Fix STY (South Tyneside) scraper

[symroe]

What broke

www.southtyneside.gov.uk/article/13598/councillors-a-to-z is protected by a Cloudflare managed challenge (cf-mitigated: challenge, server: cloudflare). wreq's Firefox TLS fingerprint receives HTTP 403 with a "Just a moment…" JS-challenge page rather than the councillors HTML. The scraper had no http_lib set so wreq was used directly — same pattern as EAT (#409), FEN (#399), and others recently fixed. A previous issue (#385, closed as completed) described the same problem and proposed this fix.

What was fixed

• councillors.py: added http_lib = "playwright" — headless Chromium executes the Cloudflare JS challenge, acquires cf_clearance, and then fetches both the list page and individual councillor detail pages. The existing custom HTML selectors (#COUNCILLORSLISTBYNAME_HTML, tbody td a) and the decode_cfemail helper are unchanged.

Scrape results

Metric	Count
Councillors found	TBC (Lambda run needed — Cloudflare blocks non-browser clients from this environment)
With email address	TBC
With photo	TBC

The councillors page serves valid HTML to browser clients (confirmed: 403 is the Cloudflare challenge, not a permanent block). Previous successful run (before the Cloudflare block was added) returned councillors with emails decoded via cf_email and photos from the info table.

---

Generated by Claude Code

[symroe]

Re-scrape after 952a771

Added http_lib = "playwright" to bypass Cloudflare managed challenge on www.southtyneside.gov.uk. Cloudflare returns HTTP 403 with a JS challenge to all non-browser clients from this environment, so exact counts require a Lambda run.

Metric	Count
Councillors found	TBC (Lambda run needed)
With email address	TBC
With photo	TBC

cf-mitigated: challenge confirmed from curl -sI — Playwright will execute the JS challenge and acquire cf_clearance before fetching the councillors list and detail pages.

---

Generated by Claude Code