nuxt-link-checker 0.1.6 → 0.3.1

package/README.md

@@ -10,7 +10,7 @@
 
 
 <p align="center">
-Improve your sites SEO by identifying and fixing link issues in your Nuxt v3 app.
+Identify and fix link issues for prerendered Nuxt 3 apps.
 </p>
 
 <p align="center">
@@ -18,7 +18,7 @@ Improve your sites SEO by identifying and fixing link issues in your Nuxt v3 app
 <tbody>
 <td align="center">
 <img width="800" height="0" /><br>
-<i>Status:</i> Early Access</b> <br>
+<i>Status:</i> Stable</b> <br>
 <sup> Please report any issues 🐛</sup><br>
 <sub>Made possible by my <a href="https://github.com/sponsors/harlan-zw">Sponsor Program 💖</a><br> Follow me <a href="https://twitter.com/harlan_zw">@harlan_zw</a> 🐦 • Join <a href="https://discord.gg/275MBUBvgP">Discord</a> for help</sub><br>
 <img width="800" height="0" />
@@ -27,15 +27,16 @@ Improve your sites SEO by identifying and fixing link issues in your Nuxt v3 app
 </table>
 </p>
 
+ℹ️ Looking for a complete SEO solution? Check out [Nuxt SEO Kit](https://github.com/harlan-zw/nuxt-seo-kit).
+
 ## Features
 
 - ✅ Discover broken links - 404s and internal redirects
+- 🚩 Warnings for bad practice links - absolute instead of relative and wrong trailing slash
 - 🕵️ Fail on build if broken links are found (optional)
 
 ## Install
 
-⚠️ The module is in early access and only works when pre-rendering.
-
 ```bash
 npm install --save-dev nuxt-link-checker
 
@@ -72,7 +73,6 @@ export default defineNuxtConfig({
 })
 ```  
 
-
 ### Set host (optional)
 
 You'll need to provide the host of your site so that the crawler can resolve absolute URLs that may be internal.
@@ -84,12 +84,40 @@ export default defineNuxtConfig({
     siteUrl: 'https://example.com',
   },
   // OR 
-  sitemap: {
+  linkChecker: {
     host: 'https://example.com',
   },
 })
 ```
 
+### Exclude URLs from throwing errors
+
+You can exclude URLs from throwing errors by adding them to the `exclude` array.
+
+For example, if you have an `/admin` route that is a separate application, you can ignore all `/admin` links with:
+
+```ts
+export default defineNuxtConfig({
+  linkChecker: {
+    exclude: [
+      '/admin/**'
+    ],
+  },
+})
+```
+
+### Disable errors on broken links
+
+You can disable errors on broken links by setting `failOn404` to `false`.
+
+```ts
+export default defineNuxtConfig({
+  linkChecker: {
+    failOn404: false,
+  },
+})
+```
+
 ## Module Config
 
 ### `failOn404`
@@ -99,10 +127,19 @@ export default defineNuxtConfig({
 
 If set to `true`, the build will fail if any broken links are found.
 
+### `exclude`
+
+- Type: `string[]`
+- Default: `[]`
+
+An array of URLs to exclude from the check. 
+
+This can be useful if you have a route that is not pre-rendered, but you know it will be valid.
+
 ### `host`
 
 - Type: `string`
-- Default: `runtimeConfig.public.siteUrl`
+- Default: `runtimeConfig.public.siteUrl || localhost`
 - Required: `false`
 
 The host of your site. This is required to validate absolute URLs which may be internal.
@@ -125,4 +162,4 @@ Whether internal links should have a trailing slash or not.
 
 ## License
 
-MIT License © 2022-PRESENT [Harlan Wilton](https://github.com/harlan-zw)
+MIT License © 2023-PRESENT [Harlan Wilton](https://github.com/harlan-zw)

package/dist/module.d.ts

@@ -10,11 +10,15 @@ interface ModuleOptions {
     /**
      * Your site hostname. Used to determine if absolute links are internal.
      */
-    host?: string;
+    host: string;
     /**
      * Whether the build should fail when a 404 is encountered.
      */
     failOn404: boolean;
+    /**
+     * Paths to ignore when checking links.
+     */
+    exclude: string[];
 }
 interface ModuleHooks {
 }

package/dist/module.json

@@ -5,5 +5,5 @@
     "bridge": false
   },
   "configKey": "linkChecker",
-  "version": "0.1.6"
+  "version": "0.3.1"
 }

package/dist/module.mjs

@@ -1,7 +1,8 @@
 import { defineNuxtModule } from '@nuxt/kit';
 import chalk from 'chalk';
-import { isRelative, parseURL } from 'ufo';
+import { parseURL, hasProtocol } from 'ufo';
 import { load } from 'cheerio';
+import { toRouteMatcher, createRouter } from 'radix3';
 
 const linkMap = {};
 const EXT_REGEX = /\.[\da-z]+$/;
@@ -11,39 +12,55 @@ function getExtension(path) {
 }
 function extractLinks(html, from, { host, trailingSlash }) {
   const links = [];
-  const _links = [];
+  const hostname = parseURL(host).host;
   const $ = load(html);
   $("[href]").each((i, el) => {
     const href = $(el).attr("href");
     if (!href)
       return;
-    if (host && !isRelative(href) && href.startsWith(host))
+    const url = parseURL(href);
+    if (hasProtocol(href) && !href.startsWith("/") && url.host !== hostname)
       return;
     if (!allowedExtensions.has(getExtension(href)))
       return;
-    _links.push({
-      href,
-      badTrailingSlash: href !== "/" && (trailingSlash && !href.endsWith("/") || !trailingSlash && href.endsWith("/")),
+    links.push({
+      pathname: url.pathname || "/",
+      url,
+      badAbsolute: Boolean(hostname) && hostname === url.host,
+      badTrailingSlash: url.pathname !== "/" && (trailingSlash && !url.pathname.endsWith("/") || !trailingSlash && url.pathname.endsWith("/")),
       element: $.html(el) || ""
     });
   });
-  for (const link of _links.filter(Boolean)) {
-    const parsed = parseURL(link.href);
-    if (parsed.protocol)
-      continue;
-    let { pathname } = parsed;
-    if (!pathname.startsWith("/")) {
-      const fromURL = new URL(from, "http://localhost");
-      pathname = new URL(pathname, fromURL).pathname;
-    }
-    links.push({
-      ...link,
-      href: pathname
-    });
-  }
   return links;
 }
 
+function createFilter(options = {}) {
+  const include = options.include || [];
+  const exclude = options.exclude || [];
+  if (include.length === 0 && exclude.length === 0)
+    return () => true;
+  return function(path) {
+    for (const v of [{ rules: exclude, result: false }, { rules: include, result: true }]) {
+      const regexRules = v.rules.filter((r) => r instanceof RegExp);
+      if (regexRules.some((r) => r.test(path)))
+        return v.result;
+      const stringRules = v.rules.filter((r) => typeof r === "string");
+      if (stringRules.length > 0) {
+        const routes = {};
+        for (const r of stringRules) {
+          if (r === path)
+            return v.result;
+          routes[r] = true;
+        }
+        const routeRulesMatcher = toRouteMatcher(createRouter({ routes, ...options }));
+        if (routeRulesMatcher.matchAll(path).length > 0)
+          return Boolean(v.result);
+      }
+    }
+    return include.length === 0;
+  };
+}
+
 const invalidStatusCodes = [404, 302, 301, 307, 303];
 const module = defineNuxtModule({
   meta: {
@@ -56,14 +73,18 @@ const module = defineNuxtModule({
   },
   defaults(nuxt) {
     return {
-      host: nuxt.options.runtimeConfig.public?.siteUrl,
+      host: nuxt.options.runtimeConfig.public?.siteUrl || "localhost",
       trailingSlash: nuxt.options.runtimeConfig.public?.trailingSlash || false,
-      failOn404: true
+      failOn404: true,
+      exclude: []
     };
   },
   setup(config, nuxt) {
     if (nuxt.options.dev)
       return;
+    const urlFilter = createFilter({
+      exclude: config.exclude
+    });
     nuxt.hooks.hook("nitro:init", async (nitro) => {
       const invalidRoutes = {};
       nitro.hooks.hook("prerender:generate", async (ctx) => {
@@ -76,16 +97,16 @@ const module = defineNuxtModule({
         const links = Object.entries(linkMap);
         if (!links.length)
           return;
-        nitro.logger.info("Scanning routes for broken links...");
+        nitro.logger.info(`Scanning routes for broken links... ${chalk.gray(`trailingSlashes: ${config.trailingSlash ? "`true`" : "`false`"}`)}`);
         let routeCount = 0;
         let badLinkCount = 0;
         links.forEach(([route, routes]) => {
           const brokenLinks = routes.map((r) => {
             return {
               ...r,
-              statusCode: invalidRoutes[r.href] || 200
+              statusCode: invalidRoutes[r.pathname] || 200
             };
-          }).filter((r) => r.statusCode !== 200 || r.badTrailingSlash);
+          }).filter((r) => r.statusCode !== 200 || r.badTrailingSlash || r.badAbsolute).filter((r) => urlFilter(r.pathname));
           if (brokenLinks.length) {
             nitro.logger.log(chalk.gray(
               `  ${Number(++routeCount) === links.length - 1 ? "\u2514\u2500" : "\u251C\u2500"} ${chalk.white(route)}`
@@ -97,9 +118,13 @@ const module = defineNuxtModule({
                 nitro.logger.log(chalk.red(
                   `   ${link.statusCode} ${link.statusCode === 404 ? "Not Found" : "Redirect"}`
                 ));
+              } else if (link.badAbsolute) {
+                nitro.logger.log(chalk.yellow(
+                  "   Absolute link, should be relative"
+                ));
               } else if (link.badTrailingSlash) {
                 nitro.logger.log(chalk.yellow(
-                  `   ${config.trailingSlash ? "Missing" : "Has added"} trailing slash`
+                  "   Incorrect trailing slash"
                 ));
               }
               nitro.logger.log(`   ${chalk.gray(link.element)}`);

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "nuxt-link-checker",
   "type": "module",
-  "version": "0.1.6",
+  "version": "0.3.1",
   "packageManager": "pnpm@7.18.0",
   "license": "MIT",
   "funding": "https://github.com/sponsors/harlan-zw",
@@ -29,20 +29,21 @@
     "@nuxt/kit": "3.0.0",
     "chalk": "^5.2.0",
     "cheerio": "1.0.0-rc.12",
+    "radix3": "^1.0.0",
     "ufo": "^1.0.1"
   },
   "devDependencies": {
-    "@antfu/eslint-config": "^0.34.0",
+    "@antfu/eslint-config": "^0.34.1",
     "@nuxt/kit": "3.0.0",
     "@nuxt/module-builder": "^0.2.1",
     "@nuxt/test-utils": "3.0.0",
     "@nuxtjs/eslint-config-typescript": "^12.0.0",
     "bumpp": "^8.2.1",
-    "eslint": "8.30.0",
+    "eslint": "8.32.0",
     "execa": "^6.1.0",
     "nuxt": "^3.0.0",
-    "pathe": "^1.0.0",
-    "vitest": "^0.25.8"
+    "pathe": "^1.1.0",
+    "vitest": "^0.28.1"
   },
   "scripts": {
     "lint": "eslint \"**/*.{ts,vue,json,yml}\" --fix",