Adapt reference extractor to modern IETF HTML-formatted RFCs (#676)

Some of the references contain link to email addresses, which needs to be ignored to hit the actual useful link
w3c · Jul 1, 2021 · ee87a7f · ee87a7f
1 parent 0bd7d82
commit ee87a7f
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/src/browserlib/extract-references.mjs b/src/browserlib/extract-references.mjs
@@ -96,7 +96,7 @@ function parseReferences(referenceList, options) {
     if (!desc || !ref.name) {
       return;
     }
-    ref.url = desc.querySelector("a[href]") ? desc.querySelector("a[href]").href : "";
+    ref.url = desc.querySelector('a[href^="http"]')?.href ?? "";
     if (options.filterInformative &&
         desc.textContent.match(/non-normative/i)) {
       return informativeRef.push(ref);
@@ -195,4 +195,4 @@ function extractReferences(rules) {
   });
 
   return references;
-}
+}