# ThaiDeka — Thai Supreme Court judgments
# Public site is crawlable. Auth-walled, generated, and JSON endpoints aren't.

User-agent: *
Allow: /

# Auth-walled or user-specific
Disallow: /admin/
Disallow: /dashboard
Disallow: /bookmarks
Disallow: /account/
Disallow: /subscription
Disallow: /ask
Disallow: /ask/history
Disallow: /saved-searches/
Disallow: /highlights/
Disallow: /keyword-alerts/

# Generated / parameter-heavy URLs that waste crawl budget.
# Results pages don't add canonical content beyond what /deka already covers.
Disallow: /search
Disallow: /*?q=*
Disallow: /*?categories[]=
Disallow: /*?year_from=
# Allow pagination so crawlers can discover deep content beyond Page 1.
# We handle duplicate content via canonical tags and noindex,follow on Page 2+.
Allow: /deka?page=
Allow: /deka/featured?page=
Allow: /deka/popular?page=
Allow: /deka/latest?page=
Allow: /articles?page=
Allow: /admincourt?page=
Allow: /krisdika?page=
Allow: /laws?page=
Allow: /categories/*?page=

# Storage + raw assets
Disallow: /storage/
Disallow: /*.json$

# Legacy WordPress URLs from the old thaideka.com site — gone for good.
# Caddy returns 410 for these too (Caddyfile.prod @legacy_wp), but adding
# them here lets well-behaved crawlers drop them from their indexes faster.
Disallow: /tag/
Disallow: /category/
Disallow: /author/
Disallow: /page/
Disallow: /feed
Disallow: /comments/
Disallow: /wp-admin/
Disallow: /wp-json/
Disallow: /wp-content/
Disallow: /wp-includes/
Disallow: /wp-login.php
Disallow: /wp-cron.php
Disallow: /xmlrpc.php
# Legacy deka permalinks (/deka/<no>-<year>.html) 301-redirect to the
# canonical /deka/<year>-<no> in Caddy. Allow them through so Googlebot
# follows the 301 and consolidates link equity into the new URL —
# disallowing here would prevent crawls and the old URLs would stay
# stuck in the index pointing at nothing useful. The blanket
# `Disallow: /*.html$` below catches the rest of the legacy WP HTML.
Allow: /deka/*.html$
Disallow: /*.html$
Disallow: /deka-diagnosis/
Disallow: /deka-request/
Disallow: /deka-noflesh/
Disallow: /deka-headnote/

# Bots that hammer hard enough to threaten site availability. They all
# claim to honor robots.txt — these blanket disallows take effect after
# their next robots.txt fetch (cached up to 24h). Caddy @abusive_bots
# 429s them in the meantime so the site stays up regardless.
User-agent: Amazonbot
Disallow: /

User-agent: Bytespider
Disallow: /

User-agent: ClaudeBot
Disallow: /

User-agent: PetalBot
Disallow: /

User-agent: AhrefsBot
Disallow: /

User-agent: SemrushBot
Disallow: /

User-agent: MJ12bot
Disallow: /

User-agent: DotBot
Disallow: /

User-agent: SeekBot
Disallow: /

# Sitemap index — points at sitemaps/static.xml, laws-{1..N}.xml,
# articles-{1..N}.xml, and dekas-{1..N}.xml. Every sub-file is capped
# at 25,000 URLs and 45 MB, well under Google's 50K / 50 MB limits.
Sitemap: https://deka.in.th/sitemap.xml