feat: add Markdown export for conference and workshop pages

scripts/markdown_test.mjs

@@ -0,0 +1,84 @@
+#!/usr/bin/env node
+/**
+ * Unit tests for site/src/lib/markdown.ts formatting logic.
+ * Run: node scripts/markdown_test.mjs
+ */
+import { formatWorkshop, formatConferenceYear, formatSingleWorkshopInfo, isAutoTopicsNote } from '../site/src/lib/markdown.ts';
+
+let failed = 0;
+function check(label, got, expect) {
+  const ok = typeof expect === 'boolean' ? got === expect : (got.includes ? got.includes(expect) : got === expect);
+  if (!ok) {
+    failed++;
+    console.error(`✗ ${label}\n  Got: ${JSON.stringify(got)}\n  Expected to match: ${JSON.stringify(expect)}`);
+  } else {
+    console.log(`✓ ${label}`);
+  }
+}
+
+// 1. Test auto topics note filtering
+check('isAutoTopicsNote identifies exact auto-suggested note',
+  isAutoTopicsNote('Topics were auto-suggested and may be imprecise — edits welcome.'), true);
+check('isAutoTopicsNote identifies historical auto note',
+  isAutoTopicsNote('Auto-imported from the OpenReview venue record on 2026-06-20 — please verify and enrich (topics are keyword-guessed).'), true);
+check('isAutoTopicsNote identifies SEED DATA note',
+  isAutoTopicsNote('SEED DATA: unverified entry'), true);
+check('isAutoTopicsNote allows legitimate custom description',
+  isAutoTopicsNote('This workshop focuses on efficient LLM reasoning.'), false);
+
+// 2. Test formatWorkshop with abstract deadline, bold key-values, and filtered notes
+const sampleWs = {
+  name: 'Efficient Reasoning Workshop',
+  statusLabel: 'Open call',
+  abstractDeadlineWallClock: 'Aug 20, 2026 23:59 UTC',
+  abstractDeadlinePassed: false,
+  deadlineWallClock: 'Aug 29, 2026 23:59 UTC',
+  notificationDateLabel: 'Sep 20, 2026',
+  workshopDateLabel: 'Oct 25, 2026',
+  topics: ['llms', 'efficiency'],
+  website: 'https://example.com/ws',
+  openreview_venue_id: 'NeurIPS.cc/2026/Workshop/ER',
+  notes: 'Topics were auto-suggested and may be imprecise — edits welcome.',
+  deadlineChange: {
+    kind: 'extended',
+    days: 5,
+    fromWallClock: 'Aug 24, 2026 23:59 UTC',
+  },
+};
+
+const output = formatWorkshop(sampleWs, 'NeurIPS');
+
+check('formatWorkshop includes H2 header', output, '## Efficient Reasoning Workshop');
+check('formatWorkshop formats bold Status label', output, '- **Status:** Open call');
+check('formatWorkshop includes abstract deadline', output, '- **Abstract Deadline:** Aug 20, 2026 23:59 UTC');
+check('formatWorkshop formats bold Submission Deadline', output, '- **Submission Deadline:** Aug 29, 2026 23:59 UTC');
+check('formatWorkshop formats bold Website link', output, '- **Website:** [https://example.com/ws](https://example.com/ws)');
+check('formatWorkshop formats bold OpenReview link', output, '- **OpenReview:** [https://openreview.net/group?id=NeurIPS.cc/2026/Workshop/ER](https://openreview.net/group?id=NeurIPS.cc/2026/Workshop/ER)');
+check('formatWorkshop formats Deadline History', output, '**Deadline History:** Extended by 5 days (previously Aug 24, 2026 23:59 UTC)');
+check('formatWorkshop excludes auto-suggested maintenance notes from Description', output.includes('Description'), false);
+
+// 3. Test custom notes inclusion
+const wsWithCustomNotes = {
+  ...sampleWs,
+  notes: 'We welcome papers on efficient attention mechanisms.',
+};
+const outputCustom = formatWorkshop(wsWithCustomNotes, 'NeurIPS');
+check('formatWorkshop includes valid custom description', outputCustom, '**Description:**\nWe welcome papers on efficient attention mechanisms.');
+
+// 4. Test formatConferenceYear
+const confObj = { name: 'NeurIPS', full_name: 'Neural Information Processing Systems' };
+const confOutput = formatConferenceYear(confObj, 2026, [sampleWs]);
+check('formatConferenceYear contains conference header', confOutput, '# Neural Information Processing Systems 2026 Workshops');
+check('formatConferenceYear contains Total Workshops count', confOutput, 'Total Workshops: 1');
+
+// 5. Test formatSingleWorkshopInfo
+const singleOutput = formatSingleWorkshopInfo(sampleWs, confObj);
+check('formatSingleWorkshopInfo contains single workshop header', singleOutput, '# Efficient Reasoning Workshop');
+check('formatSingleWorkshopInfo contains Conference field', singleOutput, 'Conference: Neural Information Processing Systems');
+
+if (failed > 0) {
+  console.error(`\nTest suite failed with ${failed} failure(s).`);
+  process.exit(1);
+} else {
+  console.log('\nAll markdown tests passed successfully!');
+}

site/package.json

@@ -5,7 +5,7 @@
   "type": "module",
   "scripts": {
     "dev": "astro dev",
-    "build": "astro build && pagefind --site dist --root-selector '[data-pf-ws]' && pagefind --site dist --output-subdir pagefind-papers --root-selector '[data-pf-papers]'",
+    "build": "astro build && pagefind --site dist --root-selector \"[data-pf-ws]\" && pagefind --site dist --output-subdir pagefind-papers --root-selector \"[data-pf-papers]\"",
     "preview": "astro preview"
   },
   "dependencies": {

site/src/components/MarkdownExport.astro

@@ -0,0 +1,139 @@
+---
+export interface Props {
+  content?: string;
+  exportUrl?: string;
+  filename?: string;
+  variant?: 'button' | 'link';
+}
+
+const { content, exportUrl, filename, variant = 'button' } = Astro.props;
+const encodedContent = content ? encodeURIComponent(content) : undefined;
+---
+
+{variant === 'link' ? (
+  <button type="button" class="btn-copy as-link markdown-export-btn" data-md={encodedContent} data-url={exportUrl}>
+    <span class="btn-text">📋 Copy as Markdown</span>
+  </button>
+) : (
+  <div class="markdown-export">
+    <button type="button" class="btn-copy markdown-export-btn" data-md={encodedContent} data-url={exportUrl}>
+      <span class="btn-text">📋 Copy as Markdown</span>
+    </button>
+
+    {exportUrl && filename ? (
+      <a href={exportUrl} download={filename} class="btn-download markdown-export-btn">
+        ⬇ Download .md
+      </a>
+    ) : filename && (
+      <button type="button" class="btn-download markdown-export-btn" data-filename={filename} data-md={encodedContent}>
+        ⬇ Download .md
+      </button>
+    )}
+  </div>
+)}
+
+<script>
+  document.addEventListener('click', async (e) => {
+    const target = e.target as HTMLElement | null;
+    if (!target) return;
+
+    const copyBtn = target.closest('.btn-copy');
+    if (copyBtn) {
+      try {
+        let md = '';
+        const url = copyBtn.getAttribute('data-url');
+        const inlineData = copyBtn.getAttribute('data-md');
+
+        if (url) {
+          const res = await fetch(url);
+          md = await res.text();
+        } else if (inlineData) {
+          md = decodeURIComponent(inlineData);
+        }
+
+        if (!md) return;
+
+        await navigator.clipboard.writeText(md);
+
+        const textSpan = copyBtn.querySelector('.btn-text') || copyBtn;
+        const originalText = textSpan.textContent;
+        textSpan.textContent = '✓ Copied!';
+        (textSpan as HTMLElement).style.color = 'var(--accent, #10b981)';
+        setTimeout(() => {
+          textSpan.textContent = originalText;
+          (textSpan as HTMLElement).style.color = '';
+        }, 2000);
+      } catch (err) {
+        console.error('Failed to copy: ', err);
+      }
+      return;
+    }
+
+    const dlBtn = target.closest('button.btn-download');
+    if (dlBtn) {
+      const md = decodeURIComponent(dlBtn.getAttribute('data-md') || '');
+      const filename = dlBtn.getAttribute('data-filename') || 'export.md';
+
+      const blob = new Blob([md], { type: 'text/markdown;charset=utf-8' });
+      const url = window.URL.createObjectURL(blob);
+
+      const a = document.createElement('a');
+      a.href = url;
+      a.download = filename;
+      a.style.display = 'none';
+      document.body.appendChild(a);
+      a.click();
+
+      setTimeout(() => {
+        document.body.removeChild(a);
+        window.URL.revokeObjectURL(url);
+      }, 1000);
+    }
+  });
+</script>
+
+<style>
+  .markdown-export {
+    display: flex;
+    gap: 0.5rem;
+    margin: 1rem 0;
+    flex-wrap: wrap;
+  }
+  .markdown-export button,
+  .markdown-export a.btn-download {
+    display: inline-flex;
+    align-items: center;
+    padding: 0.4rem 0.8rem;
+    font-size: 0.9rem;
+    font-family: var(--font-body, inherit);
+    color: var(--ink, inherit);
+    background: var(--surface, transparent);
+    border: 1px solid var(--line, currentColor);
+    border-radius: 4px;
+    cursor: pointer;
+    text-decoration: none;
+    transition: border-color 0.2s, background-color 0.2s, color 0.2s;
+  }
+  .markdown-export button:hover,
+  .markdown-export a.btn-download:hover {
+    border-color: var(--accent);
+    background: var(--accent-soft);
+  }
+
+  /* Link variant styles */
+  button.as-link {
+    background: none;
+    border: none;
+    padding: 0;
+    font-size: inherit;
+    font-family: inherit;
+    color: var(--accent);
+    text-decoration: underline;
+    text-underline-offset: 2px;
+    cursor: pointer;
+    transition: opacity 0.2s;
+  }
+  button.as-link:hover {
+    opacity: 0.8;
+  }
+</style>

site/src/lib/markdown.ts

@@ -0,0 +1,132 @@
+import { topicById } from './data.ts';
+
+export const AUTO_TOPICS_NOTE = 'Topics were auto-suggested and may be imprecise — edits welcome.';
+
+/**
+ * True if `notes` is the bot's auto-suggested-topics note or a seed note,
+ * so we drop it from the exported Description.
+ */
+export function isAutoTopicsNote(notes?: string | null): boolean {
+  if (!notes || typeof notes !== 'string') return false;
+  const trimmed = notes.trim();
+  if (trimmed === AUTO_TOPICS_NOTE) return true;
+  if (trimmed.includes('auto-suggested and may be imprecise')) return true;
+  if (trimmed.includes('SEED DATA')) return true;
+  if (/^Auto-imported from the OpenReview venue record on \d{4}-\d{2}-\d{2} — please verify and enrich \(topics are keyword-guessed\)\.?$/.test(trimmed)) return true;
+  return false;
+}
+
+export function formatWorkshop(w: any, confName?: string): string {
+  const parts: string[] = [];
+
+  parts.push(`## ${w.name}`);
+
+  const details: string[] = [];
+
+  if (w.statusLabel) {
+    details.push(`- **Status:** ${w.statusLabel}`);
+  }
+
+  if (w.abstractDeadlineWallClock) {
+    const passedTag = w.abstractDeadlinePassed ? ' (closed)' : '';
+    details.push(`- **Abstract Deadline:** ${w.abstractDeadlineWallClock}${passedTag}`);
+  }
+
+  if (w.deadlineWallClock) {
+    details.push(`- **Submission Deadline:** ${w.deadlineWallClock}`);
+  } else {
+    details.push(`- **Submission Deadline:** TBA`);
+  }
+
+  if (w.notificationDateLabel) {
+    details.push(`- **Notification:** ${w.notificationDateLabel}`);
+  }
+
+  if (w.workshopDateLabel) {
+    details.push(`- **Workshop Date:** ${w.workshopDateLabel}`);
+  }
+
+  if (w.website) {
+    details.push(`- **Website:** [${w.website}](${w.website})`);
+  }
+
+  if (w.openreview_venue_id) {
+    const url = `https://openreview.net/group?id=${w.openreview_venue_id}`;
+    details.push(`- **OpenReview:** [${url}](${url})`);
+  }
+
+  if (details.length > 0) {
+    parts.push(details.join('\n'));
+  }
+
+  if (w.topics && w.topics.length > 0) {
+    const topicLabels = w.topics.map((t: string) => `- ${topicById.get(t)?.label ?? t}`).join('\n');
+    parts.push(`**Topics:**\n${topicLabels}`);
+  }
+
+  if (w.deadlineChange) {
+    if (w.deadlineChange.kind === 'extended') {
+      parts.push(`**Deadline History:** Extended by ${w.deadlineChange.days} ${w.deadlineChange.days === 1 ? 'day' : 'days'} (previously ${w.deadlineChange.fromWallClock})`);
+    } else if (w.deadlineChange.kind === 'earlier') {
+      parts.push(`**Deadline History:** Moved ${w.deadlineChange.days} ${w.deadlineChange.days === 1 ? 'day' : 'days'} earlier (previously ${w.deadlineChange.fromWallClock})`);
+    } else if (w.deadlineChange.kind === 'announced') {
+      parts.push(`**Deadline History:** Deadline just announced`);
+    }
+  } else if (Array.isArray(w.deadlineHistoryView) && w.deadlineHistoryView.length > 1) {
+    const historyLines = w.deadlineHistoryView.map((h: any, i: number) => {
+      const tag = i === w.deadlineHistoryView.length - 1 ? 'first recorded' : 'changed';
+      return `- ${h.recordedLabel}: ${h.wallClock ?? 'no date published'} (${tag})`;
+    }).join('\n');
+    parts.push(`**Deadline History:**\n${historyLines}`);
+  }
+
+  if (w.notes && !isAutoTopicsNote(w.notes)) {
+    parts.push(`**Description:**\n${w.notes}`);
+  }
+
+  return parts.join('\n\n');
+}
+
+export function formatConferenceYear(conf: any, year: number, wsList: any[]): string {
+  const confFull = conf.full_name || conf.name;
+  const genDate = new Date().toISOString().split('T')[0];
+
+  const header = [
+    `# ${confFull} ${year} Workshops`,
+    `Generated from AI Workshop Tracker`,
+    ``,
+    `Conference: ${conf.name}`,
+    `Edition: ${year}`,
+    `Generated: ${genDate}`,
+    `Total Workshops: ${wsList.length}`,
+    ``,
+    `---`
+  ].join('\n');
+
+  const body = wsList.map(w => formatWorkshop(w, conf.name)).join('\n\n---\n\n');
+
+  return `${header}\n\n${body}`;
+}
+
+export function formatSingleWorkshopInfo(w: any, conf: any): string {
+  const confFull = conf.full_name || conf.name;
+  const genDate = new Date().toISOString().split('T')[0];
+
+  const header = [
+    `# ${w.name}`,
+    `Generated from AI Workshop Tracker`,
+    ``,
+    `Conference: ${confFull}`,
+    `Edition: ${w.year}`,
+    `Generated: ${genDate}`,
+    ``,
+    `---`
+  ].join('\n');
+
+  // Strip the '## ' line from formatWorkshop output since header already has '# w.name'
+  const bodyParts = formatWorkshop(w, conf.name).split('\n\n');
+  bodyParts.shift();
+  const body = bodyParts.join('\n\n');
+
+  return `${header}\n\n${body}`;
+}

site/src/pages/conference/[conf].astro

@@ -2,6 +2,8 @@
 import Base from '../../components/Base.astro';
 import { workshops, conferences } from '../../lib/data';
 import { href, REPO_URL } from '../../lib/site';
+import MarkdownExport from '../../components/MarkdownExport.astro';
+import { formatConferenceYear } from '../../lib/markdown';
 
 // One hub page per conference, generated from the same data the rest of the
 // site uses. Intentionally carries NO data-pf-ws / data-pf-papers attributes,
@@ -144,6 +146,7 @@ const statusLines = (w: any): string[] =>
   {yearGroups.map(([year, ws]) => (
     <section class="conf-year">
       <h2>{confName} {year}</h2>
+      <MarkdownExport content={formatConferenceYear(conf, year, ws)} filename={`${conf.id}-${year}-workshops.md`} />
       <ul class="conf-ws-list">
         {ws.map((w) => (
           <li>