rules.yaml

# Address Parsing Rules 
address_parts:

  # Unfiltered set of address parts.  Currently maps directly to usaddress parts.
  standard:
    - id: &address_number address_number
      usaddress: AddressNumber
      description: 'The address number, e.g. "1234" Main St.'
  
    - id: &address_number_prefix address_number_prefix
      usaddress: AddressNumberPrefix
      description: 'A modifier before an address number, e.g. "Mile", "#"'
  
    - id: &address_number_suffix address_number_suffix
      usaddress: AddressNumberSuffix
      description: 'A modifier after an address number, e.g "B", "1/2"'
  
    - id: &building_name building_name
      usaddress: BuildingName
      description: 'The name of a building, e.g. "Atlanta Financial Center"'
  
    - id: &corner_of corner_of
      usaddress: CornerOf
      description: 'Words indicating that an address is a corner, e.g. "Junction", "corner of"'
  
    - id: &intersection_separator intersection_separator
      usaddress: IntersectionSeparator
      description: 'A conjunction connecting parts of an intersection, e.g. "and", "&"'
  
    - id: &landmark_name landmark_name
      usaddress: LandmarkName
      description: 'The name of a landmark, e.g. "Wrigley Field", "Union Station"'
  
    - id: &not_address not_address
      usaddress: NotAddress
      description: "A non-address component that doesn't refer to a recipient"
  
    - id: &occupancy_type occupancy_type
      usaddress: OccupancyType
      description: 'A type of occupancy within a building, e.g. "Suite", "Apt", "Floor"'
  
    - id: &occupancy_id occupancy_id
      usaddress: OccupancyIdentifier
      description: 'The identifier of an occupancy, often a number or letter'
  
    - id: &city_name city_name
      usaddress: PlaceName
      description: 'The name of the city'
  
    - id: &recipient recipient
      usaddress: Recipient
      description: 'A non-address recipient, e.g. the name of a person/organization'
  
    - id: &state_name state_name
      usaddress: StateName
      description: 'The name of the state'
 
    - id: &street_name street_name
      usaddress: StreetName
      description: 'The street name, excluding type & direction'
  
    - id: &street_name_pre_directional street_name_pre_directional
      usaddress: StreetNamePreDirectional
      description: 'A direction before a street name, e.g. "North", "S"'
  
    - id: &street_name_pre_modifier street_name_pre_modifier
      usaddress: StreetNamePreModifier
      description: 'A modifier before a street name that is not a direction, e.g. "Old"'
  
    - id: &street_name_pre_type street_name_pre_type
      usaddress: StreetNamePreType
      description: 'A street type that comes before a street name, e.g. "Route", "Ave"'
  
    - id: &street_name_post_directional street_name_post_directional
      usaddress: StreetNamePostDirectional
      description: 'A direction after a street name, e.g. "North", "S"'
  
    - id: &street_name_post_modifier street_name_post_modifier
      usaddress: StreetNamePostModifier
      description: 'A modifier adter a street name, e.g. "Ext"'
  
    - id: &street_name_post_type street_name_post_type
      usaddress: StreetNamePostType
      description: 'A street type that comes after a street name, e.g. "Avenue", "Rd"'
  
    - id: &subaddress_id subaddress_id
      usaddress: SubaddressIdentifier
      description: 'The name/identifier of a subaddress component'
  
    - id: &subaddress_type subaddress_type
      usaddress: SubaddressType
      description: 'A level of detail in an address that is not an occupancy within a building, e.g. "Building", "Tower"'
  
    - id: &usps_box_group_id usps_box_group_id
      usaddress: USPSBoxGroupID
      description: 'The identifier of a USPS box group, usually a number'
  
    - id: &usps_box_group_type usps_box_group_type
      usaddress: USPSBoxGroupType
      description: 'A name for a group of USPS boxes, e.g. "R"'
  
    - id: &postal_box_id postal_box_id
      usaddress: USPSBoxID
      description: 'The identifier of a USPS box, usually a number'
  
    - id: &usps_box_type usps_box_type
      usaddress: USPSBoxType
      description: 'A USPS box, e.g. "P.O. Box"'
  
    - id: &zip_code zip_code
      usaddress: ZipCode
      description: 'The zip code'

  # Alternate address parts made up of multiple "standard" parts
  derived:
    - id: &address_number_full address_number_full
      description: 'All address number parts combined, e.g. "123 1/2", "12 B"'
      parts:
        - *address_number_prefix
        - *address_number
        - *address_number_suffix

    - id: &street_name_full street_name_full
      description: 'All address name parts combined, e.g. "Pennsylvania Ave NW"'
      parts:
        - *street_name_pre_directional
        - *street_name_pre_modifier
        - *street_name_pre_type
        - *street_name
        - *street_name_post_type
        - *street_name_post_directional
        - *street_name_post_modifier
  
# Alternate parsing rulesets
profiles:
  - id: grasshopper
    required:
      - *address_number_full
      - *street_name_full
      - *state_name
      - *zip_code
    # TODO: "optional" may be useful if we start excluding all other parts
    # optional:
    #  - *city_name
    # TODO: Include "invalid" address parts for added validation
    # invalid:
    #  - *corner_of
    #  - *intersection_indicator