Plaintext RSS fixer upper

src/lib/rss-transform.ts

@@ -0,0 +1,84 @@
+/**
+ * Starting around episode 223, the RSS feed changed from HTML to plain text.
+ * This transformer converts the new plain-text format to match the old HTML structure exactly.
+ */
+export function transformPlainTextToHtml(text: string): string {
+  const lines = text.split('\n').map(l => l.trim()).filter(Boolean);
+  const html: string[] = [];
+  let i = 0;
+
+  while (i < lines.length) {
+    const line = lines[i];
+
+    // Check if this is a timestamp line like "(00:00) - Intro"
+    if (/^\(\d{2}:\d{2}(?::\d{2})?\)\s*-/.test(line)) {
+      // Start collecting all consecutive timestamp lines into a list
+      const listItems: string[] = [];
+      while (i < lines.length && /^\(\d{2}:\d{2}(?::\d{2})?\)\s*-/.test(lines[i])) {
+        listItems.push(escapeHtml(lines[i]));
+        i++;
+      }
+      html.push('<ul>');
+      listItems.forEach(item => html.push(`<li>${item}</li>`));
+      html.push('</ul>');
+      continue;
+    }
+
+    // Check if this is a section header (bold text like "**Links**" or just "Links")
+    if (/^\*\*(.+?)\*\*$/.test(line)) {
+      const text = line.replace(/^\*\*(.+?)\*\*$/, '$1');
+      html.push(`<p><strong>${escapeHtml(text)}</strong></p>`);
+      i++;
+      continue;
+    }
+
+    // Check if this looks like a link list item (e.g., "CodeRabbit: https://...")
+    if (/:?\s*https?:\/\//.test(line)) {
+      // Collect all consecutive link lines into a list
+      const linkItems: string[] = [];
+      while (i < lines.length && /:?\s*https?:\/\//.test(lines[i])) {
+        linkItems.push(lines[i]);
+        i++;
+      }
+      html.push('<ul>');
+      linkItems.forEach(item => {
+        // Parse "Label: URL" or just "URL"
+        const match = item.match(/^(.+?):\s*(https?:\/\/.+)$/);
+        if (match) {
+          const label = escapeHtml(match[1].trim());
+          const url = escapeHtml(match[2].trim());
+          html.push(`<li>${label}: <a href="${url}">${url}</a></li>`);
+        } else {
+          const urlMatch = item.match(/(https?:\/\/.+)/);
+          if (urlMatch) {
+            const url = escapeHtml(urlMatch[1].trim());
+            html.push(`<li><a href="${url}">${url}</a></li>`);
+          } else {
+            html.push(`<li>${escapeHtml(item)}</li>`);
+          }
+        }
+      });
+      html.push('</ul>');
+      continue;
+    }
+
+    // Default: regular paragraph
+    html.push(`<p>${escapeHtml(line)}</p>`);
+    i++;
+  }
+
+  return html.join('\n');
+}
+
+export function escapeHtml(str: string): string {
+  return str
+    .replace(/&/g, '&amp;')
+    .replace(/</g, '&lt;')
+    .replace(/>/g, '&gt;')
+    .replace(/\"/g, '&quot;')
+    .replace(/'/g, '&#039;');
+}
+
+export function looksLikeHtml(text: string): boolean {
+  return /<[a-z][\s\S]*>/i.test(text.trim());
+}

src/lib/rss.ts

@@ -3,10 +3,16 @@ import parseFeed from 'rss-to-json';
 import { array, number, object, optional, parse, string } from 'valibot';
 
 import { optimizeImage } from './optimize-episode-image';
+import {
+  escapeHtml,
+  looksLikeHtml,
+  transformPlainTextToHtml
+} from './rss-transform';
 import { dasherize } from '../utils/dasherize';
 import { truncate } from '../utils/truncate';
 import starpodConfig from '../../starpod.config';
 
+
 export interface Show {
   title: string;
   description: string;
@@ -102,7 +108,9 @@ export async function getAllEpisodes() {
           return {
             id,
             title: `${title}`,
-            content: description,
+            content: looksLikeHtml(description)
+              ? description
+              : transformPlainTextToHtml(description),
             description: truncate(htmlToText(description), 260),
             duration: itunes_duration,
             episodeImage: itunes_image?.href,
@@ -121,4 +129,4 @@ export async function getAllEpisodes() {
 
   episodesCache = episodes;
   return episodes;
-}
+}

tests/unit/rss.test.ts

@@ -0,0 +1,82 @@
+import { describe, expect, it } from 'vitest';
+import {
+  escapeHtml,
+  looksLikeHtml,
+  transformPlainTextToHtml
+} from '../../src/lib/rss-transform';
+
+describe('RSS Transformation Functions', () => {
+  describe('escapeHtml', () => {
+    it('escapes multiple special characters', () => {
+      expect(escapeHtml('<script>alert("XSS & stuff")</script>')).toBe(
+        '&lt;script&gt;alert(&quot;XSS &amp; stuff&quot;)&lt;/script&gt;'
+      );
+    });
+
+    it('returns unchanged string without special characters', () => {
+      expect(escapeHtml('Hello World')).toBe('Hello World');
+    });
+  });
+
+  describe('looksLikeHtml', () => {
+    it('detects HTML tags vs plain text', () => {
+      expect(looksLikeHtml('<p>Hello</p>')).toBe(true);
+      expect(looksLikeHtml('Just plain text')).toBe(false);
+      expect(looksLikeHtml('5 < 10 and 10 > 5')).toBe(false);
+    });
+  });
+
+  describe('transformPlainTextToHtml', () => {
+    it('handles mixed content with lists, headers, and paragraphs', () => {
+      const input = `**Episode Summary**
+
+This episode covers many topics & details.
+
+**Timestamps**
+
+(00:00) - Introduction
+(05:30) - Main discussion
+(00:00:15) - With seconds
+
+**Links**
+
+GitHub: https://github.com/example
+https://example.com
+Company & Co: https://example.com?foo=bar&baz=qux
+
+Thanks for listening!`;
+
+      const output = transformPlainTextToHtml(input);
+
+      expect(output).toContain('<p><strong>Episode Summary</strong></p>');
+      expect(output).toContain('<p>This episode covers many topics &amp; details.</p>');
+      expect(output).toContain('<li>(00:00) - Introduction</li>');
+      expect(output).toContain('<li>(00:00:15) - With seconds</li>');
+      expect(output.match(/<ul>/g)).toHaveLength(2);
+      expect(output).toContain(
+        '<li>GitHub: <a href="https://github.com/example">https://github.com/example</a></li>'
+      );
+      expect(output).toContain(
+        '<li>Company &amp; Co: <a href="https://example.com?foo=bar&amp;baz=qux">https://example.com?foo=bar&amp;baz=qux</a></li>'
+      );
+      expect(output).toContain('<p>Thanks for listening!</p>');
+    });
+
+    it('separates non-consecutive timestamp groups', () => {
+      const input = `(00:00) - Intro
+(05:00) - Part 1
+
+Some text in between
+
+(10:00) - Part 2
+(15:00) - Part 3`;
+      const output = transformPlainTextToHtml(input);
+      expect(output.match(/<ul>/g)).toHaveLength(2);
+    });
+
+    it('returns empty string for blank input', () => {
+      expect(transformPlainTextToHtml('')).toBe('');
+      expect(transformPlainTextToHtml('   \n  \n  ')).toBe('');
+    });
+  });
+});