# TerraNova — allow crawling of public content (SEO), block the data API and
# paid/expensive endpoints. Honest crawlers obey this; edge/rate-limit rules
# handle the rest.
User-agent: *
Allow: /
Disallow: /api/
Disallow: /dossier/download
Disallow: /dossier/checkout
Disallow: /report/download
Disallow: /listing/download
Disallow: /listing/checkout
# The listing report itself is generated from one visitor's pasted price and pin.
# Nothing to index, and a crawler walking the parameters would be pure load.
Disallow: /listing/report
# /reports itself is indexable and wants to be — it is the page that explains the
# paid documents. Only the sample generator is blocked: each hit renders a full PDF
# through dompdf, which is the most expensive thing this app does.
Disallow: /reports/sample/
Disallow: /go/
# Stripe's webhook receiver. Nothing to index, and a crawler hitting it just
# generates rejected-signature noise in the logs.
Disallow: /stripe/
# Honeypot — see routes/web.php. Nothing here is real; requesting it flags the
# caller as a bot that ignores this file.
Disallow: /data/

# /sitemap.xml is a sitemap INDEX (51k+ municipality URLs is past Google's
# 50k-per-file limit) — it points at /sitemap-pages.xml and the
# /sitemap-communes-N.xml chunks, which need no line of their own here.
#
# ⚠ This directive is IGNORED unless the URL is fully qualified, and this file is
# static (nginx serves it directly, so it cannot be rendered from APP_URL). Replace
# the line below with the real host at deploy time — it is on the go-live checklist
# in docs/DEPLOYMENT.md §13.2. Submitting the sitemap in Search Console covers Google
# regardless; this line is what everyone else discovers it by.
Sitemap: /sitemap.xml