Complete catalog standardization: generations, banners, and Fakemon roster tag.

Adds standardize-vault-pages pipeline, exports intentional empty generations, and refreshes the 396-hack catalog with full banner coverage, status classification, and Fakemon generation tags.

scripts/enrich_web.py

@@ -20,7 +20,50 @@ import enrich_vault as E
 
 FACTS_GLOB = os.path.join(E.SCRAPE, "web_facts*.json")
 SCALAR_FIELDS = ["tagline", "developer", "version", "status", "release_date",
-                 "base", "platform", "generation", "homepage"]
+                 "base", "platform", "generation", "homepage", "source", "fakemon",
+                 "banner"]
+LIST_FIELDS = ["generations"]
+
+
+def normalize_link_fact(link):
+    if isinstance(link, str):
+        return "Link", link
+    if isinstance(link, dict) and link.get("url"):
+        return link.get("label") or "Link", link["url"]
+    return None
+
+
+def merge_links_section(body, links):
+    normalized = [x for x in (normalize_link_fact(link) for link in links) if x]
+    if not normalized:
+        return body
+    existing = ""
+    match = re.search(r"(?ms)^## Links\s*\n(.*?)(?=^## |\n\[\[Index|\Z)", body)
+    if match:
+        existing = match.group(1).strip()
+    pairs = []
+    for line in existing.splitlines():
+        url_match = re.search(r"(https?://\S+)", line)
+        if not url_match:
+            continue
+        label_match = re.match(r"^\s*-\s*([^:]+):", line)
+        pairs.append(((label_match.group(1).strip() if label_match else "Link"), url_match.group(1).rstrip(".,)")))
+    pairs.extend(normalized)
+    seen = set()
+    lines = []
+    for label, url in pairs:
+        if url in seen:
+            continue
+        seen.add(url)
+        lines.append(f"- {label}: {url}")
+    return set_section(body, "Links", "\n".join(lines))
+
+
+def facts_sort_key(path: str) -> int:
+    match = re.search(r"web_facts(\d*)\.json$", os.path.basename(path))
+    if not match or not match.group(1):
+        return 0
+    return int(match.group(1))
 
 
 def set_section(body, heading, content):
@@ -57,6 +100,9 @@ def apply_one(stem, facts):
     for k in SCALAR_FIELDS:
         if facts.get(k):
             fm[k] = facts[k]
+    for k in LIST_FIELDS:
+        if k in facts:
+            fm[k] = facts[k]
     if facts.get("type"):
         fm["type"] = facts["type"]
     fm["tags"] = E.derive_tags(fm)
@@ -77,6 +123,8 @@ def apply_one(stem, facts):
         body = set_section(body, "Features", feats)
     if facts.get("notability"):
         body = set_section(body, "Why it stands out", facts["notability"].strip())
+    if facts.get("links"):
+        body = merge_links_section(body, facts["links"])
 
     return f"---\n{E.emit_fm(fm)}\n---\n{body}", "ok"
 
@@ -84,7 +132,7 @@ def apply_one(stem, facts):
 def main():
     apply = "--apply" in sys.argv
     facts = {}
-    for fp in sorted(glob.glob(FACTS_GLOB)):
+    for fp in sorted(glob.glob(FACTS_GLOB), key=facts_sort_key):
         facts.update(json.load(open(fp, encoding="utf-8")))
     ok = miss = 0
     for stem, f in facts.items():