feedcanon 1.2.2 → 1.3.0

package/README.md

@@ -41,8 +41,8 @@ This is a simplified flow. For complete details, see [How It Works](https://feed
 1. Fetch the input URL and parse the feed to establish reference content.
 2. Extract the feed's declared self URL (if present).
 3. Validate the self URL by fetching and comparing content.
-4. Generate URL variants ordered from cleanest to least clean.
-5. Test variants in order—the first one serving identical content wins.
+4. Generate URL candidates ordered from cleanest to least clean.
+5. Test candidates in order—the first one serving identical content wins.
 6. Upgrade HTTP to HTTPS if both serve identical content.
 
 ### Customization
@@ -53,7 +53,7 @@ Feedcanon is designed to be flexible. Every major component can be replaced or e
 - **Database lookup** — use `existsFn` to check if a URL already exists in your database.
 - **Custom fetch** — use your own HTTP client (Axios, Got, Ky, etc.)
 - **Custom parser** — bring your own parser (Feedsmith by default).
-- **Custom tiers** — define your own URL normalization variants.
+- **Custom tiers** — define your own URL normalization tiers.
 - **Custom platforms** — add handlers to normalize domain aliases (like FeedBurner).
 
 ## Quick Start

package/dist/defaults.cjs

@@ -1,3 +1,4 @@
+const require_utils = require('./utils.cjs');
 let feedsmith = require("feedsmith");
 
 //#region src/defaults.ts
@@ -183,15 +184,6 @@ const defaultFetch = async (url, options) => {
 		status: response.status
 	};
 };
-const trailingSlashPattern = /("(?:https?:\/\/|\/)[^"]+)\/([?"])/g;
-const neutralizeFeedUrls = (signature, url) => {
-	try {
-		const escapedHost = new URL("/", url).host.replace(/^www\./, "").replaceAll(".", "\\.");
-		return signature.replace(new RegExp(`https?://(?:www\\.)?${escapedHost}(?=[/"])(/)?`, "g"), "/").replace(trailingSlashPattern, "$1$2");
-	} catch {
-		return signature;
-	}
-};
 const retrieveSelfLink = (parsed) => {
 	switch (parsed.format) {
 		case "atom": return parsed.feed.links?.find((link) => link.rel === "self");
@@ -209,49 +201,48 @@ const defaultParser = {
 		return parsed.format === "json" ? parsed.feed.feed_url : retrieveSelfLink(parsed)?.href;
 	},
 	getSignature: (parsed, url) => {
-		if (parsed.format === "json") {
-			const originalSelfUrl = parsed.feed.feed_url;
-			parsed.feed.feed_url = void 0;
-			const signature$1 = JSON.stringify(parsed.feed);
-			parsed.feed.feed_url = originalSelfUrl;
-			return neutralizeFeedUrls(signature$1, url);
-		}
 		let signature;
-		let originalBuildDate;
-		let originalPubDate;
-		let originalLink;
-		if (parsed.format === "rss") {
-			originalBuildDate = parsed.feed.lastBuildDate;
-			originalPubDate = parsed.feed.pubDate;
-			originalLink = parsed.feed.link;
-			parsed.feed.lastBuildDate = void 0;
-			parsed.feed.pubDate = void 0;
-			parsed.feed.link = void 0;
-		} else if (parsed.format === "rdf") {
-			originalLink = parsed.feed.link;
-			parsed.feed.link = void 0;
-		} else if (parsed.format === "atom") {
-			originalBuildDate = parsed.feed.updated;
-			parsed.feed.updated = void 0;
-		}
-		const link = retrieveSelfLink(parsed);
-		if (!link) signature = JSON.stringify(parsed.feed);
-		else {
-			const originalSelfUrl = link.href;
-			link.href = void 0;
-			signature = JSON.stringify(parsed.feed);
-			link.href = originalSelfUrl;
+		let contentUrl;
+		if (parsed.format === "json") {
+			contentUrl = parsed.feed.home_page_url;
+			signature = require_utils.createSignature(parsed.feed, ["feed_url"]);
+		} else {
+			const selfLink = retrieveSelfLink(parsed);
+			const savedSelfHref = selfLink?.href;
+			if (selfLink) selfLink.href = void 0;
+			if (parsed.format === "rss") {
+				contentUrl = parsed.feed.link;
+				signature = require_utils.createSignature(parsed.feed, [
+					"lastBuildDate",
+					"pubDate",
+					"link",
+					"generator"
+				]);
+			} else if (parsed.format === "rdf") {
+				contentUrl = parsed.feed.link;
+				signature = require_utils.createSignature(parsed.feed, ["link"]);
+			} else signature = require_utils.createSignature(parsed.feed, ["updated", "generator"]);
+			if (selfLink) selfLink.href = savedSelfHref;
 		}
-		if (parsed.format === "rss") {
-			parsed.feed.lastBuildDate = originalBuildDate;
-			parsed.feed.pubDate = originalPubDate;
-			parsed.feed.link = originalLink;
-		} else if (parsed.format === "rdf") parsed.feed.link = originalLink;
-		else if (parsed.format === "atom") parsed.feed.updated = originalBuildDate;
-		return neutralizeFeedUrls(signature, url);
+		return require_utils.neutralizeUrls(signature, contentUrl ? [url, contentUrl] : [url]);
 	}
 };
 const defaultTiers = [
+	{
+		stripProtocol: false,
+		stripAuthentication: false,
+		stripWww: true,
+		stripTrailingSlash: true,
+		stripRootSlash: true,
+		collapseSlashes: true,
+		stripHash: true,
+		sortQueryParams: false,
+		stripQuery: true,
+		stripEmptyQuery: true,
+		normalizeEncoding: true,
+		normalizeUnicode: true,
+		convertToPunycode: true
+	},
 	{
 		stripProtocol: false,
 		stripAuthentication: false,
@@ -304,5 +295,4 @@ exports.defaultFetch = defaultFetch;
 exports.defaultNormalizeOptions = defaultNormalizeOptions;
 exports.defaultParser = defaultParser;
 exports.defaultStrippedParams = defaultStrippedParams;
-exports.defaultTiers = defaultTiers;
-exports.neutralizeFeedUrls = neutralizeFeedUrls;
+exports.defaultTiers = defaultTiers;

package/dist/defaults.d.cts

@@ -4,8 +4,7 @@ import { DefaultParserResult, FetchFn, NormalizeOptions, ParserAdapter, Tier } f
 declare const defaultStrippedParams: string[];
 declare const defaultNormalizeOptions: NormalizeOptions;
 declare const defaultFetch: FetchFn;
-declare const neutralizeFeedUrls: (signature: string, url: string) => string;
 declare const defaultParser: ParserAdapter<DefaultParserResult>;
 declare const defaultTiers: Array<Tier>;
 //#endregion
-export { defaultFetch, defaultNormalizeOptions, defaultParser, defaultStrippedParams, defaultTiers, neutralizeFeedUrls };
+export { defaultFetch, defaultNormalizeOptions, defaultParser, defaultStrippedParams, defaultTiers };

package/dist/defaults.d.ts

@@ -4,8 +4,7 @@ import { DefaultParserResult, FetchFn, NormalizeOptions, ParserAdapter, Tier } f
 declare const defaultStrippedParams: string[];
 declare const defaultNormalizeOptions: NormalizeOptions;
 declare const defaultFetch: FetchFn;
-declare const neutralizeFeedUrls: (signature: string, url: string) => string;
 declare const defaultParser: ParserAdapter<DefaultParserResult>;
 declare const defaultTiers: Array<Tier>;
 //#endregion
-export { defaultFetch, defaultNormalizeOptions, defaultParser, defaultStrippedParams, defaultTiers, neutralizeFeedUrls };
+export { defaultFetch, defaultNormalizeOptions, defaultParser, defaultStrippedParams, defaultTiers };

package/dist/defaults.js

@@ -1,3 +1,4 @@
+import { createSignature, neutralizeUrls } from "./utils.js";
 import { parseFeed } from "feedsmith";
 
 //#region src/defaults.ts
@@ -183,15 +184,6 @@ const defaultFetch = async (url, options) => {
 		status: response.status
 	};
 };
-const trailingSlashPattern = /("(?:https?:\/\/|\/)[^"]+)\/([?"])/g;
-const neutralizeFeedUrls = (signature, url) => {
-	try {
-		const escapedHost = new URL("/", url).host.replace(/^www\./, "").replaceAll(".", "\\.");
-		return signature.replace(new RegExp(`https?://(?:www\\.)?${escapedHost}(?=[/"])(/)?`, "g"), "/").replace(trailingSlashPattern, "$1$2");
-	} catch {
-		return signature;
-	}
-};
 const retrieveSelfLink = (parsed) => {
 	switch (parsed.format) {
 		case "atom": return parsed.feed.links?.find((link) => link.rel === "self");
@@ -209,49 +201,48 @@ const defaultParser = {
 		return parsed.format === "json" ? parsed.feed.feed_url : retrieveSelfLink(parsed)?.href;
 	},
 	getSignature: (parsed, url) => {
-		if (parsed.format === "json") {
-			const originalSelfUrl = parsed.feed.feed_url;
-			parsed.feed.feed_url = void 0;
-			const signature$1 = JSON.stringify(parsed.feed);
-			parsed.feed.feed_url = originalSelfUrl;
-			return neutralizeFeedUrls(signature$1, url);
-		}
 		let signature;
-		let originalBuildDate;
-		let originalPubDate;
-		let originalLink;
-		if (parsed.format === "rss") {
-			originalBuildDate = parsed.feed.lastBuildDate;
-			originalPubDate = parsed.feed.pubDate;
-			originalLink = parsed.feed.link;
-			parsed.feed.lastBuildDate = void 0;
-			parsed.feed.pubDate = void 0;
-			parsed.feed.link = void 0;
-		} else if (parsed.format === "rdf") {
-			originalLink = parsed.feed.link;
-			parsed.feed.link = void 0;
-		} else if (parsed.format === "atom") {
-			originalBuildDate = parsed.feed.updated;
-			parsed.feed.updated = void 0;
-		}
-		const link = retrieveSelfLink(parsed);
-		if (!link) signature = JSON.stringify(parsed.feed);
-		else {
-			const originalSelfUrl = link.href;
-			link.href = void 0;
-			signature = JSON.stringify(parsed.feed);
-			link.href = originalSelfUrl;
+		let contentUrl;
+		if (parsed.format === "json") {
+			contentUrl = parsed.feed.home_page_url;
+			signature = createSignature(parsed.feed, ["feed_url"]);
+		} else {
+			const selfLink = retrieveSelfLink(parsed);
+			const savedSelfHref = selfLink?.href;
+			if (selfLink) selfLink.href = void 0;
+			if (parsed.format === "rss") {
+				contentUrl = parsed.feed.link;
+				signature = createSignature(parsed.feed, [
+					"lastBuildDate",
+					"pubDate",
+					"link",
+					"generator"
+				]);
+			} else if (parsed.format === "rdf") {
+				contentUrl = parsed.feed.link;
+				signature = createSignature(parsed.feed, ["link"]);
+			} else signature = createSignature(parsed.feed, ["updated", "generator"]);
+			if (selfLink) selfLink.href = savedSelfHref;
 		}
-		if (parsed.format === "rss") {
-			parsed.feed.lastBuildDate = originalBuildDate;
-			parsed.feed.pubDate = originalPubDate;
-			parsed.feed.link = originalLink;
-		} else if (parsed.format === "rdf") parsed.feed.link = originalLink;
-		else if (parsed.format === "atom") parsed.feed.updated = originalBuildDate;
-		return neutralizeFeedUrls(signature, url);
+		return neutralizeUrls(signature, contentUrl ? [url, contentUrl] : [url]);
 	}
 };
 const defaultTiers = [
+	{
+		stripProtocol: false,
+		stripAuthentication: false,
+		stripWww: true,
+		stripTrailingSlash: true,
+		stripRootSlash: true,
+		collapseSlashes: true,
+		stripHash: true,
+		sortQueryParams: false,
+		stripQuery: true,
+		stripEmptyQuery: true,
+		normalizeEncoding: true,
+		normalizeUnicode: true,
+		convertToPunycode: true
+	},
 	{
 		stripProtocol: false,
 		stripAuthentication: false,
@@ -300,4 +291,4 @@ const defaultTiers = [
 ];
 
 //#endregion
-export { defaultFetch, defaultNormalizeOptions, defaultParser, defaultStrippedParams, defaultTiers, neutralizeFeedUrls };
+export { defaultFetch, defaultNormalizeOptions, defaultParser, defaultStrippedParams, defaultTiers };

package/dist/exports.cjs

@@ -1,6 +1,7 @@
-const require_defaults = require('./defaults.cjs');
 const require_utils = require('./utils.cjs');
+const require_defaults = require('./defaults.cjs');
 const require_index = require('./index.cjs');
+const require_wordpress = require('./probes/wordpress.cjs');
 const require_blogger = require('./rewrites/blogger.cjs');
 const require_feedburner = require('./rewrites/feedburner.cjs');
 
@@ -15,4 +16,5 @@ exports.findCanonical = require_index.findCanonical;
 exports.fixMalformedProtocol = require_utils.fixMalformedProtocol;
 exports.normalizeUrl = require_utils.normalizeUrl;
 exports.resolveFeedProtocol = require_utils.resolveFeedProtocol;
-exports.resolveUrl = require_utils.resolveUrl;
+exports.resolveUrl = require_utils.resolveUrl;
+exports.wordpressProbe = require_wordpress.wordpressProbe;

package/dist/exports.d.cts

@@ -1,7 +1,8 @@
-import { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Rewrite } from "./types.cjs";
+import { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Probe, Rewrite, Tier } from "./types.cjs";
 import { defaultFetch, defaultParser, defaultStrippedParams, defaultTiers } from "./defaults.cjs";
 import { findCanonical } from "./index.cjs";
+import { wordpressProbe } from "./probes/wordpress.cjs";
 import { bloggerRewrite } from "./rewrites/blogger.cjs";
 import { feedburnerRewrite } from "./rewrites/feedburner.cjs";
 import { addMissingProtocol, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl } from "./utils.cjs";
-export { type DefaultParserResult, type ExistsFn, type FetchFn, type FetchFnOptions, type FetchFnResponse, type FindCanonicalOptions, type NormalizeOptions, type OnExistsFn, type OnFetchFn, type OnMatchFn, type ParserAdapter, type Rewrite, addMissingProtocol, bloggerRewrite, defaultFetch, defaultParser, defaultStrippedParams, defaultTiers, feedburnerRewrite, findCanonical, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl };
+export { type DefaultParserResult, type ExistsFn, type FetchFn, type FetchFnOptions, type FetchFnResponse, type FindCanonicalOptions, type NormalizeOptions, type OnExistsFn, type OnFetchFn, type OnMatchFn, type ParserAdapter, type Probe, type Rewrite, type Tier, addMissingProtocol, bloggerRewrite, defaultFetch, defaultParser, defaultStrippedParams, defaultTiers, feedburnerRewrite, findCanonical, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl, wordpressProbe };

package/dist/exports.d.ts

@@ -1,7 +1,8 @@
-import { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Rewrite } from "./types.js";
+import { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Probe, Rewrite, Tier } from "./types.js";
 import { defaultFetch, defaultParser, defaultStrippedParams, defaultTiers } from "./defaults.js";
 import { findCanonical } from "./index.js";
+import { wordpressProbe } from "./probes/wordpress.js";
 import { bloggerRewrite } from "./rewrites/blogger.js";
 import { feedburnerRewrite } from "./rewrites/feedburner.js";
 import { addMissingProtocol, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl } from "./utils.js";
-export { type DefaultParserResult, type ExistsFn, type FetchFn, type FetchFnOptions, type FetchFnResponse, type FindCanonicalOptions, type NormalizeOptions, type OnExistsFn, type OnFetchFn, type OnMatchFn, type ParserAdapter, type Rewrite, addMissingProtocol, bloggerRewrite, defaultFetch, defaultParser, defaultStrippedParams, defaultTiers, feedburnerRewrite, findCanonical, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl };
+export { type DefaultParserResult, type ExistsFn, type FetchFn, type FetchFnOptions, type FetchFnResponse, type FindCanonicalOptions, type NormalizeOptions, type OnExistsFn, type OnFetchFn, type OnMatchFn, type ParserAdapter, type Probe, type Rewrite, type Tier, addMissingProtocol, bloggerRewrite, defaultFetch, defaultParser, defaultStrippedParams, defaultTiers, feedburnerRewrite, findCanonical, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl, wordpressProbe };

package/dist/exports.js

@@ -1,7 +1,8 @@
-import { defaultFetch, defaultParser, defaultStrippedParams, defaultTiers } from "./defaults.js";
 import { addMissingProtocol, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl } from "./utils.js";
+import { defaultFetch, defaultParser, defaultStrippedParams, defaultTiers } from "./defaults.js";
 import { findCanonical } from "./index.js";
+import { wordpressProbe } from "./probes/wordpress.js";
 import { bloggerRewrite } from "./rewrites/blogger.js";
 import { feedburnerRewrite } from "./rewrites/feedburner.js";
 
-export { addMissingProtocol, bloggerRewrite, defaultFetch, defaultParser, defaultStrippedParams, defaultTiers, feedburnerRewrite, findCanonical, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl };
+export { addMissingProtocol, bloggerRewrite, defaultFetch, defaultParser, defaultStrippedParams, defaultTiers, feedburnerRewrite, findCanonical, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl, wordpressProbe };

package/dist/index.cjs

@@ -1,15 +1,15 @@
-const require_defaults = require('./defaults.cjs');
 const require_utils = require('./utils.cjs');
+const require_defaults = require('./defaults.cjs');
 
 //#region src/index.ts
 async function findCanonical(inputUrl, options) {
-	const { parser = require_defaults.defaultParser, fetchFn = require_defaults.defaultFetch, existsFn, tiers = require_defaults.defaultTiers, rewrites, stripQueryParams = require_defaults.defaultStrippedParams, onFetch, onMatch, onExists } = options ?? {};
+	const { parser = require_defaults.defaultParser, fetchFn = require_defaults.defaultFetch, existsFn, tiers = require_defaults.defaultTiers, rewrites, probes, stripQueryParams = require_defaults.defaultStrippedParams, onFetch, onMatch, onExists } = options ?? {};
 	const stripParams = (url) => {
-		return stripQueryParams?.length ? require_utils.normalizeUrl(url, {
+		return require_utils.normalizeUrl(url, {
 			stripQueryParams,
 			sortQueryParams: true,
 			stripEmptyQuery: true
-		}) : url;
+		});
 	};
 	const resolveAndApplyRewrites = (url, baseUrl) => {
 		const resolved = require_utils.resolveUrl(url, baseUrl);
@@ -73,7 +73,7 @@ async function findCanonical(inputUrl, options) {
 		if (!await compareWithInitialResponse(response.body, response.url)) return;
 		return response;
 	};
-	let variantSourceUrl = initialResponseUrl;
+	let candidateSourceUrl = initialResponseUrl;
 	if (selfRequestUrl && selfRequestUrl !== initialResponseUrl) {
 		const urlsToTry = [selfRequestUrl];
 		if (selfRequestUrl.startsWith("https://")) urlsToTry.push(selfRequestUrl.replace("https://", "http://"));
@@ -86,42 +86,53 @@ async function findCanonical(inputUrl, options) {
 					response,
 					feed: initialResponseFeed
 				});
-				variantSourceUrl = resolveAndApplyRewrites(response.url) ?? initialResponseUrl;
-				variantSourceUrl = stripParams(variantSourceUrl);
+				candidateSourceUrl = resolveAndApplyRewrites(response.url) ?? initialResponseUrl;
+				candidateSourceUrl = stripParams(candidateSourceUrl);
 				break;
 			}
 		}
 	}
-	const variantUrls = new Set(tiers.map((tier) => resolveAndApplyRewrites(require_utils.normalizeUrl(variantSourceUrl, tier))).filter((variantUrl) => !!variantUrl));
-	variantUrls.add(variantSourceUrl);
-	let winningUrl = variantSourceUrl;
-	for (const variantUrl of variantUrls) {
+	if (probes?.length) candidateSourceUrl = await require_utils.applyProbes(candidateSourceUrl, probes, async (candidateUrl) => {
+		const response = await fetchAndCompare(candidateUrl);
+		if (response) {
+			onMatch?.({
+				url: candidateUrl,
+				response,
+				feed: initialResponseFeed
+			});
+			return stripParams(resolveAndApplyRewrites(response.url) ?? candidateUrl);
+		}
+	});
+	const candidateUrls = new Set(tiers.map((tier) => resolveAndApplyRewrites(require_utils.normalizeUrl(candidateSourceUrl, tier))).filter((candidateUrl) => !!candidateUrl));
+	candidateUrls.add(candidateSourceUrl);
+	let winningUrl = candidateSourceUrl;
+	for (const candidateUrl of candidateUrls) {
 		if (existsFn) {
-			const data = await existsFn(variantUrl);
+			const data = await existsFn(candidateUrl);
 			if (data !== void 0) {
 				onExists?.({
-					url: variantUrl,
+					url: candidateUrl,
 					data
 				});
-				return variantUrl;
+				return candidateUrl;
 			}
 		}
-		if (variantUrl === variantSourceUrl) continue;
-		if (variantUrl === initialResponseUrl) {
+		if (candidateUrl === candidateSourceUrl) continue;
+		if (candidateUrl === initialResponseUrl) {
 			winningUrl = initialResponseUrl;
 			break;
 		}
-		const variantResponse = await fetchAndCompare(variantUrl);
-		if (variantResponse) {
-			let variantResponseUrl = resolveAndApplyRewrites(variantResponse.url);
-			if (variantResponseUrl) variantResponseUrl = stripParams(variantResponseUrl);
-			if (variantResponseUrl === variantSourceUrl || variantResponseUrl === initialResponseUrl) continue;
+		const candidateResponse = await fetchAndCompare(candidateUrl);
+		if (candidateResponse) {
+			let candidateResponseUrl = resolveAndApplyRewrites(candidateResponse.url);
+			if (candidateResponseUrl) candidateResponseUrl = stripParams(candidateResponseUrl);
+			if (candidateResponseUrl === candidateSourceUrl || candidateResponseUrl === initialResponseUrl) continue;
 			onMatch?.({
-				url: variantUrl,
-				response: variantResponse,
+				url: candidateUrl,
+				response: candidateResponse,
 				feed: initialResponseFeed
 			});
-			winningUrl = variantUrl;
+			winningUrl = candidateUrl;
 			break;
 		}
 	}

package/dist/index.js

@@ -1,15 +1,15 @@
+import { applyProbes, applyRewrites, normalizeUrl, resolveUrl } from "./utils.js";
 import { defaultFetch, defaultParser, defaultStrippedParams, defaultTiers } from "./defaults.js";
-import { applyRewrites, normalizeUrl, resolveUrl } from "./utils.js";
 
 //#region src/index.ts
 async function findCanonical(inputUrl, options) {
-	const { parser = defaultParser, fetchFn = defaultFetch, existsFn, tiers = defaultTiers, rewrites, stripQueryParams = defaultStrippedParams, onFetch, onMatch, onExists } = options ?? {};
+	const { parser = defaultParser, fetchFn = defaultFetch, existsFn, tiers = defaultTiers, rewrites, probes, stripQueryParams = defaultStrippedParams, onFetch, onMatch, onExists } = options ?? {};
 	const stripParams = (url) => {
-		return stripQueryParams?.length ? normalizeUrl(url, {
+		return normalizeUrl(url, {
 			stripQueryParams,
 			sortQueryParams: true,
 			stripEmptyQuery: true
-		}) : url;
+		});
 	};
 	const resolveAndApplyRewrites = (url, baseUrl) => {
 		const resolved = resolveUrl(url, baseUrl);
@@ -73,7 +73,7 @@ async function findCanonical(inputUrl, options) {
 		if (!await compareWithInitialResponse(response.body, response.url)) return;
 		return response;
 	};
-	let variantSourceUrl = initialResponseUrl;
+	let candidateSourceUrl = initialResponseUrl;
 	if (selfRequestUrl && selfRequestUrl !== initialResponseUrl) {
 		const urlsToTry = [selfRequestUrl];
 		if (selfRequestUrl.startsWith("https://")) urlsToTry.push(selfRequestUrl.replace("https://", "http://"));
@@ -86,42 +86,53 @@ async function findCanonical(inputUrl, options) {
 					response,
 					feed: initialResponseFeed
 				});
-				variantSourceUrl = resolveAndApplyRewrites(response.url) ?? initialResponseUrl;
-				variantSourceUrl = stripParams(variantSourceUrl);
+				candidateSourceUrl = resolveAndApplyRewrites(response.url) ?? initialResponseUrl;
+				candidateSourceUrl = stripParams(candidateSourceUrl);
 				break;
 			}
 		}
 	}
-	const variantUrls = new Set(tiers.map((tier) => resolveAndApplyRewrites(normalizeUrl(variantSourceUrl, tier))).filter((variantUrl) => !!variantUrl));
-	variantUrls.add(variantSourceUrl);
-	let winningUrl = variantSourceUrl;
-	for (const variantUrl of variantUrls) {
+	if (probes?.length) candidateSourceUrl = await applyProbes(candidateSourceUrl, probes, async (candidateUrl) => {
+		const response = await fetchAndCompare(candidateUrl);
+		if (response) {
+			onMatch?.({
+				url: candidateUrl,
+				response,
+				feed: initialResponseFeed
+			});
+			return stripParams(resolveAndApplyRewrites(response.url) ?? candidateUrl);
+		}
+	});
+	const candidateUrls = new Set(tiers.map((tier) => resolveAndApplyRewrites(normalizeUrl(candidateSourceUrl, tier))).filter((candidateUrl) => !!candidateUrl));
+	candidateUrls.add(candidateSourceUrl);
+	let winningUrl = candidateSourceUrl;
+	for (const candidateUrl of candidateUrls) {
 		if (existsFn) {
-			const data = await existsFn(variantUrl);
+			const data = await existsFn(candidateUrl);
 			if (data !== void 0) {
 				onExists?.({
-					url: variantUrl,
+					url: candidateUrl,
 					data
 				});
-				return variantUrl;
+				return candidateUrl;
 			}
 		}
-		if (variantUrl === variantSourceUrl) continue;
-		if (variantUrl === initialResponseUrl) {
+		if (candidateUrl === candidateSourceUrl) continue;
+		if (candidateUrl === initialResponseUrl) {
 			winningUrl = initialResponseUrl;
 			break;
 		}
-		const variantResponse = await fetchAndCompare(variantUrl);
-		if (variantResponse) {
-			let variantResponseUrl = resolveAndApplyRewrites(variantResponse.url);
-			if (variantResponseUrl) variantResponseUrl = stripParams(variantResponseUrl);
-			if (variantResponseUrl === variantSourceUrl || variantResponseUrl === initialResponseUrl) continue;
+		const candidateResponse = await fetchAndCompare(candidateUrl);
+		if (candidateResponse) {
+			let candidateResponseUrl = resolveAndApplyRewrites(candidateResponse.url);
+			if (candidateResponseUrl) candidateResponseUrl = stripParams(candidateResponseUrl);
+			if (candidateResponseUrl === candidateSourceUrl || candidateResponseUrl === initialResponseUrl) continue;
 			onMatch?.({
-				url: variantUrl,
-				response: variantResponse,
+				url: candidateUrl,
+				response: candidateResponse,
 				feed: initialResponseFeed
 			});
-			winningUrl = variantUrl;
+			winningUrl = candidateUrl;
 			break;
 		}
 	}

package/dist/probes/wordpress.cjs

@@ -0,0 +1,49 @@
+
+//#region src/probes/wordpress.ts
+const feedTypes = [
+	"atom",
+	"rss2",
+	"rss",
+	"rdf"
+];
+const wordpressProbe = {
+	match: (url) => {
+		const feed = url.searchParams.get("feed")?.toLowerCase();
+		if (!feed) return false;
+		const type = feed.startsWith("comments-") ? feed.slice(9) : feed;
+		return feedTypes.includes(type);
+	},
+	getCandidates: (url) => {
+		const feed = url.searchParams.get("feed")?.toLowerCase();
+		if (!feed) return [];
+		const candidates = [];
+		const isComment = feed.startsWith("comments-");
+		const type = isComment ? feed.slice(9) : feed;
+		if ((isComment ? /\/comments\/feed(\/|$)/ : /\/feed(\/|$)/).test(url.pathname)) {
+			const withoutSlash = new URL(url);
+			withoutSlash.pathname = url.pathname.replace(/\/$/, "");
+			withoutSlash.searchParams.delete("feed");
+			candidates.push(withoutSlash.href);
+			const withSlash$1 = new URL(url);
+			withSlash$1.pathname = url.pathname.replace(/\/?$/, "/");
+			withSlash$1.searchParams.delete("feed");
+			candidates.push(withSlash$1.href);
+			return candidates;
+		}
+		const basePath = url.pathname.replace(/\/$/, "");
+		const feedSegment = type === "atom" ? "/feed/atom" : "/feed";
+		const feedPath = isComment ? `/comments${feedSegment}` : feedSegment;
+		const primary = new URL(url);
+		primary.pathname = basePath + feedPath;
+		primary.searchParams.delete("feed");
+		candidates.push(primary.href);
+		const withSlash = new URL(url);
+		withSlash.pathname = `${basePath}${feedPath}/`;
+		withSlash.searchParams.delete("feed");
+		candidates.push(withSlash.href);
+		return candidates;
+	}
+};
+
+//#endregion
+exports.wordpressProbe = wordpressProbe;

package/dist/probes/wordpress.d.cts

@@ -0,0 +1,6 @@
+import { Probe } from "../types.cjs";
+
+//#region src/probes/wordpress.d.ts
+declare const wordpressProbe: Probe;
+//#endregion
+export { wordpressProbe };

package/dist/probes/wordpress.d.ts

@@ -0,0 +1,6 @@
+import { Probe } from "../types.js";
+
+//#region src/probes/wordpress.d.ts
+declare const wordpressProbe: Probe;
+//#endregion
+export { wordpressProbe };

package/dist/probes/wordpress.js

@@ -0,0 +1,48 @@
+//#region src/probes/wordpress.ts
+const feedTypes = [
+	"atom",
+	"rss2",
+	"rss",
+	"rdf"
+];
+const wordpressProbe = {
+	match: (url) => {
+		const feed = url.searchParams.get("feed")?.toLowerCase();
+		if (!feed) return false;
+		const type = feed.startsWith("comments-") ? feed.slice(9) : feed;
+		return feedTypes.includes(type);
+	},
+	getCandidates: (url) => {
+		const feed = url.searchParams.get("feed")?.toLowerCase();
+		if (!feed) return [];
+		const candidates = [];
+		const isComment = feed.startsWith("comments-");
+		const type = isComment ? feed.slice(9) : feed;
+		if ((isComment ? /\/comments\/feed(\/|$)/ : /\/feed(\/|$)/).test(url.pathname)) {
+			const withoutSlash = new URL(url);
+			withoutSlash.pathname = url.pathname.replace(/\/$/, "");
+			withoutSlash.searchParams.delete("feed");
+			candidates.push(withoutSlash.href);
+			const withSlash$1 = new URL(url);
+			withSlash$1.pathname = url.pathname.replace(/\/?$/, "/");
+			withSlash$1.searchParams.delete("feed");
+			candidates.push(withSlash$1.href);
+			return candidates;
+		}
+		const basePath = url.pathname.replace(/\/$/, "");
+		const feedSegment = type === "atom" ? "/feed/atom" : "/feed";
+		const feedPath = isComment ? `/comments${feedSegment}` : feedSegment;
+		const primary = new URL(url);
+		primary.pathname = basePath + feedPath;
+		primary.searchParams.delete("feed");
+		candidates.push(primary.href);
+		const withSlash = new URL(url);
+		withSlash.pathname = `${basePath}${feedPath}/`;
+		withSlash.searchParams.delete("feed");
+		candidates.push(withSlash.href);
+		return candidates;
+	}
+};
+
+//#endregion
+export { wordpressProbe };

package/dist/types.d.cts

@@ -11,6 +11,10 @@ type Rewrite = {
   match: (url: URL) => boolean;
   rewrite: (url: URL) => URL;
 };
+type Probe = {
+  match: (url: URL) => boolean;
+  getCandidates: (url: URL) => Array<string>;
+};
 type NormalizeOptions = {
   stripProtocol?: boolean;
   stripAuthentication?: boolean;
@@ -45,8 +49,9 @@ type FindCanonicalOptions<TFeed = DefaultParserResult, TResponse extends FetchFn
   parser?: ParserAdapter<TFeed>;
   fetchFn?: FetchFn<TResponse>;
   existsFn?: ExistsFn<TExisting>;
-  tiers?: Array<Tier>;
   rewrites?: Array<Rewrite>;
+  probes?: Array<Probe>;
+  tiers?: Array<Tier>;
   stripQueryParams?: Array<string>;
   onFetch?: OnFetchFn<TResponse>;
   onMatch?: OnMatchFn<TFeed, TResponse>;
@@ -65,4 +70,4 @@ type FetchFnResponse = {
 };
 type FetchFn<TResponse extends FetchFnResponse = FetchFnResponse> = (url: string, options?: FetchFnOptions) => Promise<TResponse>;
 //#endregion
-export { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Rewrite, Tier };
+export { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Probe, Rewrite, Tier };

package/dist/types.d.ts

@@ -11,6 +11,10 @@ type Rewrite = {
   match: (url: URL) => boolean;
   rewrite: (url: URL) => URL;
 };
+type Probe = {
+  match: (url: URL) => boolean;
+  getCandidates: (url: URL) => Array<string>;
+};
 type NormalizeOptions = {
   stripProtocol?: boolean;
   stripAuthentication?: boolean;
@@ -45,8 +49,9 @@ type FindCanonicalOptions<TFeed = DefaultParserResult, TResponse extends FetchFn
   parser?: ParserAdapter<TFeed>;
   fetchFn?: FetchFn<TResponse>;
   existsFn?: ExistsFn<TExisting>;
-  tiers?: Array<Tier>;
   rewrites?: Array<Rewrite>;
+  probes?: Array<Probe>;
+  tiers?: Array<Tier>;
   stripQueryParams?: Array<string>;
   onFetch?: OnFetchFn<TResponse>;
   onMatch?: OnMatchFn<TFeed, TResponse>;
@@ -65,4 +70,4 @@ type FetchFnResponse = {
 };
 type FetchFn<TResponse extends FetchFnResponse = FetchFnResponse> = (url: string, options?: FetchFnOptions) => Promise<TResponse>;
 //#endregion
-export { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Rewrite, Tier };
+export { DefaultParserResult, ExistsFn, FetchFn, FetchFnOptions, FetchFnResponse, FindCanonicalOptions, NormalizeOptions, OnExistsFn, OnFetchFn, OnMatchFn, ParserAdapter, Probe, Rewrite, Tier };

package/dist/utils.cjs

@@ -155,11 +155,51 @@ const applyRewrites = (url, rewrites) => {
 		return url;
 	}
 };
+const applyProbes = async (url, probes, testCandidate) => {
+	try {
+		const parsed = new URL(url);
+		for (const probe of probes) {
+			if (!probe.match(parsed)) continue;
+			for (const candidate of probe.getCandidates(parsed)) {
+				const result = await testCandidate(candidate);
+				if (result) return result;
+			}
+			break;
+		}
+		return url;
+	} catch {
+		return url;
+	}
+};
+const createSignature = (object, fields) => {
+	const saved = fields.map((key) => [key, object[key]]);
+	for (const key of fields) object[key] = void 0;
+	const signature = JSON.stringify(object);
+	for (const [key, val] of saved) object[key] = val;
+	return signature;
+};
+const trailingSlashPattern = /("(?:https?:\/\/|\/)[^"]+)\/([?"])/g;
+const neutralizeUrls = (text, urls) => {
+	const escapeHost = (url) => {
+		try {
+			return new URL("/", url).host.replace(/^www\./, "").replaceAll(".", "\\.");
+		} catch {
+			return;
+		}
+	};
+	const hosts = urls.map(escapeHost).filter(Boolean);
+	if (hosts.length === 0) return text;
+	const hostPattern = hosts.length === 1 ? hosts[0] : `(?:${hosts.join("|")})`;
+	return text.replace(new RegExp(`https?://(?:www\\.)?${hostPattern}(?=[/"])(/)?`, "g"), "/").replace(trailingSlashPattern, "$1$2");
+};
 
 //#endregion
 exports.addMissingProtocol = addMissingProtocol;
+exports.applyProbes = applyProbes;
 exports.applyRewrites = applyRewrites;
+exports.createSignature = createSignature;
 exports.fixMalformedProtocol = fixMalformedProtocol;
+exports.neutralizeUrls = neutralizeUrls;
 exports.normalizeUrl = normalizeUrl;
 exports.resolveFeedProtocol = resolveFeedProtocol;
 exports.resolveUrl = resolveUrl;

package/dist/utils.js

@@ -155,6 +155,43 @@ const applyRewrites = (url, rewrites) => {
 		return url;
 	}
 };
+const applyProbes = async (url, probes, testCandidate) => {
+	try {
+		const parsed = new URL(url);
+		for (const probe of probes) {
+			if (!probe.match(parsed)) continue;
+			for (const candidate of probe.getCandidates(parsed)) {
+				const result = await testCandidate(candidate);
+				if (result) return result;
+			}
+			break;
+		}
+		return url;
+	} catch {
+		return url;
+	}
+};
+const createSignature = (object, fields) => {
+	const saved = fields.map((key) => [key, object[key]]);
+	for (const key of fields) object[key] = void 0;
+	const signature = JSON.stringify(object);
+	for (const [key, val] of saved) object[key] = val;
+	return signature;
+};
+const trailingSlashPattern = /("(?:https?:\/\/|\/)[^"]+)\/([?"])/g;
+const neutralizeUrls = (text, urls) => {
+	const escapeHost = (url) => {
+		try {
+			return new URL("/", url).host.replace(/^www\./, "").replaceAll(".", "\\.");
+		} catch {
+			return;
+		}
+	};
+	const hosts = urls.map(escapeHost).filter(Boolean);
+	if (hosts.length === 0) return text;
+	const hostPattern = hosts.length === 1 ? hosts[0] : `(?:${hosts.join("|")})`;
+	return text.replace(new RegExp(`https?://(?:www\\.)?${hostPattern}(?=[/"])(/)?`, "g"), "/").replace(trailingSlashPattern, "$1$2");
+};
 
 //#endregion
-export { addMissingProtocol, applyRewrites, fixMalformedProtocol, normalizeUrl, resolveFeedProtocol, resolveUrl };
+export { addMissingProtocol, applyProbes, applyRewrites, createSignature, fixMalformedProtocol, neutralizeUrls, normalizeUrl, resolveFeedProtocol, resolveUrl };

package/package.json

@@ -63,5 +63,5 @@
     "tsdown": "^0.18.4",
     "vitepress": "^1.6.4"
   },
-  "version": "1.2.2"
+  "version": "1.3.0"
 }