name: caribbeancom
sceneByURL:
  - action: scrapeXPath
    url:
      - en.caribbeancom.com/eng/moviepages
      - en.caribbeancompr.com/eng/moviepages
    scraper: sceneScraper
sceneByFragment:
  action: scrapeXPath
  queryURL: https://en.caribbeancom{filename}
  # constructs the movie URL from the filename, provided that the filename includes the movie id
  queryURLReplace:
    filename:
      - regex: .*(\d{6}-\d{3}).* # carribeancom uses ids with form like 062212-055
        with: .com/eng/moviepages/$1
      - regex: .*(\d{6}_\d{3}).* # caribbeancompr uses 062212_055
        with: pr.com/eng/moviepages/$1
      - regex: $
        with: /index.html
  scraper: sceneScraper

xPathScrapers:
  sceneScraper:
    common:
      $movieinfo: //div[@class="movie-info section divider"]
    scene:
      Title: //div[contains(@class,"heading")]/h1/text()
      Details: $movieinfo/p
      URL:
        selector: //link[@hreflang="ja-JP"]/@href
        postProcess:
          - replace:
              - regex: "https://www.caribbeancom.com"
                with: "https://en.caribbeancom.com/eng"
      Date:
        selector: //ul/li/span[contains(.,"Release Date")]/../span[@class="spec-content"]
        postProcess:
          - replace:
              - regex: (\d{4}).(\d{2}).(\d{2})
                with: $1-$2-$3
          - parseDate: 2006-01-02
      Performers:
        Name:
          selector: //ul/li/span[contains(.,"Starring")]/..//a
          concat: "," # caribbeancom splits name/surname for some performers
          postProcess:
            - replace:
                - regex: "&" # in some cases & is used instead of , to split performers
                  with: ","
                - regex: "([^,]),([^,])"
                  with: "$1 $2"
                - regex: ",,"
                  with: ","
          split: ","
      Tags:
        Name: //ul/li/span[contains(.,"Tags")]/../span/a[contains(@class,"spec")]
      Image:
        selector: //link[@hreflang="ja-JP"]/@href|//script[contains(.,"posterImage = '/moviepages/'+movie_id+'/images/")]
        postProcess:
          - replace:
              - regex: index\.html$
                with: images/l.jpg
              - regex: .*posterImage\s*=\s*\'/moviepages/\'\+movie_id\+\'/images/([^\']+)\'(.|\s)*MoviePlayer\.setImage..movie_id\s.*:\s\'(\d+_\d+)\'.*
                with: https://en.caribbeancompr.com/moviepages/$3/images/$1
      Studio:
        Name:
          selector: //ul[@class="footer-copyright"]/li[contains(.,"©")]
          postProcess:
            - replace:
                - regex: .*\s([\w]+)\.com.*
                  with: $1
            - map:
                caribbeancompr: Caribbeancom Premium
                caribbeancom: Caribbeancom
# Last Updated May 20, 2021