From 6e05484307a2b8bff5b6ef32f7fa3e5ad1144475 Mon Sep 17 00:00:00 2001
From: Harry Rose <harryrose@protonmail.com>
Date: Mon, 16 Mar 2026 20:59:00 +0000
Subject: [PATCH] Handle podcasts which use html lists and also have html tags
 in the chapter titles

---
 .../parsers/parsePodcastDescriptionForChapters.js  | 10 +++++++---
 .../parsePodcastDescriptionForChapters.test.js     | 14 +++++++++++++-
 2 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/server/utils/parsers/parsePodcastDescriptionForChapters.js b/server/utils/parsers/parsePodcastDescriptionForChapters.js
index 3f1b4c75..8a83f149 100644
--- a/server/utils/parsers/parsePodcastDescriptionForChapters.js
+++ b/server/utils/parsers/parsePodcastDescriptionForChapters.js
@@ -1,3 +1,4 @@
+const sanitizeHtml = require('../../libs/sanitizeHtml')
 const Logger = require('../../Logger')
 
 /**
@@ -27,13 +28,16 @@ module.exports.parse = (podcastDescription, audioDurationSecs) => {
 
   const timestampRegex = /\b(\d{1,2}):(\d{1,2})(?::(\d{1,2}))?\b/
   const chapterTitleRegex = /\b\d{1,2}:\d{1,2}(?::\d{1,2})?\b(?:\s+|\))(.+)$/
-  const descriptionLineSplitRegex = /\<\s*\/\s*p\s*\>|\<\s*br\s*\s*\/\>|\n/
+
+  // Split on "</p>", "<br />", "\n", </li>
+  const descriptionLineSplitRegex = /\<\s*\/\s*p\s*\>|\<\s*br\s*\/\>|\n|\<\s*\/\s*li\s*\>/
 
   var descriptionLines = podcastDescription.split(descriptionLineSplitRegex)
   var newChapters = []
 
   for (let i = 0; i < descriptionLines.length; i++) {
-    let line = descriptionLines[i]
+    // Strip all HTML tags out
+    let line = sanitizeHtml(descriptionLines[i], { allowedTags: [] })
 
     let match = timestampRegex.exec(line)
     if (match == null) continue
@@ -85,7 +89,7 @@ module.exports.parse = (podcastDescription, audioDurationSecs) => {
     newChapters[newChapters.length - 1].end = audioDurationSecs
   }
 
-  Logger.info(`[PodcastEpisode] Successfully generated ${newChapters.length} chapters`)
+  Logger.info(`Successfully generated ${newChapters.length} chapters`)
 
   if (newChapters.length == 1) {
     throw new Error('Only one chapter found, treating as invalid description')
diff --git a/test/server/utils/parsers/parsePodcastDescriptionForChapters.test.js b/test/server/utils/parsers/parsePodcastDescriptionForChapters.test.js
index 0fff4574..662c131d 100644
--- a/test/server/utils/parsers/parsePodcastDescriptionForChapters.test.js
+++ b/test/server/utils/parsers/parsePodcastDescriptionForChapters.test.js
@@ -1,5 +1,4 @@
 const chai = require('chai')
-const PodcastEpisode = require('../../../../server/models/PodcastEpisode')
 const expect = chai.expect
 const parsePodcastDescriptionForChapters = require('../../../../server/utils/parsers/parsePodcastDescriptionForChapters')
 
@@ -69,6 +68,19 @@ describe('parsePodcastDescriptionForChapters', () => {
         { title: 'Chapter 7', id: 7, start: 2492, end: 2803 },
         { title: 'Chapter 8', id: 8, start: 2803, end: 3060 }
       ]
+    },
+    {
+      testName: 'Should handle html lists and chapters with html tags in the title',
+      description: '<p>Introduction</p><p><br /></p><p><br /></p>Chapters<ul><li><strong>00:00:00</strong> Intro</li><li><strong>00:03:55</strong> Chapter 1</li><li><strong>00:09:52</strong> Chapter 2    </li><li><strong>00:16:11</strong> Chapter 3</li><li><strong>00:20:03</strong> Chapter 4</li><li><strong>00:24:08</strong> Chapter 5</li>',
+      audioDuration: 4000,
+      expectedChapters: [
+        { title: 'Intro', id: 1, start: 0, end: 235 },
+        { title: 'Chapter 1', id: 2, start: 235, end: 592 },
+        { title: 'Chapter 2', id: 3, start: 592, end: 971 },
+        { title: 'Chapter 3', id: 4, start: 971, end: 1203 },
+        { title: 'Chapter 4', id: 5, start: 1203, end: 1448 },
+        { title: 'Chapter 5', id: 6, start: 1448, end: 4000 }
+      ]
     }
   ]
   testCasesTestingSuccess.forEach(function (testCase) {