reffy 16.0.0 → 16.0.2

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "reffy",
-  "version": "16.0.0",
+  "version": "16.0.2",
   "description": "W3C/WHATWG spec dependencies exploration companion. Features a short set of tools to study spec references as well as WebIDL term definitions and references found in W3C specifications.",
   "repository": {
     "type": "git",
@@ -36,14 +36,14 @@
     "ajv-formats": "3.0.1",
     "commander": "12.1.0",
     "fetch-filecache-for-crawling": "5.1.1",
-    "puppeteer": "22.10.0",
+    "puppeteer": "22.11.2",
     "semver": "^7.3.5",
     "web-specs": "3.9.1",
     "webidl2": "24.4.1"
   },
   "devDependencies": {
     "mocha": "10.4.0",
-    "respec": "35.1.0",
+    "respec": "35.1.1",
     "respec-hljs": "2.1.1",
     "rollup": "4.18.0",
     "undici": "^6.1.0"

package/src/browserlib/extract-links.mjs

@@ -28,7 +28,8 @@ export default function () {
     if (n.closest('.head, del')) return;
     const pageUrl = n.href.split('#')[0];
     // links generated by authoring tools have data-link-type or data-xref-type set
-    let linkSet = n.dataset.linkType || n.dataset.xrefType ? autolinks : rawlinks;
+    // Bikeshed also adds automatic untyped links in the generatedindex ("ul.index aside")
+    let linkSet = n.dataset.linkType || n.dataset.xrefType || n.closest("ul.index aside") ? autolinks : rawlinks;
     if (!linkSet[pageUrl]) {
       linkSet[pageUrl] = {anchors: new Set()};
     }

package/src/lib/specs-crawler.js

@@ -84,8 +84,15 @@ async function crawlSpec(spec, crawlOptions) {
     try {
         const fallback = crawlOptions.fallbackData?.results?.find(s => s.url === spec.url);
         let cacheInfo = {};
-        if (crawlOptions.fallbackData?.crawler === `reffy-${reffyVersion}`) {
-          cacheInfo = Object.assign({}, fallback?.crawlCacheInfo);
+        if (fallback && !fallback.error &&
+                crawlOptions.fallbackData?.crawler === `reffy-${reffyVersion}`) {
+            // Note: we don't want to reuse the previous crawl results if
+            // there was an error because we don't really know whether these
+            // results come from that previous crawl (in which case we should
+            // crawl the spec again), or from a an earlier crawl where
+            // everything went fine (in which case we could reuse the results
+            // if the spec wasn't updated in the meantime).
+            cacheInfo = Object.assign({}, fallback.crawlCacheInfo);
         }
         let result = null;
         if (crawlOptions.useCrawl) {
@@ -348,6 +355,11 @@ async function crawlList(speclist, crawlOptions) {
     const crawlQueue = new ThrottledQueue({
         maxParallel: 4,
         sleepInterval: origin => {
+            if (crawlOptions.useCrawl) {
+                // Not an actual crawl, we're going to reuse previous crawl
+                // results instead. No need to sleep!
+                return 0;
+            }
             switch (origin) {
             case 'https://csswg.org': return 2000;
             case 'https://www.w3.org': return 1000;