link title to proposal

convert_workshops_to_md.py

@@ -10,6 +10,7 @@ from pathlib import Path
 
 
 TITLE_RE = re.compile(r'<div id="block-pagetitle".*?<h1 class="page-title"><span[^>]*>(.*?)</span>.*?</h1>', re.DOTALL)
+CANONICAL_RE = re.compile(r'<link rel="canonical" href="([^"]+)"\s*/?>', re.DOTALL)
 SESSION_RE = re.compile(
     r'<details[^>]*id="edit-group-session"[^>]*>.*?<div class="details-wrapper">(.*?)</div>\s*</details>',
     re.DOTALL,
@@ -133,14 +134,22 @@ def extract_title(source: str) -> str:
     return normalize_text(title_match.group(1))
 
 
+def extract_canonical_url(source: str) -> str:
+    canonical_match = CANONICAL_RE.search(source)
+    if not canonical_match:
+        raise ValueError('Could not find canonical URL for proposal page')
+    return html.unescape(canonical_match.group(1)).strip()
+
+
 def extract_markdown(source: str) -> tuple[str, str]:
     title = extract_title(source)
+    canonical_url = extract_canonical_url(source)
     session_match = SESSION_RE.search(source)
     if not session_match:
         raise ValueError('Could not find session details block by id edit-group-session')
 
     details_html = session_match.group(1)
-    sections: list[str] = [f'# {title}']
+    sections: list[str] = [f'# [{title}]({canonical_url})']
 
     for classes, body in extract_field_blocks(details_html):
         label_match = LABEL_RE.search(body)