fix(patreon): handle the /cw/ creator-URL prefix in vanity extraction

A source URL like https://www.patreon.com/cw/Atole resolved vanity='cw' — the
vanity regex only skipped a /c/ prefix, so Patreon's current /cw/ ("creator
workspace") form fell through to the bare-vanity branch and captured the prefix
instead of the slug. Every /cw/ source then failed campaign-id resolution
(API + page-scrape both looked up "cw"). Operator-confirmed 2026-06-07 via the
new error text: source_url='.../cw/Atole'; vanity='cw'.

Add cw/ to the optional prefix group (ordered before c/ so the longer prefix
wins), and have the creator-page fallback try the /cw/ form too. Test covers
bare / c/ / cw/ extraction and that id: URLs stay non-vanity.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

backend/app/services/patreon_resolver.py

@@ -30,9 +30,17 @@ _CAMPAIGNS_URL = "https://www.patreon.com/api/campaigns"
 # A source URL of the form `.../id:<digits>` already carries the campaign id
 # (no lookup needed). The vanity regex deliberately EXCLUDES the id: form so the
 # two paths don't overlap.
+#
+# Patreon serves the creator vanity under several path prefixes — bare
+# (`patreon.com/Atole`), `c/` (`patreon.com/c/Atole`), and `cw/`
+# (`patreon.com/cw/Atole`, its current "creator workspace" URL). The optional
+# prefix group must list `cw/` BEFORE `c/` so the longer prefix wins — otherwise
+# `cw/Atole` matches the bare branch and yields vanity="cw" (operator-flagged
+# 2026-06-07: every `/cw/` source failed resolution on vanity="cw").
 _ID_URL_RE = re.compile(r"/id:(\d+)")
+_VANITY_PREFIXES = ("cw/", "c/")
 _VANITY_RE = re.compile(
-    r"^https?://(?:www\.)?patreon\.com/(?:c/)?(?!id:)([^/?#]+)",
+    r"^https?://(?:www\.)?patreon\.com/(?:cw/|c/)?(?!id:)([^/?#]+)",
     re.IGNORECASE,
 )
 _USER_AGENT = (
@@ -141,10 +149,12 @@ def _lookup_via_page(vanity: str, cookies_path: str | None) -> str | None:
     Patreon redirects between. Never raises."""
     jar = _load_cookie_jar(cookies_path)
     headers = {"User-Agent": _USER_AGENT, "Accept": "text/html"}
-    for page_url in (
+    # Try the bare vanity and every known creator-path prefix Patreon
-        f"https://www.patreon.com/{vanity}",
+    # redirects between (c/, cw/) — the one the source used isn't known here
-        f"https://www.patreon.com/c/{vanity}",
+    # (extract_vanity already stripped it).
-    ):
+    page_urls = [f"https://www.patreon.com/{vanity}"]
+    page_urls += [f"https://www.patreon.com/{p}{vanity}" for p in _VANITY_PREFIXES]
+    for page_url in page_urls:
         try:
             resp = requests.get(
                 page_url,

tests/test_patreon_resolver.py

@@ -169,3 +169,17 @@ async def test_for_source_unresolvable_returns_none():
     cid, resolved = await resolve_campaign_id_for_source("not-a-patreon-url", None, {})
     assert cid is None
     assert resolved is None
+
+
+def test_extract_vanity_handles_all_path_prefixes():
+    """Patreon serves the vanity bare and under c/ and cw/ prefixes; all must
+    yield the creator slug, not the prefix (operator-flagged 2026-06-07: a
+    /cw/Atole source resolved vanity='cw')."""
+    from backend.app.services.patreon_resolver import extract_vanity
+
+    assert extract_vanity("https://www.patreon.com/Atole") == "Atole"
+    assert extract_vanity("https://www.patreon.com/c/Atole") == "Atole"
+    assert extract_vanity("https://www.patreon.com/cw/Atole") == "Atole"
+    assert extract_vanity("https://patreon.com/cw/Atole") == "Atole"
+    # id: URLs are NOT vanities (handled by the id path).
+    assert extract_vanity("https://www.patreon.com/id:123") is None