1  #!/usr/bin/env node
  2  /**
  3   * Scrape podcast episodes from RSS feed → one markdown file per episode.
  4   *
  5   * Usage:
  6   *   node scrape-rss.mjs <rss-url>                # all episodes from RSS
  7   *   node scrape-rss.mjs <rss-url> --limit 5      # first 5 episodes only
  8   *   node scrape-rss.mjs <rss-url> --dry-run      # preview without writing files
  9   *   node scrape-rss.mjs <rss-url> --out ./eps     # custom output directory
 10   */
 11  
 12  import { parseArgs } from 'node:util';
 13  import { writeFileSync, mkdirSync } from 'node:fs';
 14  import { join } from 'node:path';
 15  
 16  const { values: flags, positionals } = parseArgs({
 17    options: {
 18      limit: { type: 'string', short: 'l' },
 19      'dry-run': { type: 'boolean', short: 'd' },
 20      out: { type: 'string', short: 'o' },
 21    },
 22    allowPositionals: true,
 23  });
 24  
 25  const RSS_URL = positionals[0];
 26  if (!RSS_URL) {
 27    console.error('Usage: node scrape-rss.mjs <rss-url> [--limit N] [--dry-run] [--out dir]');
 28    process.exit(1);
 29  }
 30  
 31  const EPISODES_DIR = flags.out || join(process.cwd(), 'episodes');
 32  
 33  // --- Helpers ---
 34  
 35  function sanitizeFilename(title) {
 36    return title
 37      .replace(/[<>:"/\\|?*]/g, '-')
 38      .replace(/\s+/g, ' ')
 39      .replace(/\.+$/, '')
 40      .trim();
 41  }
 42  
 43  function formatDuration(seconds) {
 44    const h = Math.floor(seconds / 3600);
 45    const m = Math.floor((seconds % 3600) / 60);
 46    const s = seconds % 60;
 47    return h > 0
 48      ? `${h}:${String(m).padStart(2, '0')}:${String(s).padStart(2, '0')}`
 49      : `${m}:${String(s).padStart(2, '0')}`;
 50  }
 51  
 52  function extractText(xml, tag) {
 53    const re = new RegExp(`<${tag}[^>]*>(?:<!\\[CDATA\\[)?([\\s\\S]*?)(?:\\]\\]>)?</${tag}>`, 'i');
 54    const m = xml.match(re);
 55    return m ? m[1].trim() : '';
 56  }
 57  
 58  function extractAttr(xml, tag, attr) {
 59    const re = new RegExp(`<${tag}[^>]*${attr}="([^"]*)"`, 'i');
 60    const m = xml.match(re);
 61    return m ? m[1] : '';
 62  }
 63  
 64  function decodeEntities(text) {
 65    return text
 66      .replace(/&amp;/g, '&')
 67      .replace(/&lt;/g, '<')
 68      .replace(/&gt;/g, '>')
 69      .replace(/&quot;/g, '"')
 70      .replace(/&#39;/g, "'")
 71      .replace(/&apos;/g, "'")
 72      .replace(/&nbsp;/g, ' ')
 73      .replace(/&#\d+;/g, m => String.fromCharCode(parseInt(m.slice(2, -1))));
 74  }
 75  
 76  function stripHtml(html) {
 77    return decodeEntities(
 78      html
 79        .replace(/<br\s*\/?>/gi, '\n')
 80        .replace(/<\/p>/gi, '\n\n')
 81        .replace(/<[^>]+>/g, '')
 82    )
 83      .replace(/\n{3,}/g, '\n\n')
 84      .trim();
 85  }
 86  
 87  function parseDate(dateStr) {
 88    const d = new Date(dateStr);
 89    return d.toISOString().split('T')[0];
 90  }
 91  
 92  // --- Main ---
 93  
 94  async function main() {
 95    console.log(`Fetching RSS feed: ${RSS_URL}`);
 96    const res = await fetch(RSS_URL);
 97    if (!res.ok) throw new Error(`HTTP ${res.status}`);
 98    const xml = await res.text();
 99  
100    // Extract channel metadata
101    const channelTitle = decodeEntities(extractText(xml, 'title'));
102    const channelAuthor = decodeEntities(extractText(xml, 'itunes:author'));
103    console.log(`Podcast: ${channelTitle} by ${channelAuthor}`);
104  
105    // Split into items
106    const items = xml.split('<item>').slice(1).map(chunk => '<item>' + chunk.split('</item>')[0] + '</item>');
107  
108    const limit = flags.limit ? parseInt(flags.limit) : items.length;
109    const episodes = items.slice(0, limit);
110  
111    console.log(`Found ${items.length} episodes in feed, processing ${episodes.length}`);
112  
113    if (!flags['dry-run']) {
114      mkdirSync(EPISODES_DIR, { recursive: true });
115    }
116  
117    for (const item of episodes) {
118      const title = decodeEntities(extractText(item, 'title'));
119      const description = extractText(item, 'description');
120      const summary = extractText(item, 'itunes:summary');
121      const pubDate = extractText(item, 'pubDate');
122      const duration = parseInt(extractText(item, 'itunes:duration')) || 0;
123      const season = extractText(item, 'itunes:season');
124      const episode = extractText(item, 'itunes:episode');
125      const audioUrl = extractAttr(item, 'enclosure', 'url');
126  
127      const date = parseDate(pubDate);
128      const descClean = stripHtml(description || summary);
129  
130      const meta = [`- **Date**: ${date}`];
131      if (season || episode) {
132        const parts = [];
133        if (season) parts.push(`**Season**: ${season}`);
134        if (episode) parts.push(`**Episode**: ${episode}`);
135        meta.push(`- ${parts.join(' | ')}`);
136      }
137      meta.push(`- **Duration**: ${formatDuration(duration)}`);
138      meta.push(`- **Audio**: ${audioUrl}`);
139  
140      const md = `# ${title}
141  
142  ## Metadata
143  ${meta.join('\n')}
144  
145  ## Description
146  
147  ${descClean}
148  
149  ## Transcript
150  
151  _Not yet transcribed._
152  `;
153  
154      const filename = sanitizeFilename(title) + '.md';
155  
156      if (flags['dry-run']) {
157        console.log(`\n--- ${filename} ---`);
158        console.log(md.slice(0, 300) + '...');
159      } else {
160        const filepath = join(EPISODES_DIR, filename);
161        writeFileSync(filepath, md);
162        console.log(`  ✓ ${filename}`);
163      }
164    }
165  
166    console.log(`\nDone. ${episodes.length} episodes processed.`);
167  }
168  
169  main().catch(err => { console.error(err); process.exit(1); });