Handle all cases of email protection to prevent blinking and ease mai…

…ntenance (#959)

CHANGELOG.md

@@ -5,7 +5,13 @@ All notable changes to this project will be documented in this file.
 This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html) and the format is based on [Common Changelog](https://common-changelog.org).\
 Unlike Common Changelog, the `unreleased` section of [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) is preserved with the addition of a tag to specify which type of release should be published and to foster discussions about it inside pull requests. This tag should be one of the names mandated by SemVer, within brackets: `[patch]`, `[minor]` or `[major]`. For example: `## Unreleased [minor]`.
 
-## Unreleased
+## Unreleased [minor]
+
+Full changeset and discussions: [#959](https://github.com/ambanum/OpenTermsArchive/pull/959)._
+
+### Changed
+
+- Improved reliability and expanded coverage of email protection global filter
 
 ## 0.19.1 - 2022-12-13
 ### Fixed

src/archivist/filter/index.js

@@ -84,9 +84,12 @@ export async function filterHTML({ content, pageDeclaration }) {
 
   // clean code from common changing patterns - initially for Windstream
   domFragment.querySelectorAll('a[href*="/email-protection"]').forEach(node => {
-    if (node.href.match(/((.*?)\/email-protection#)[0-9a-fA-F]+/gim)) {
-      node.href = `${node.href.split('#')[0]}#removed`;
-    }
+    const newProtectedLink = webPageDOM.createElement('a');
+    const [href] = node.href.split('#');
+
+    newProtectedLink.href = href;
+    newProtectedLink.innerHTML = '[email protected]';
+    node.parentNode.replaceChild(newProtectedLink, node);
   });
 
   const markdownContent = transform(domFragment);

src/archivist/filter/index.test.js

@@ -63,6 +63,7 @@ const rawHTMLWithCommonChangingItems = `
     <p><a id="link3" href="http://absolute.url/link">link 3</a></p>
     <p><a id="link4" href="">link 4</a></p>
     <a href="/cdn-cgi/l/email-protection#3b4c52555f484f495e5a56154b49524d5a584215484f5a4f5e565e554f7b4c52555f484f495e5a5615585456">[email&#160;protected]</a>
+    <p><a href="/cdn-cgi/l/email-protection#2d4e4243594c4e596d4e4459545e4e424259034858">conta<span>[email&#160;protected]</span></a></p>
   </body>
 </html>`;
 
@@ -78,7 +79,9 @@ const expectedFilteredWithCommonChangingItems = `Title
 
 link 4
 
-[\\[email protected\\]](https://exemple.com/cdn-cgi/l/email-protection#removed)`;
+[\\[email protected\\]](https://exemple.com/cdn-cgi/l/email-protection)
+
+[\\[email protected\\]](https://exemple.com/cdn-cgi/l/email-protection)`;
 /* eslint-enable no-irregular-whitespace */
 
 const additionalFilter = {