (e) => {
 	id: e.id,
 	url: e.url,
 	raw: a ? e.raw : void 0
-}), V = async (e, t) => {
-	let n = "metadata" in e ? e.metadata : e, r = t.limit ?? ne, i = t.minScore ?? re, a = t.includeRaw === !0, o = await Promise.all(t.providers.map(async (e) => {
+}), pe = async (e, t) => {
+	let n = "metadata" in e ? e.metadata : e, r = t.limit ?? le, i = t.minScore ?? ue, a = t.includeRaw === !0, o = await Promise.all(t.providers.map(async (e) => {
 		try {
 			return {
 				provider: e,
@@ -289,12 +256,12 @@ var n = (e) => {
 			};
 		} catch (n) {
 			if (t.signal?.aborted === !0) throw n;
-			return B.error(`fetchMetadata: provider "${e.id}" failed`, n), {
+			return N.error(`fetchMetadata: provider "${e.id}" failed`, n), {
 				provider: e,
 				candidates: void 0,
-				failure: I({
+				failure: P({
 					id: e.id,
-					status: z(n)
+					status: M(n)
 				})
 			};
 		}
@@ -302,8 +269,8 @@ var n = (e) => {
 	for (let { provider: e, candidates: t } of o) {
 		if (t === void 0) continue;
 		let o = t.map((t) => {
-			let { score: r, signals: o } = F(n, t.metadata);
-			return ae({
+			let { score: r, signals: o } = ce(n, t.metadata);
+			return fe({
 				candidate: t,
 				providerId: e.id,
 				score: r,
@@ -312,7 +279,7 @@ var n = (e) => {
 				includeRaw: a
 			});
 		}).sort((e, t) => t.score - e.score).slice(0, r);
-		c[e.id] !== void 0 && B.warn(`fetchMetadata: several providers share the id "${e.id}"`), c[e.id] = {
+		c[e.id] !== void 0 && N.warn(`fetchMetadata: several providers share the id "${e.id}"`), c[e.id] = {
 			provider: {
 				id: e.id,
 				name: e.name
@@ -320,63 +287,36 @@ var n = (e) => {
 			matches: o
 		}, l.push(...o);
 	}
-	let u = [...l].sort((e, t) => t.score - e.score);
 	return {
-		version: ie,
-		metadata: L(...u.filter((e) => e.accepted).map((e) => e.metadata)),
-		matches: u,
+		version: de,
+		matches: [...l].sort((e, t) => t.score - e.score),
 		sources: c,
 		failedProviders: s
 	};
-}, H = (e) => typeof e == "object" && !!e && !Array.isArray(e), U = (e, t) => {
-	let n = e[t];
-	if (typeof n != "string") return;
-	let r = n.trim();
-	return r.length > 0 ? r : void 0;
-}, W = (e, t) => {
-	let n = e[t];
-	return typeof n == "number" && Number.isFinite(n) ? n : void 0;
-}, G = (e, t) => {
-	let n = e[t];
-	if (!Array.isArray(n)) return;
-	let r = n.filter((e) => typeof e == "string").map((e) => e.trim()).filter((e) => e.length > 0);
-	return r.length > 0 ? r : void 0;
-}, K = (e, t) => {
-	let n = e[t];
-	return Array.isArray(n) ? n.filter(H) : [];
-}, q = (e) => ({
-	key: U(e, "key"),
-	title: U(e, "title"),
-	subtitle: U(e, "subtitle"),
-	author_name: G(e, "author_name"),
-	first_publish_year: W(e, "first_publish_year"),
-	publisher: G(e, "publisher"),
-	language: G(e, "language"),
-	subject: G(e, "subject"),
-	number_of_pages_median: W(e, "number_of_pages_median"),
-	cover_i: W(e, "cover_i"),
-	id_project_gutenberg: G(e, "id_project_gutenberg")
-}), J = (e) => H(e) ? K(e, "docs").map(q) : [], oe = /* @__PURE__ */ new Set(["gutenberg.org", "www.gutenberg.org"]), se = [
+}, F = "ProjectGutenberg", me = /* @__PURE__ */ new Set(["gutenberg.org", "www.gutenberg.org"]), he = [
 	/^\/ebooks\/(\d+)(?:[./]|$)/i,
 	/^\/files\/(\d+)(?:\/|$)/i,
 	/^\/cache\/epub\/(\d+)(?:\/|$)/i,
 	/^\/(\d+)(?:\/|$)/
-], ce = (e) => {
+], I = (e) => {
+	let t = e.trim();
+	if (!/^\d+$/.test(t)) return;
+	let n = t.replace(/^0+(?=\d)/, "");
+	return n === "0" ? void 0 : n;
+}, ge = (e) => {
 	try {
 		let t = new URL(e.trim());
-		if (t.protocol !== "http:" && t.protocol !== "https:" || !oe.has(t.hostname.toLowerCase())) return;
-		for (let e of se) {
+		if (t.protocol !== "http:" && t.protocol !== "https:" || !me.has(t.hostname.toLowerCase())) return;
+		for (let e of he) {
 			let n = e.exec(t.pathname)?.[1];
-			if (n !== void 0) return n;
+			if (n !== void 0) return I(n);
 		}
 	} catch {
 		return;
 	}
-}, le = (e) => {
+}, L = (e) => {
 	for (let t of e.identifiers ?? []) {
-		let e = t.scheme?.trim().toLowerCase();
-		if (e !== void 0 && e !== "url" && e !== "uri") continue;
-		let n = ce(t.value);
+		let e = t.scheme?.trim().toLowerCase(), n = e === "projectgutenberg" ? I(t.value) : e === void 0 || e === "url" || e === "uri" ? ge(t.value) : void 0;
 		if (n !== void 0) return {
 			id: n,
 			identifier: {
@@ -385,7 +325,34 @@ var n = (e) => {
 			}
 		};
 	}
-}, ue = {
+}, R = (e) => typeof e == "object" && !!e && !Array.isArray(e), z = (e, t) => {
+	let n = e[t];
+	if (typeof n != "string") return;
+	let r = n.trim();
+	return r.length > 0 ? r : void 0;
+}, B = (e, t) => {
+	let n = e[t];
+	return typeof n == "number" && Number.isFinite(n) ? n : void 0;
+}, V = (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;
+}, _e = (e, t) => {
+	let n = e[t];
+	return Array.isArray(n) ? n.filter(R) : [];
+}, ve = (e) => ({
+	key: z(e, "key"),
+	title: z(e, "title"),
+	subtitle: z(e, "subtitle"),
+	author_name: V(e, "author_name"),
+	first_publish_year: B(e, "first_publish_year"),
+	language: V(e, "language"),
+	subject: V(e, "subject"),
+	number_of_pages_median: B(e, "number_of_pages_median"),
+	cover_i: B(e, "cover_i"),
+	id_project_gutenberg: V(e, "id_project_gutenberg")
+}), H = (e) => R(e) ? _e(e, "docs").map(ve) : [], ye = {
 	afr: "af",
 	alb: "sq",
 	amh: "am",
@@ -490,28 +457,27 @@ var n = (e) => {
 	wel: "cy",
 	yid: "yi",
 	zho: "zh"
-}, de = /* @__PURE__ */ new Set([
+}, be = /* @__PURE__ */ new Set([
 	"mul",
 	"und",
 	"zxx",
 	"sgn"
-]), Y = (e) => {
+]), U = (e) => {
 	let t = e.trim().toLowerCase();
-	if (t.length !== 0 && !de.has(t)) return ue[t] ?? t;
-}, fe = 25, pe = {
+	if (t.length !== 0 && !be.has(t)) return ye[t] ?? t;
+}, xe = 25, Se = {
 	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",
 	id_project_gutenberg: "identifiers"
-}, X = "OpenLibrary", me = "ProjectGutenberg", Z = (e) => e.length > 0 ? e : void 0, Q = (e, t) => {
-	let n = e.title !== void 0 && e.subtitle !== void 0 ? `${e.title}: ${e.subtitle}` : e.title, r = Z([...new Set((e.language ?? []).map(Y).filter((e) => e !== void 0))]), i = [
+}, W = "OpenLibrary", Ce = "ProjectGutenberg", G = (e) => e.length > 0 ? e : void 0, K = (e, t) => {
+	let n = e.title !== void 0 && e.subtitle !== void 0 ? `${e.title}: ${e.subtitle}` : e.title, r = G([...new Set((e.language ?? []).map(U).filter((e) => e !== void 0))]), i = [
 		...t.isbn === void 0 ? [] : [{
 			value: t.isbn,
 			scheme: "ISBN"
@@ -519,57 +485,55 @@ var n = (e) => {
 		...t.matchedProjectGutenbergIdentifier === void 0 ? [] : [t.matchedProjectGutenbergIdentifier],
 		...(e.id_project_gutenberg ?? []).map((e) => ({
 			value: e,
-			scheme: me
+			scheme: Ce
 		})),
 		...e.key === void 0 ? [] : [{
 			value: e.key,
-			scheme: X
+			scheme: W
 		}]
 	];
-	return I({
+	return P({
 		title: n,
 		cover: e.cover_i === void 0 ? void 0 : {
 			uri: `${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: G((e.subject ?? []).slice(0, 25)),
+		contributors: G((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: G(i)
 	});
-}, $ = "openLibrary", he = "https://openlibrary.org", ge = "https://covers.openlibrary.org", _e = [
+}, we = "openLibrary", Te = "https://openlibrary.org", Ee = "https://covers.openlibrary.org", De = [
 	"key",
 	"title",
 	"subtitle",
 	"author_name",
 	"first_publish_year",
-	"publisher",
 	"language",
 	"subject",
 	"number_of_pages_median",
 	"cover_i",
 	"id_project_gutenberg"
-].join(","), ve = (e) => {
+].join(","), Oe = (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 }
 	};
-}, ye = (e = {}) => {
-	let t = e.baseUrl ?? he, n = e.coversBaseUrl ?? ge, r = async (n, r) => {
+}, ke = (e = {}) => {
+	let t = e.baseUrl ?? Te, n = e.coversBaseUrl ?? Ee, 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", _e), i.searchParams.set("limit", String(r.limit));
+		i.searchParams.set("fields", De), i.searchParams.set("limit", String(r.limit));
 		let a = await (e.fetch ?? globalThis.fetch)(i.toString(), {
 			signal: r.signal,
 			headers: {
@@ -577,10 +541,10 @@ var n = (e) => {
 				...e.userAgent === void 0 ? {} : { "User-Agent": e.userAgent }
 			}
 		});
-		if (!a.ok) throw new R(a.status, `Open Library search failed with status ${a.status}`);
-		return J(await a.json());
+		if (!a.ok) throw new j(a.status, `Open Library search failed with status ${a.status}`);
+		return H(await a.json());
 	}, i = (e, r = {}) => e.map((e) => ({
-		metadata: Q(e, {
+		metadata: K(e, {
 			coversBaseUrl: n,
 			isbn: r.isbn,
 			matchedProjectGutenbergIdentifier: r.confirmedProjectGutenberg?.identifier
@@ -590,10 +554,10 @@ var n = (e) => {
 		raw: e
 	}));
 	return {
-		id: $,
+		id: we,
 		name: "Open Library",
 		search: async (e, t) => {
-			let n = e.isbn?.trim() || void 0, a = le(e);
+			let n = e.isbn?.trim() || void 0, a = L(e);
 			if (n !== void 0) {
 				let e = await r({ isbn: n }, t);
 				if (e.length > 0) return i(e, { isbn: n });
@@ -602,17 +566,204 @@ var n = (e) => {
 				let e = await r({ q: `id_project_gutenberg:${a.id}` }, t);
 				if (e.length > 0) return i(e, { confirmedProjectGutenberg: a });
 			}
-			let o = ve(e);
+			let o = Oe(e);
 			return o === void 0 ? [] : i(await r(o, t));
 		}
 	};
-}, be = (e) => [
+}, q = (e) => e.includes(":") ? e.slice(e.lastIndexOf(":") + 1) : e, Ae = (e) => e.type === "element", J = (e, t) => e.children.filter((e) => Ae(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;
+	}
+}, je = (e, t) => J(e, t).flatMap((e) => {
+	let t = X(Y(Y(e, "Description") ?? e, "value"));
+	return t === void 0 ? [] : [t];
+}), Me = (e) => {
+	let t = Z(e, "about"), n = t === void 0 ? void 0 : /(?:^|\/)ebooks\/(\d+)$/.exec(t)?.[1];
+	if (n === void 0) return;
+	let r = n.replace(/^0+(?=\d)/, "");
+	return r === "0" ? void 0 : r;
+}, Ne = (e) => {
+	let t = [];
+	for (let n of e.children) {
+		if (!Ae(n)) continue;
+		let e = n.name.toLowerCase() === "dcterms:creator", r = n.name.toLowerCase().startsWith("marcrel:");
+		if (!e && !r) continue;
+		let i = X(Y(Y(n, "agent") ?? n, "name"));
+		i !== void 0 && t.push({
+			name: i,
+			role: e ? "aut" : q(n.name).toLowerCase()
+		});
+	}
+	return t;
+}, Pe = (e) => J(e, "subject").flatMap((e) => {
+	let t = Y(e, "Description"), n = t === void 0 ? void 0 : Z(Y(t, "memberOf") ?? t, "resource"), r = X(t === void 0 ? void 0 : Y(t, "value"));
+	return r !== void 0 && n?.endsWith("/LCSH") === !0 ? [r] : [];
+}), Fe = (e) => {
+	let t = J(e, "hasFormat").flatMap((e) => {
+		let t = Y(e, "file");
+		if (t === void 0) return [];
+		let n = Z(t, "about"), r = J(t, "format").map((e) => X(Y(Y(e, "Description") ?? e, "value"))).find((e) => e?.startsWith("image/") === !0);
+		return n === void 0 || r === void 0 || !/\.cover\.(?:medium|small)\.[a-z0-9]+(?:$|[?#])/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, Ie = (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: ${t}`, { cause: e });
+	}
+	if (q(t.name).toLowerCase() !== "rdf") return;
+	let r = Y(t, "ebook");
+	if (r === void 0) return;
+	let i = Me(r);
+	if (i === void 0) return;
+	let a = [...new Set(je(r, "language"))], o = [...new Set(Pe(r))], s = [...new Set(je(r, "bookshelf"))], c = Ne(r);
+	return {
+		id: i,
+		title: X(Y(r, "title")),
+		publisher: X(Y(r, "publisher")),
+		issued: X(Y(r, "issued")),
+		originalPublication: X(Y(r, "marc260")),
+		rights: X(Y(r, "rights")),
+		description: X(Y(r, "description")),
+		summary: X(Y(r, "marc520")),
+		languages: Q(a),
+		subjects: Q(o),
+		bookshelves: Q(s),
+		contributors: Q(c),
+		cover: Fe(r)
+	};
+}, Le = 25, Re = {
+	id: "identifiers",
+	title: "title",
+	publisher: "publication.edition.publisher",
+	issued: "publication.edition.date",
+	originalPublication: "publication.original",
+	rights: "rights",
+	description: "description",
+	summary: "description",
+	languages: "languages",
+	subjects: "subjects",
+	bookshelves: "subjects",
+	contributors: "contributors",
+	cover: "cover"
+}, ze = {
+	art: "artist",
+	aut: "author",
+	clr: "colorist",
+	ctb: "contributor",
+	edt: "editor",
+	ill: "illustrator",
+	nrt: "narrator",
+	trl: "translator"
+}, Be = (e) => {
+	let t = /* @__PURE__ */ new Map();
+	for (let n of e.contributors ?? []) {
+		let e = n.name.trim().toLowerCase(), r = ze[n.role.trim().toLowerCase()] ?? n.role, i = t.get(e);
+		i === void 0 ? t.set(e, {
+			name: n.name,
+			roles: [r]
+		}) : i.roles.includes(r) || i.roles.push(r);
+	}
+	return t.size > 0 ? [...t.values()] : void 0;
+}, Ve = (e) => {
+	if (e === void 0) return;
+	let t = /^(\d{4})(?:-(\d{2})(?:-(\d{2}))?)?$/.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 }
+	};
+}, $ = (e, t) => e.split("$").slice(1).flatMap((e) => e[0]?.toLowerCase() === t.toLowerCase() ? [e.slice(1).trim()] : []).filter((e) => e.length > 0), He = (e) => {
+	if (e === void 0) return;
+	let t = $(e, "b"), n = (t.length === 1 ? t[0] : void 0)?.replace(/[\s,;:/]+$/u, "").trim(), r = $(e, "c"), i = [...new Set(r.flatMap((e) => [...e.matchAll(/\d{4}/g)].flatMap((e) => e[0] === void 0 ? [] : [Number(e[0])])))], a = P({
+		date: i.length === 1 ? { year: i[0] } : void 0,
+		publisher: n !== void 0 && n.length > 0 && !/^\[?s\.n\.?\]?$/i.test(n) ? n : void 0
+	});
+	return a.date !== void 0 || a.publisher !== void 0 ? a : void 0;
+}, Ue = (e, t) => {
+	try {
+		let n = new URL(e, t);
+		return n.protocol === "http:" || n.protocol === "https:" ? n.toString() : void 0;
+	} catch {
+		return;
+	}
+}, We = (e) => e.length > 0 ? e : void 0, Ge = (e, t) => {
+	let n = [t.matchedIdentifier, {
+		value: e.id,
+		scheme: F
+	}].filter((e, t, n) => n.findIndex((t) => t.value.trim().toLowerCase() === e.value.trim().toLowerCase() && t.scheme?.trim().toLowerCase() === e.scheme?.trim().toLowerCase()) === t), r = [.../* @__PURE__ */ new Set([...e.subjects ?? [], ...e.bookshelves ?? []])].slice(0, 25), i = e.cover === void 0 ? void 0 : Ue(e.cover.uri, t.baseUrl), a = i === void 0 ? void 0 : {
+		uri: i,
+		mediaType: e.cover?.mediaType,
+		confidence: "derived"
+	}, o = P({
+		date: Ve(e.issued),
+		publisher: e.publisher
+	}), s = P({
+		original: He(e.originalPublication),
+		edition: o.date !== void 0 || o.publisher !== void 0 ? o : void 0
+	});
+	return P({
+		title: e.title,
+		publication: s.original !== void 0 || s.edition !== void 0 ? s : void 0,
+		description: e.summary ?? e.description,
+		rights: e.rights,
+		languages: We([...new Set(e.languages ?? [])]),
+		subjects: We(r),
+		contributors: Be(e),
+		identifiers: n,
+		cover: a
+	});
+}, Ke = "projectGutenberg", qe = "https://www.gutenberg.org", Je = (e = {}) => {
+	let t = e.baseUrl ?? qe;
+	return {
+		id: Ke,
+		name: "Project Gutenberg",
+		search: async (n, r) => {
+			let i = L(n);
+			if (i === void 0) return [];
+			let a = new URL(`/cache/epub/${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 j(o.status, `Project Gutenberg metadata lookup failed with status ${o.status}`);
+			let s = Ie(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: Ge(s, {
+					baseUrl: t,
+					matchedIdentifier: i.identifier
+				})
+			}];
+		}
+	};
+}, Ye = (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 { E as METADATA_MATCH_WEIGHTS, R as MetadataProviderResponseError, X as OPEN_LIBRARY_IDENTIFIER_SCHEME, fe as OPEN_LIBRARY_MAX_SUBJECTS, $ as OPEN_LIBRARY_PROVIDER_ID, ye as createOpenLibraryProvider, V as fetchMetadata, be as hasSearchTerms, Y as marcLanguageToBcp47, L as mergeResolvedMetadata, n as metadataAuthors, c as normalizeForComparison, pe as openLibraryMetadataHomes, J as parseOpenLibrarySearchResponse, T as personNameSimilarity, Q as resolveOpenLibraryDoc, z as responseErrorStatus, F as scoreMetadataCandidate, u as textSimilarity, C as titleSimilarity, a as toIsbn13 };
+export { D as METADATA_MATCH_WEIGHTS, j as MetadataProviderResponseError, W as OPEN_LIBRARY_IDENTIFIER_SCHEME, xe as OPEN_LIBRARY_MAX_SUBJECTS, we as OPEN_LIBRARY_PROVIDER_ID, F as PROJECT_GUTENBERG_IDENTIFIER_SCHEME, Le as PROJECT_GUTENBERG_MAX_SUBJECTS, Ke as PROJECT_GUTENBERG_PROVIDER_ID, ke as createOpenLibraryProvider, Je as createProjectGutenbergProvider, pe as fetchMetadata, Ye as hasSearchTerms, U as marcLanguageToBcp47, r as metadataAuthors, l as normalizeForComparison, Se as openLibraryMetadataHomes, H as parseOpenLibrarySearchResponse, Ie as parseProjectGutenbergRdf, E as personNameSimilarity, L as projectGutenbergLookupFromMetadata, Re as projectGutenbergMetadataHomes, K as resolveOpenLibraryDoc, Ge as resolveProjectGutenbergRecord, M as responseErrorStatus, ce as scoreMetadataCandidate, d as textSimilarity, w as titleSimilarity, o as toIsbn13 };
 
 //# sourceMappingURL=index.js.map