Geo::StreetAddress::US - Perl extension for parsing US street addresses
use Geo::StreetAddress::US; my $hashref = Geo::StreetAddress::US->parse_location( "1005 Gravenstein Hwy N, Sebastopol CA 95472" ); my $hashref = Geo::StreetAddress::US->parse_location( "Hollywood & Vine, Los Angeles, CA" ); my $hashref = Geo::StreetAddress::US->parse_address( "1600 Pennsylvania Ave, Washington, DC" ); my $hashref = Geo::StreetAddress::US->parse_intersection( "Mission Street at Valencia Street, San Francisco, CA" ); my $normal = Geo::StreetAddress::US->normalize_address( \%spec ); # the parse_* methods call this automatically...
Geo::StreetAddress::US is a regex-based street address and street intersection parser for the United States. Its basic goal is to be as forgiving as possible when parsing user-provided address strings. Geo::StreetAddress::US knows about directional prefixes and suffixes, fractional building numbers, building units, grid-based addresses (such as those used in parts of Utah), 5 and 9 digit ZIP codes, and all of the official USPS abbreviations for street types and state names.
Most Geo::StreetAddress::US methods return a reference to a hash containing address or intersection information as one of their arguments. This "address specifier" hash may contain any of the following fields for a given address. If a given field is not present in the address, the corresponding key will be set to
undef in the hash.
House or street number.
Directional prefix for the street, such as N, NE, E, etc. A given prefix should be one to two characters long.
Name of the street, without directional or type qualifiers.
Abbreviated street type, e.g. Rd, St, Ave, etc. See the USPS official type abbreviations at http://www.usps.com/ncsc/lookups/abbr_suffix.txt for a list of abbreviations used.
Directional suffix for the street, as above.
Name of the city, town, or other locale that the address is situated in.
The state which the address is situated in, given as its two-letter postal abbreviation. See http://www.usps.com/ncsc/lookups/abbr_state.txt for a list of abbreviations used.
Five digit ZIP postal code for the address, including leading zero, if needed.
Directional prefixes for the streets in question.
Names of the streets in question.
Street types for the streets in question.
Directional suffixes for the streets in question.
City or locale containing the intersection, as above.
State abbreviation, as above.
Five digit ZIP code, as above.
Geo::StreetAddress::US contains a number of global variables which it uses to recognize different bits of US street addresses. Although you will probably not need them, they are documented here for completeness's sake.
Maps directional names (north, northeast, etc.) to abbreviations (N, NE, etc.).
Maps directional abbreviations to directional names.
Maps lowercased USPS standard street types to their canonical postal abbreviations as found in TIGER/Line. See eg/get_street_abbrev.pl in the distrbution for how this map was generated.
Maps lowercased US state and territory names to their canonical two-letter postal abbreviations. See eg/get_state_abbrev.pl in the distrbution for how this map was generated.
Maps two-digit FIPS-55 US state and territory codes (including the leading zero!) as found in TIGER/Line to the state's canonical two-letter postal abbreviation. See eg/get_state_fips.pl in the distrbution for how this map was generated. Yes, I know the FIPS data also has the state names. Oops.
A hash of compiled regular expressions corresponding to different types of address or address portions. Defined regexen include type, number, fraction, state, direct(ion), dircode, zip, corner, street, place, address, and intersection.
Parses any address or intersection string and returns the appropriate specifier, by calling parse_intersection() or parse_address() as needed.
Parses a street address into an address specifier, returning undef if the address cannot be parsed. You probably want to use parse_location() instead.
Parses an intersection string into an intersection specifier, returning undef if the address cannot be parsed. You probably want to use parse_location() instead.
Takes an address or intersection specifier, and normalizes its components, stripping out all leading and trailing whitespace and punctuation, and substituting official abbreviations for prefix, suffix, type, and state values. Also, city names that are prefixed with a directional abbreviation (e.g. N, NE, etc.) have the abbreviation expanded. The normalized specifier is returned.
Typically, you won't need to use this method, as the
parse_*() methods call it for you.
normalize_address() crops 9-digit ZIP codes to 5 digits. This is for the benefit of Geo::Coder::US and may not be what you want. E-mail me if this is a problem and I'll see what I can do to fix it.
Geo::StreetAddress::US might not correctly parse house numbers that contain hyphens, such as those used in parts of Queens, New York. Also, some addresses in rural Michigan and Illinois may contain letter prefixes to the building number that may cause problems. Fixing these edge cases is on the to-do list, to be sure. Patches welcome!
This software was originally part of Geo::Coder::US (q.v.) but was split apart into an independent module for your convenience. Therefore it has some behaviors which were designed for Geo::Coder::US, but which may not be right for your purposes. If this turns out to be the case, please let me know.
Geo::StreetAddress::US does NOT perform USPS-certified address normalization.
This software was originally part of Geo::Coder::US(3pm).
Lingua::EN::AddressParse(3pm) and Geo::PostalAddress(3pm) both do something very similar to Geo::StreetAddress::US, but are either too strict/limited in their address parsing, or not really specific enough in how they break down addresses (for my purposes). If you want USPS-style address standardization, try Scrape::USPS::ZipLookup(3pm). Be aware, however, that it scrapes a form on the USPS website in a way that may not be officially permitted and might break at any time. If this module does not do what you want, you might give the othersa try. All three modules are available from the CPAN.
Many thanks to Dave Rolsky for submitting a very useful patch to fix fractional house numbers, dotted directionals, and other kinds of edge cases, e.g. South St. He even submitted additional tests!
Schuyler D. Erle <email@example.com>
Copyright (C) 2005 by Schuyler D. Erle.
This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.8.4 or, at your option, any later version of Perl 5 you may have available.