# STACK Infrastructure — campus pages (campus + building records) + news/press feed
id: stack
name: STACK Infrastructure
domain: www.stackinfra.com
kind: operator
mode: hybrid
priority: 1
license: "Publicly accessible operator pages; factual data only; © STACK Infrastructure"
attribution: "STACK Infrastructure"
homepage: https://www.stackinfra.com/locations/
notes: >
  Open WordPress site. Yoast sitemaps: campuses-sitemap.xml (/locations/<region>/<metro>/<code>/), location-sitemap.xml
  (metro indexes), news-sitemap.xml; RSS at /feed/. Campus pages publish campus MW + acreage and a card per building
  (code, SQ FT, MW). No street address or coordinates. The parser emits one campus record plus one record per building
  (campusName = STACK <code>). Japanese pages under /japan/ are excluded (duplicates of KIX01/TKY01).
  robots.txt has no disallows.
  maxLevel is capped at 2 (direct HTTP only): every stackinfra.com page embeds Cloudflare's "/cdn-cgi/challenge-platform/"
  beacon, which the shared looksBlockedOrEmpty() heuristic misreads as a bot wall and would escalate each of the ~50
  campus pages to Scrapfly (30 credits each) although direct fetches return the full page (HTTP 200, verified).
coverage: { global: true, operators: ["STACK Infrastructure"] }
fetch:
  level: 1
  maxLevel: 2
  rpm: 20
  concurrency: 2
  respectRobots: true
  maxCreditsPerRun: 40
discovery:
  sitemap: ["/campuses-sitemap.xml", "/news-sitemap.xml"]
  rss: ["/feed/"]
  include: ["/locations/", "/about/news-press/"]
  exclude: ["\\?", "/japan/", "/take-a-tour", "/about/news-press/?$"]
  classify:
    - { pattern: "/locations/(americas|emea|asia-pacific)/[a-z-]+/[a-z0-9-]+/?$", group: facility_pages, pageType: facility_page, priority: 60 }
    - { pattern: "/locations/(americas|emea|asia-pacific)/[a-z-]+/?$", group: index, pageType: facility_index, priority: 20 }
    - { pattern: "/locations/?$", group: index, pageType: facility_index, priority: 10 }
    - { pattern: "/about/news-press/(press-releases|news)/[a-z0-9._%-]+/?$", group: newsroom, pageType: press_release, priority: 70 }
  maxUrlsPerRun: 1000
schedule:
  facility_pages: weekly
  newsroom: daily
  index: monthly
  sitemap: weekly
  rss: daily
extractors:
  campus:
    # selected by URL only (classifyPage() may mislabel deep facility URLs; the URL is authoritative)
    match: ["/locations/(americas|emea|asia-pacific)/[a-z-]+/[a-z0-9-]+/?$"]
    kind: facility
    parser: stack_campus_v1
defaults:
  operatorName: STACK Infrastructure
  operatorKind: developer
  facilityType: wholesale
  isHyperscale: true
parserVersion: v1