Handle article:author meta tag. Fixes #938 (#942)

* Handle article:author meta tag. Fixes #938 * Add newly found BBC byline, revert apparently unnecessarily regex change. --------- Co-authored-by: Gijs Kruitbosch <[email protected]>
mozilla · Jan 2, 2025 · b6ff1b6 · b6ff1b6
1 parent d9c588a
commit b6ff1b6
Show file tree

Hide file tree

Showing 5 changed files with 2,538 additions and 2 deletions.
diff --git a/Readability.js b/Readability.js
@@ -432,6 +432,20 @@ Readability.prototype = {
     }
   },
 
+  /**
+   * Tests whether a string is a URL or not.
+   *
+   * @param {string} str The string to test
+   * @return {boolean} true if str is a URL, false if not
+   */
+  _isUrl(str) {
+    try {
+      new URL(str);
+      return true;
+    } catch {
+      return false;
+    }
+  },
   /**
    * Converts each <a> and <img> uri in the given element to an absolute URI,
    * ignoring #ref URIs.
@@ -1771,13 +1785,20 @@ Readability.prototype = {
       metadata.title = this._getArticleTitle();
     }
 
+    const articleAuthor =
+      typeof values["article:author"] === "string" &&
+      !this._isUrl(values["article:author"])
+        ? values["article:author"]
+        : undefined;
+
     // get author
     metadata.byline =
       jsonld.byline ||
       values["dc:creator"] ||
       values["dcterm:creator"] ||
       values.author ||
-      values["parsely-author"];
+      values["parsely-author"] ||
+      articleAuthor;
 
     // get description
     metadata.excerpt =

diff --git a/test/test-pages/article-author-tag/expected-metadata.json b/test/test-pages/article-author-tag/expected-metadata.json
@@ -0,0 +1,10 @@
+{
+  "title": "If You Can Picture A Tarot Card, It's Because of These 3 People",
+  "byline": "Laura June Topolsky",
+  "dir": null,
+  "lang": "en",
+  "excerpt": "Picture a deck of tarot cards. What do you see? Maybe the Magician in his rich red robes, right arm raised high above him. Or the skeleton on horseback for...",
+  "siteName": "Atlas Obscura",
+  "publishedTime": "2015-07-10 09:53:00 -0400",
+  "readerable": true
+}