@docusaurus/core 0.0.0-5797 → 0.0.0-5801

package/lib/client/BrokenLinksContext.js

@@ -11,10 +11,10 @@ export const createStatefulBrokenLinks = () => {
     const allLinks = new Set();
     return {
         collectAnchor: (anchor) => {
-            allAnchors.add(anchor);
+            typeof anchor !== 'undefined' && allAnchors.add(anchor);
         },
         collectLink: (link) => {
-            allLinks.add(link);
+            typeof link !== 'undefined' && allLinks.add(link);
         },
         getCollectedAnchors: () => [...allAnchors],
         getCollectedLinks: () => [...allLinks],

package/lib/client/exports/Link.js

@@ -97,11 +97,16 @@ function Link({ isNavLink, to, href, activeClassName, isActive, 'data-noBrokenLi
             }
         };
     }, [ioRef, targetLink, IOSupported, isInternal]);
+    // It is simple local anchor link targeting current page?
     const isAnchorLink = targetLink?.startsWith('#') ?? false;
+    // Should we use a regular <a> tag instead of React-Router Link component?
     const isRegularHtmlLink = !targetLink || !isInternal || isAnchorLink;
-    if (!isRegularHtmlLink && !noBrokenLinkCheck) {
+    if (!noBrokenLinkCheck && (isAnchorLink || !isRegularHtmlLink)) {
         brokenLinks.collectLink(targetLink);
     }
+    if (props.id) {
+        brokenLinks.collectAnchor(props.id);
+    }
     return isRegularHtmlLink ? (
     // eslint-disable-next-line jsx-a11y/anchor-has-content, @docusaurus/no-html-links
     <a ref={innerRef} href={targetLink} {...(targetLinkUnprefixed &&

package/lib/server/brokenLinks.js

@@ -14,15 +14,20 @@ const react_router_config_1 = require("react-router-config");
 const utils_1 = require("@docusaurus/utils");
 const utils_2 = require("./utils");
 function getBrokenLinksForPage({ collectedLinks, pagePath, pageLinks, routes, }) {
-    // console.log('routes:', routes);
+    const allCollectedPaths = new Set(Object.keys(collectedLinks));
     function isPathBrokenLink(linkPath) {
-        const matchedRoutes = [linkPath.pathname, decodeURI(linkPath.pathname)]
+        const pathnames = [linkPath.pathname, decodeURI(linkPath.pathname)];
+        const matchedRoutes = pathnames
             // @ts-expect-error: React router types RouteConfig with an actual React
             // component, but we load route components with string paths.
             // We don't actually access component here, so it's fine.
             .map((l) => (0, react_router_config_1.matchRoutes)(routes, l))
             .flat();
-        return matchedRoutes.length === 0;
+        // The link path is broken if:
+        // - it doesn't match any route
+        // - it doesn't match any collected path
+        return (matchedRoutes.length === 0 &&
+            !pathnames.some((p) => allCollectedPaths.has(p)));
     }
     function isAnchorBrokenLink(linkPath) {
         const { pathname, hash } = linkPath;
@@ -30,6 +35,12 @@ function getBrokenLinksForPage({ collectedLinks, pagePath, pageLinks, routes, })
         if (hash === undefined) {
             return false;
         }
+        // Link has empty hash ("#", "/page#"...): we do not report it as broken
+        // Empty hashes are used for various weird reasons, by us and other users...
+        // See for example: https://github.com/facebook/docusaurus/pull/6003
+        if (hash === '') {
+            return false;
+        }
         const targetPage = collectedLinks[pathname] || collectedLinks[decodeURI(pathname)];
         // link with anchor to a page that does not exist (or did not collect any
         // link/anchor) is considered as a broken anchor
@@ -38,7 +49,8 @@ function getBrokenLinksForPage({ collectedLinks, pagePath, pageLinks, routes, })
         }
         // it's a broken anchor if the target page exists
         // but the anchor does not exist on that page
-        return !targetPage.anchors.includes(hash);
+        const hashes = [hash, decodeURIComponent(hash)];
+        return !targetPage.anchors.some((anchor) => hashes.includes(anchor));
     }
     const brokenLinks = pageLinks.flatMap((link) => {
         const linkPath = (0, utils_1.parseURLPath)(link, pagePath);
@@ -76,13 +88,22 @@ function filterIntermediateRoutes(routesInput) {
 }
 function getBrokenLinks({ collectedLinks, routes, }) {
     const filteredRoutes = filterIntermediateRoutes(routes);
-    return lodash_1.default.mapValues(collectedLinks, (pageCollectedData, pagePath) => getBrokenLinksForPage({
-        collectedLinks,
-        pageLinks: pageCollectedData.links,
-        pageAnchors: pageCollectedData.anchors,
-        pagePath,
-        routes: filteredRoutes,
-    }));
+    return lodash_1.default.mapValues(collectedLinks, (pageCollectedData, pagePath) => {
+        try {
+            return getBrokenLinksForPage({
+                collectedLinks,
+                pageLinks: pageCollectedData.links,
+                pageAnchors: pageCollectedData.anchors,
+                pagePath,
+                routes: filteredRoutes,
+            });
+        }
+        catch (e) {
+            throw new Error(`Unable to get broken links for page ${pagePath}.`, {
+                cause: e,
+            });
+        }
+    });
 }
 function brokenLinkMessage(brokenLink) {
     const showResolvedLink = brokenLink.link !== brokenLink.resolvedLink;
@@ -179,11 +200,26 @@ function reportBrokenLinks({ brokenLinks, onBrokenLinks, onBrokenAnchors, }) {
         logger_1.default.report(onBrokenAnchors)(anchorErrorMessage);
     }
 }
+// Users might use the useBrokenLinks() API in weird unexpected ways
+// JS users might call "collectLink(undefined)" for example
+// TS users might call "collectAnchor('#hash')" with/without #
+// We clean/normalize the collected data to avoid obscure errors being thrown
+function normalizeCollectedLinks(collectedLinks) {
+    return lodash_1.default.mapValues(collectedLinks, (pageCollectedData) => ({
+        links: pageCollectedData.links.filter(lodash_1.default.isString),
+        anchors: pageCollectedData.anchors
+            .filter(lodash_1.default.isString)
+            .map((anchor) => (anchor.startsWith('#') ? anchor.slice(1) : anchor)),
+    }));
+}
 async function handleBrokenLinks({ collectedLinks, onBrokenLinks, onBrokenAnchors, routes, }) {
     if (onBrokenLinks === 'ignore' && onBrokenAnchors === 'ignore') {
         return;
     }
-    const brokenLinks = getBrokenLinks({ routes, collectedLinks });
+    const brokenLinks = getBrokenLinks({
+        routes,
+        collectedLinks: normalizeCollectedLinks(collectedLinks),
+    });
     reportBrokenLinks({ brokenLinks, onBrokenLinks, onBrokenAnchors });
 }
 exports.handleBrokenLinks = handleBrokenLinks;

package/lib/server/routes.js

@@ -159,6 +159,14 @@ ${JSON.stringify(routeConfig)}`);
         props,
     });
 }
+/**
+ * Old stuff
+ * As far as I understand, this is what permits to SSG the 404.html file
+ * This is rendered through the catch-all ComponentCreator("*") route
+ * Note CDNs only understand the 404.html file by convention
+ * The extension probably permits to avoid emitting "/404/index.html"
+ */
+const NotFoundRoutePath = '/404.html';
 /**
  * Routes are prepared into three temp files:
  *
@@ -175,7 +183,7 @@ function loadRoutes(routeConfigs, baseUrl, onDuplicateRoutes) {
         routesConfig: '',
         routesChunkNames: {},
         registry: {},
-        routesPaths: [(0, utils_1.normalizeUrl)([baseUrl, '404.html'])],
+        routesPaths: [(0, utils_1.normalizeUrl)([baseUrl, NotFoundRoutePath])],
     };
     // `genRouteCode` would mutate `res`
     const routeConfigSerialized = routeConfigs

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@docusaurus/core",
   "description": "Easy to Maintain Open Source Documentation Websites",
-  "version": "0.0.0-5797",
+  "version": "0.0.0-5801",
   "license": "MIT",
   "publishConfig": {
     "access": "public"
@@ -43,13 +43,13 @@
     "@babel/runtime": "^7.22.6",
     "@babel/runtime-corejs3": "^7.22.6",
     "@babel/traverse": "^7.22.8",
-    "@docusaurus/cssnano-preset": "0.0.0-5797",
-    "@docusaurus/logger": "0.0.0-5797",
-    "@docusaurus/mdx-loader": "0.0.0-5797",
+    "@docusaurus/cssnano-preset": "0.0.0-5801",
+    "@docusaurus/logger": "0.0.0-5801",
+    "@docusaurus/mdx-loader": "0.0.0-5801",
     "@docusaurus/react-loadable": "5.5.2",
-    "@docusaurus/utils": "0.0.0-5797",
-    "@docusaurus/utils-common": "0.0.0-5797",
-    "@docusaurus/utils-validation": "0.0.0-5797",
+    "@docusaurus/utils": "0.0.0-5801",
+    "@docusaurus/utils-common": "0.0.0-5801",
+    "@docusaurus/utils-validation": "0.0.0-5801",
     "@slorber/static-site-generator-webpack-plugin": "^4.0.7",
     "@svgr/webpack": "^6.5.1",
     "autoprefixer": "^10.4.14",
@@ -104,8 +104,8 @@
     "webpackbar": "^5.0.2"
   },
   "devDependencies": {
-    "@docusaurus/module-type-aliases": "0.0.0-5797",
-    "@docusaurus/types": "0.0.0-5797",
+    "@docusaurus/module-type-aliases": "0.0.0-5801",
+    "@docusaurus/types": "0.0.0-5801",
     "@types/detect-port": "^1.3.3",
     "@types/react-dom": "^18.2.7",
     "@types/react-router-config": "^5.0.7",
@@ -124,5 +124,5 @@
   "engines": {
     "node": ">=18.0"
   },
-  "gitHead": "b1487520689968837819e5b837e0acbf071b9d0e"
+  "gitHead": "44100c84778ec4735110e7a878269af287facaa7"
 }