package Badger::URL; use Badger::Class version => 0.02, debug => 0, base => 'Badger::Base', import => 'CLASS class', utils => 'textlike is_object', codec => 'uri', # we use URI encoding for parameters as_text => \&text, is_true => 1, # not sure about this constants => 'HASH BLANK', constant => { SLASH => '/', TEXT => 0, SCHEME => 1, AUTHORITY => 2, USER => 3, HOST => 4, PORT => 5, PATH => 6, QUERY => 7, FRAGMENT => 8, PARAMS => 9, }, alias => { url => \&text, }, exports => { any => 'URL', }; #------------------------------------------------------------------------ # Example URL: # # scheme authority path query fragment # __ ___________________ _________ _________ __ # / \ / \/ \ / \ / \ # http://user@example.com:8042/over/there?name=ferret#nose # \__/ \_________/ \__/ # user host port # #------------------------------------------------------------------------ our @ELEMENTS = qw( scheme authority user host port path query fragment params ); our $N_ELEMS = 1; # slot 0 holds source text, so slot 1 is first field our $ELEMENT = { map { $_ => $N_ELEMS++ } @ELEMENTS }; # regexen to match basic tokens our $MATCH_SCHEME = qr{ ( [a-zA-Z][a-zA-Z0-9.+\-]* ) : }x; our $MATCH_USER = qr{ ([^@]*) @ }x; our $MATCH_HOST = qr{ ( [^:\/]* ) }x; our $MATCH_PORT = qr{ : (\d*) }x; our $MATCH_PATH = qr{ ( [^ \? \#]* ) }x; our $MATCH_QUERY = qr{ \? ( [^ \#]* ) }x; our $MATCH_FRAGMENT = qr{ \# ( .* ) }x; # compound regexen to match authority our $MATCH_AUTHORITY = qr{ // ( # $1 - authority (?: $MATCH_USER )? # $2 - user $MATCH_HOST # $3 - host (?: $MATCH_PORT )? # $4 - port ) }x; # compound regexen to match complete URL our $MATCH_URL = qr{ ^ (?: $MATCH_SCHEME )? # $1 - scheme (?: $MATCH_AUTHORITY )? # $2,$3,$4,$5 - authority,user,host,port $MATCH_PATH # $6 - path (?: $MATCH_QUERY )? # $7 - query (?: $MATCH_FRAGMENT )? # $8 - fragment }x; #------------------------------------------------------------------------ # Constructor function and methods. #------------------------------------------------------------------------ sub URL { return CLASS unless @_; return @_ == 1 && is_object(CLASS, $_[0]) ? $_[0]->copy # copy existing URL object : CLASS->new(@_); # or construct a new one } sub new { my $class = shift; my $args = @_ == 1 ? shift : { @_ }; my $self; $class = ref $class || $class; if (textlike $args) { $self = bless [$args, $args =~ $MATCH_URL], $class; } elsif (ref $args eq HASH) { $self = bless [ ], $class; $self->set($args); } else { return $class->error("Invalid URL specification: $_[0]"); } return $self; } sub copy { my $self = shift; my $copy = bless [ @$self ], ref $self; return @_ ? $copy->set(@_) : $copy; } sub set { my $self = shift; my $args = @_ && ref $_[0] eq HASH ? shift : { @_ }; my ($k, $v, $n); while (($k, $v) = each %$args) { $n = $ELEMENT->{ $k } || next; $self->[$n] = $v; } # The authority is comprised of the user, host and port fields. # We need to split any authority specified, or merge together the user, # host and port if any of them have been changed $self->split_authority if exists $args->{ authority }; $self->join_authority if exists $args->{ user } or exists $args->{ host } or exists $args->{ port }; # similar thing for query/params $self->split_query if exists $args->{ query }; $self->join_query if exists $args->{ params }; # finally reconstruct the complete url $self->join_url; return $self; } #----------------------------------------------------------------------- # split/join methods #----------------------------------------------------------------------- sub split_authority { my $self = shift; $self->[AUTHORITY] = BLANK unless defined $self->[AUTHORITY]; # this regex shouldn't ever fail as everything is optional @$self[AUTHORITY,USER,HOST,PORT] = $self->[AUTHORITY] =~ $MATCH_AUTHORITY; } sub join_authority { my $self = shift; my ($user, $host, $port) = @$self[USER,HOST,PORT]; $user = (defined $user && length $user) ? $user . '@' : BLANK; $port = (defined $port && length $port) ? ':' . $port : BLANK; $host = BLANK unless defined $host; return ($self->[AUTHORITY] = $user.$host.$port); } sub split_query { my $self = shift; $self->[QUERY] = '' unless defined $self->[QUERY]; return ($self->[PARAMS] = { map { map { decode($_) } split(/=/, $_, 2) } split(/[&;]/, $self->[QUERY]) }); } sub join_query { my $self = shift; my $params = $self->[PARAMS] || { } ; # should we call split_query()? return ($self->[QUERY] = join( '&', map { $_ . '=' . encode( $params->{ $_ } ) } sort keys %$params # sorted makes debugging easier )); } sub join_url { my $self = shift; my $scheme = $self->[SCHEME]; my $auth = $self->[AUTHORITY]; my $query = $self->[QUERY]; my $frag = $self->[FRAGMENT]; $scheme = (defined $scheme && length $scheme) ? $scheme . ':' : BLANK; $auth = (defined $auth && length $auth) ? '//' . $auth : BLANK; $query = (defined $query && length $query) ? '?' . $query : BLANK; $frag = (defined $frag && length $frag) ? '#' . $frag : BLANK; return ($self->[TEXT] = $scheme.$auth.$self->[PATH].$query.$frag); } #----------------------------------------------------------------------- # accessor/mutator methods #----------------------------------------------------------------------- sub text { $_[0]->[TEXT]; } sub scheme { my $self = shift; if (@_) { $self->[SCHEME] = shift; $self->join_url; } return $self->[SCHEME]; } sub authority { my $self = shift; if (@_) { $self->[AUTHORITY] = shift; $self->split_authority; $self->join_url; } return $self->[AUTHORITY]; } sub query { my $self = shift; if (@_) { $self->[QUERY] = shift; $self->split_query; $self->join_url; } return $self->[QUERY]; } sub params { my $self = shift; my $params = $self->[PARAMS] || $self->split_query; if (@_) { my $extra = Badger::Utils::params(@_); # NOTE: this doesn't account for multi-valued params @$params{ keys %$extra } = values %$extra; $self->join_query; $self->join_url; } return $params; } sub server { my $self = shift; my $scheme = $self->[SCHEME]; my $auth = $self->[AUTHORITY]; $scheme = (defined $scheme && length $scheme) ? $scheme.':' : BLANK; $auth = (defined $auth && length $auth) ? '//'.$auth : BLANK; return $scheme.$auth; } sub service { my $self = shift; return $self->server . $self->[PATH]; } sub request { my $self = shift; my $service = $self->service; my $query = $self->[QUERY]; $query = (defined $query && length $query) ? '?'.$query : BLANK; return $self->service . $query; } # This is a quick-hack implementation of relative() that bodges the paths. # This should be replaced with a more robust implementation. It also needs # to be integrated with the work on Badger::Filesystem::Universal. sub relative { my $self = shift; my $path = join(SLASH, @_); my $base = $self->[PATH]; $base =~ s{/$}{}; $path = join(SLASH, $base, $path) unless $path =~ m{^/}; return $self->copy( path => $path ); } sub absolute { my $self = shift; my $path = shift; $path =~ s{^/*}{/}; return $self->copy( path => $path ); } sub dump { my $self = shift; return '[URL:' . join('|', map { defined($_) ? $_ : '' } @$self) . ']'; } #----------------------------------------------------------------------- # generated accessor/mutator methods for those with similar functionality #----------------------------------------------------------------------- class->methods( map { my ($name, $slot) = @$_; $name => sub { my $self = shift; if (@_) { # if any of user, host or port are updated then we must # reconstruct the authority and complete URL $self->[$slot] = shift; $self->join_authority; $self->join_url; } return $self->[$slot]; } } [user => USER], [host => HOST], [port => PORT], ); class->methods( map { my ($name, $slot) = @$_; $name => sub { my $self = shift; if (@_) { # if either of the path or fragment are updated then we # must regenerate the complete URL $self->[$slot] = shift; $self->join_url; } return $self->[$slot]; } } [path => PATH], [fragment => FRAGMENT], ); 1; __END__ =head1 NAME Badger::URL - representation of a Uniform Resource Locator (URL) =head1 SYNOPSIS use Badger::URL; # all-in-one URL string my $url = Badger::URL->new( 'http://abw@badgerpower.com:8080/under/ground?animal=badger#stripe' ); # named parameters my $url = Badger::URL->new( scheme => 'http', user => 'abw', host => 'badgerpower.com', port => '8080', path => '/under/ground', query => 'animal=badger', fragment => 'stripe', ); # methods to access standard W3C parts of URL print $url->scheme; # http print $url->authority; # abw@badgerpower.com:8080 print $url->user; # abw print $url->host; # badgerpower.com print $url->port; # 8080 print $url->path; # /under/ground print $url->query; # animal=badger print $uri->fragment; # stripe # additional composite methods: print $url->server; # http://abw@badgerpower.com:8080 print $url->service; # http://abw@badgerpower.com:8080/under/ground print $url->request; # http://abw@badgerpower.com:8080/under/ground?animal=badger # method to return the whole URL print $url->url(); # http://abw@badgerpower.com:8080/under/ground?animal=badger#stripe # overloaded stringification operator calls url() method print $url; # http://abw@badgerpower.com:8080/under/ground?animal=badger#stripe =head1 DESCRIPTION This module implements an object for representing URLs. It can parse existing URLs to break them down into their constituent parts, and also to generate new or modified URLs. The emphasis is on simplicity and convenience for tasks related to web programming (e.g. dispatching web applications based on the URL, generating URLs for redirects or embedding as links in HTML pages). If you want more generic URI functionality then you should consider using the L module. A URL looks like this: http://abw@badgerpower.com:8080/under/ground?animal=badger#stripe \__/ \______________________/\___________/ \___________/ \____/ | | | | | scheme authority path query fragment The C part can be broken down further: abw@badgerpower.com:8080 \_/ \_____________/ \__/ | | | user host port A L object will parse a URL and store the component parts internally. You can then change any of the individual parts and regenerate the URL. my $url = Badger::URL->new( 'http://badgerpower.com/' ); $url->port('8080'); $url->path('/under/ground'); $url->query('animal=badger'); print $url; # http://badgerpower.com:8080/under/ground?animal=badger =head1 METHODS =head2 new($url) This constructor method is used to create a new URL object. my $url = Badger::URL->new( 'http://abw@badgerpower.com:8080/under/ground?animal=badger#stripe' ); You can also specify the individual parts of the URL using named paramters. my $url = Badger::URL->new( scheme => 'http', user => 'abw', host => 'badgerpower.com', port => '8080', path => '/under/ground', query => 'animal=badger', fragment => 'stripe', ); =head2 copy() This method creates and returns a new C object as a copy of the current one. my $copy = $url->copy; =head2 url() Method to return the complete URL. print $url->url; # http://abw@badgerpower.com:8080/under/ground?animal=badger#stripe This method is called automatically whenever the URL object is stringified. print $url; # same as above =head2 text() An alias for the L method. =head2 scheme() Method to get or set the scheme part of the URL. $url = Badger::URL->new('http://badgerpower.com/); print $url->scheme(); # http $url->scheme('ftp'); print $url->scheme(); # ftp =head2 authority() Method to get or set the authority part of the URL. This is comprised of a host with optional user and/or port. $url->authority('badgerpower.com'); $url->authority('abw@badgerpower.com'); $url->authority('badgerpower.com:8080'); $url->authority('abw@badgerpower.com:8080'); print $url->authority(); # abw@badgerpower.com:8080 =head2 user() Method to get or set the optional user in the authority part of the URL. $url->user('fred'); print $url->user(); # fred print $url->authority(); # fred@badgerpower.com:8080 =head2 host() Get or set the host in the authority part of the URL. $url->host('example.org'); print $url->host(); # example.org print $url->authority(); # fred@example.org:8080 =head2 port() Get or set the port in the authority part of the URL. $url->port(1234); print $url->port(); # 1234 print $url->authority(); # fred@example.org:1234 =head2 path() Get or set the path part of the URL. $url->path('/right/here'); print $url->path(); # /right/here =head2 query() Get or set the query part of the URL. The leading 'C' is not considered part of the query and should be should not be included when setting a new query. $url->query('animal=ferret'); print $url->query(); # animal=ferret =head2 params() Get or set the query parameters. # get params my $params = $url->params; # set params $url->params( x => 10 ); =head2 fragment() Get or set the fragment part of the URL. The leading '#' is not considered part of the fragment and should be should not be included when setting a new fragment. $url->fragment('feet'); print $url->fragment(); # feet =head2 server() Returns a composite of the scheme and authority. print $url->server(); # http://fred@example.org:1234 =head2 service() Returns a composite of the server (scheme and authority) and path (in other words, everything up to the query or fragment). print $url->server(); # http://fred@example.org:1234/right/here =head2 request() Returns a composite of the service (scheme, authority and path) and query (in other words, everything except the fragment). print $url->request(); # http://fred@example.org:1234/right/here?animal=badger =head2 relative($path) Returns a new URL with the relative path specified. my $base = Badger::URL->new('http://badgerpower.com/example'); my $rel = $base->relative('foo/bar'); print $rel; # http://badgerpower.com/example/foo/bar =head2 absolute($path) Returns a new URL with the absolute path specified. The leading C on the path provided as an argument is option. It will be assumed if not present. my $base = Badger::URL->new('http://badgerpower.com/example'); my $rel = $base->absolute('foo/bar'); print $rel; # http://badgerpower.com/foo/bar =head1 INTERNAL METHODS =head2 set($items) This method is used to set internal values. =head2 join_authority() This method reconstructs the C from the C, C and C. =head2 join_query() This method reconstructs the C from the query parameters. =head2 join_url() This method reconstructs the complete URL from its constituent parts. =head2 split_authority() This method splits the C into C, C and C. =head2 split_query() This method splits the C string into query parameters. =head2 dump() Return a text representation of the structure of the URL object, for debugging purposes. =head1 EXPORTABLE SUBROUTINES =head2 URL($url) This constructor function can be used to create a new URL. If the argument is already a C object then it is copied to create a new object. Otherwise a new C object is created from scratch. use Badger::URL 'URL'; my $url1 = URL('http://example.com/foo'); my $url2 = URL($url1); =head1 AUTHOR Andy Wardley L =head1 COPYRIGHT Copyright (C) 2001-2010 Andy Wardley. All Rights Reserved. This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. =head1 SEE ALSO L =cut # Local Variables: # mode: perl # perl-indent-level: 4 # indent-tabs-mode: nil # End: # # vim: expandtab shiftwidth=4: