XConfessions.yml

name: XConfessions
sceneByURL:
  - action: scrapeXPath
    url:
      - xconfessions.com
    scraper: sceneScraper
movieByURL:
  - action: scrapeXPath
    url:
      - xconfessions.com
    scraper: movieScraper
xPathScrapers:
  sceneScraper:
    common: &com
      $script: //script[contains(.,'window.__NUXT')]/text()
    scene:
      Title: &title //h1
      Date: &date
        selector: $script
        postProcess:
          - replace:
              - regex: .*release_date."([^\s]+).*
                with: $1
          - parseDate: 2006-01-02
      Details: &details
        selector: $script
        postProcess:
          - replace:
              - regex: .+synopsis_clean="(?P<full>[^"]+).+short_synopsis_clean="(?P<short>[^"]+).+
                with: |
                  $short

                  $full
      Performers:
        Name:
          selector: //p[span[contains(.,'Performers:')]]//a
          concat: ' & '  # some performers are joined
          split: ' & '  # with &
      Director: &director //p[span[contains(.,'Director:')]]//a
      Tags:
        Name: 
          selector: //div/a[contains(@href,'/categories/')] | //button/@data-test
          postProcess:
            - replace:
                - regex: 'profile-'
                  with: ''
      Image: &image
        selector: $script
        postProcess: 
          - replace:
                    # Most scenes have a cover_title_picture value in the script
              - regex: .+?cover_title_picture="([^"]+).+
                with: $1
                    # Some without the cover_title_picture will have twi consecutive images
                    # with a UID.  The first is the movie poster, the second is the scene cover image
              - regex: .+,"(https:[^"]+?-[^"]+?)","(https:[^"]+?-[^"]+)".*
                with: $2
                    # Some without the UID-style images instead have a sequence of images after
                    # a time stamp.  The first is th movie poster, the second is the scene image
              - regex: .+"\d\d\d\d-\d\d-\d\d\s+\d\d:\d\d:\d\d"[0-9a-zA-Z,"]*"(https:[^"]+?)","(https:[^"]+)".*
                with: $2
                    # Some don't have any of the above, but have a confession_cover_image value
                    # that is pretty good, but doesn't have the scene name on it.
              - regex: .+?confession_cover_image:"([^"]+).+
                with: $1
                    # If nothing above matches, take the last image in the script.  Often this
                    # is the scene image, but it's not 100% reliable.
              - regex: .+(https:[^"]+).*$
                with: $1
                    # Clean up UTF-8 coded slashes for real ones
              - regex: '\\u002F'
                with: '/'
      Studio: &studio
        Name:
          fixed: XConfessions
      Movies:
        Name: *title
        Director: *director
        Duration: &duration
          selector: $script
          postProcess:
            - replace:
                - regex: .+?.length="([^"]+).+
                  with: $1
        Date: *date
        Synopsis: *details
        Studio: *studio
        URL: &url //link[@rel='canonical']/@href
        FrontImage: &frontImage //article//div[contains(@class,'134px')]//img[contains(@sizes,'200px') or contains(@src,'jpeg') or contains(@src,'jpg')]/@src

  movieScraper:
    common: *com
    movie:
      Name: *title
      Director: *director 
      Duration: *duration
      Date: *date
      Synopsis: *details
      Studio: *studio
      URL: *url
      FrontImage: *frontImage
# Last Updated April 13, 2024