package Net::Google::Calendar;
{
$Net::Google::Calendar::VERSION = '1.05';
}
use strict;
use LWP::UserAgent;
use HTTP::Cookies;
use HTTP::Request;
use HTTP::Headers;
use HTTP::Request::Common;
use XML::Atom::Feed;
use XML::Atom::Entry;
use Data::Dumper;
use Net::Google::AuthSub;
use Net::Google::Calendar::Entry;
use Net::Google::Calendar::Person;
use Net::Google::Calendar::Calendar;
use URI;
use URI::Escape;
use Carp qw(confess);
use vars qw($VERSION $APP_NAME $REDIRECT_MAX);
$APP_NAME = $Net::Google::OAuth::APP_NAME = __PACKAGE__."-${VERSION}";
$REDIRECT_MAX = 10; #Maximum number of redirects to allow
# ABSTRACT: Interface to Google calendars
=head1 NAME
Net::Google::Calendar - programmatic access to Google's Calendar API
=head1 SYNOPSIS
# this will only get you a read only feed
my $cal = Net::Google::Calendar->new( url => $private_url );
or
# this will get you a read-write feed.
my $cal = Net::Google::Calendar->new;
$cal->login($username, $password);
or
# this will also get you a read-write feed
my $cal = Net::Google::Calendar->new;
$cal->auth($username, $auth_token);
or
# this will again get you a read-write feed
my $cal = Net::Google::Calendar->new;
$cal->oauth(Net::Google::OAuth);
or you can pass in a url to specify a particular calendar
my $cal = Net::Google::Calendar->new( url => $non_default_url );
$cal->login($username, $password);
# or $cal->auth($username, $auth_token) obviously
then
for ($cal->get_events()) {
print $_->title."\n";
print $_->content->body."\n*****\n\n";
}
my $c;
for ($cal->get_calendars) {
print $_->title."\n";
print $_->id."\n\n";
$c = $_ if ($_->title eq 'My Non Default Calendar');
}
$cal->set_calendar($c);
print $cal->id." has ".scalar($cal->get_events)." events\n";
# everything below here requires a read-write feed
my $entry = Net::Google::Calendar::Entry->new();
$entry->title($title);
$entry->content("My content");
$entry->location('London, England');
$entry->transparency('transparent');
$entry->status('confirmed');
$entry->when(DateTime->now, DateTime->now() + DateTime::Duration->new( hours => 6 ) );
my $author = Net::Google::Calendar::Person->new();
$author->name('Foo Bar');
$author->email('foo@bar.com');
$entry->author($author);
By default new or updated entries are modified in place with
any new information provided by Google.
$cal->add_entry($entry);
$entry->content('Updated');
$cal->update_entry($entry);
$cal->delete_entry($entry);
However if you don't want the entry updated in place pass
C<no_event_modification> in to the C<new()> method.
my $cal = Net::Google::Calendar->new( no_event_modification => 1 );
$cal->login($user, $pass);
my $tmp = $cal->add_entry($entry);
die "Couldn't add event: $@\n" unless defined $tmp;
print "Events=".scalar($cal->get_events())."\n";
$tmp->content('Updated');
$tmp = $cal->update_entry($tmp) || die "Couldn't update ".$tmp->id.": $@\n";
$cal->delete_entry($tmp) || die "Couldn't delete ".$tmp->id.": $@\n";
=head1 DESCRIPTION
Interact with Google's new calendar using the GData API.
=head1 AUTHENTICATION AND READ-WRITE CALENDARS
There are effectively four ways to get events from a Google calendar.
You can get any public events by querying
http://www.google.com/calendar/feeds/<email>/public/full
Then there are the three ways to get private entries. The first of these
involves a magic cookie in the url like this:
http://www.google.com/calendar/feeds/<email>/private-<key>/full
Google has information on how to find this url here
http://code.google.com/apis/calendar/developers_guide_protocol.html#find_feed_url
To use either the private or public feeds do
my $cal = Net::Google::Calendar->new( url => $url);
Both these feeds will be read only however. This means that you won't be able to
add, update or delete entries.
You can also get all the private entries in a read-write feed by either logging in
or using C<AuthSub>.
Logging in is the easiest. Simply do
my $cal = Net::Google::Calendar->new;
$cal->login($username, $password);
Where C<$username> and C<$password> are the same as if you were logging into the
Google Calendar site.
Alternatively if you don't want to use username and password (if, for example you were
providing Calendar reading as a service on your website and didn't want to have to ask
your users for their Google login details) you can use C<AuthSub>.
http://code.google.com/apis/accounts/AuthForWebApps.html
Once you have an AuthSub token (or you user has supplied you with one)
then you can login using
my $cal = Net::Google::Calendar->new;
$cal->auth($username, $token);
=head1 METHODS
=cut
=head2 new <opts>
Create a new instance. C<opts> is a hash which must contain your private Google url
as the key C<url> unless you plan to log in or authenticate.
See
http://code.google.com/apis/gdata/calendar.html#find_feed_url
for how to get that.
If you pass the option C<no_event_modification> as a psotive value then
add_entry and update_entry will not modify the entry in place.
=cut
sub new {
my ($class, %opts) = @_;
$opts{_ua} = LWP::UserAgent->new( max_redirect => 0 );
$opts{_ua}->env_proxy;
$opts{_auth} = Net::Google::AuthSub->new( service => 'cl' );
$opts{_cookie_jar} = HTTP::Cookies->new;
$opts{no_event_modification} ||= 0;
my $self = bless \%opts, $class;
$self->_find_calendar_id if $opts{url};
return $self;
}
=head2 login <username> <password> [opt[s]]
Login to google.
Can optionally take a hash of options which will override the
default login params.
=over 4
=item service
Name of the Google service for which authorization is requested.
Defaults to 'cl' for calendar.
=item source
Short string identifying your application, for logging purposes.
Defaults to 'Net::Google::Calendar-<VERSION>'
=item accountType
Type of account to be authenticated.
Defaults to 'HOSTED_OR_GOOGLE'.
=back
See http://code.google.com/apis/accounts/AuthForInstalledApps.html#ClientLogin for more details.
=cut
sub login {
my $self = shift;
my $user = shift;
my $pass = shift;
my $r = $self->{_auth}->login($user, $pass);
my $error;
if (!defined $r) {
$error = $@;
} elsif (!$r->is_success) {
$error = $r->error;
}
die "Couldn't log in - $error" if defined $error;
$self->{user} = $user;
$self->_generate_url();
return 1;
}
=head2 auth <username> <token>
Use the AuthSub method for calendar access.
See http://code.google.com/apis/accounts/AuthForWebApps.html
for details.
=cut
sub auth {
my $self = shift;
my $user = shift;
my $token = shift;
$self->{_auth}->auth($user, $token);
$self->{user} = $user;
$self->_generate_url();
return 1;
}
=head2 oauth Net::Google::OAuth
Use OAuth for calendar access
=cut
sub oauth {
my $self = shift;
$self->{_auth} = shift;
}
sub _generate_url {
my $self= shift;
$self->{url} ||= $self->_get_protocol()."://google.com/calendar/feeds/$self->{user}/private/full";
$self->{url} =~ s!/private-[^/]+!/private!;
$self->_find_calendar_id;
}
=head2 auth_object [Net::Google::AuthSub]
Get or set the current C<Net::Google::AuthSub> object.
=cut
sub auth_object {
my $self = shift;
$self->{_auth} = shift if @_;
return $self->{_auth};
}
sub _find_calendar_id {
my $self = shift;
($self->{calendar_id}) = ($self->{url} =~ m!/feeds/([^/]+)/!);
}
=head2 ssl bool
Use ssl or not. Auth tokens (AuthSub and OAuth) have a scope that includes http:// or https://. Make sure you use ssl(1) if your scope is https://www.google.com/calendar/feeds/.
=cut
sub ssl {
my $self = shift;
$self->{_use_ssl} = shift;
}
sub _get_protocol {
my $self = shift;
if ($self->{_use_ssl}) {
return 'https';
}
return 'http';
}
=head2 get_events [ %opts ]
Return a list of Net::Google::Calendar::Entry objects;
You can pass in a hash of options which map to the Google Data API's generic
searching mechanisms plus the specific calendar ones.
See
http://code.google.com/apis/gdata/protocol.html#query-requests
for more details.
=over 4
=item q
Full-text query string
When creating a query, list search terms separated by spaces, in the
form q=term1 term2 term3. (As with all of the query parameter values,
the spaces must be URL encoded.) The GData service returns all entries
that match all of the search terms (like using AND between terms). Like
Google's web search, a GData service searches on complete words (and
related words with the same stem), not substrings.
To search for an exact phrase, enclose the phrase in quotation marks:
q => '"exact phrase'
To exclude entries that match a given term, use the form
q => '-term'
The search is case-insensitive.
Example: to search for all entries that contain the exact phrase
'Elizabeth Bennet' and the word 'Darcy' but don't contain the word
'Austen', use the following query:
q => '"Elizabeth Bennet" Darcy -Austen'
=item category
Category filter
To search in just one category do
category => 'Fritz'
You can query on multiple categories by listing multiple category parameters. For example
category => [ 'Fritz', 'Laurie' ]
returns entries that match both categories.
To do an OR between terms, use a pipe character (|). For example
category => 'Fritz|Laurie'
returns entries that match either category.
To exclude entries that match a given category, use the form
category => '-categoryname'
You can, of course, mix and match
[ 'Jo', 'Fritz|Laurie', '-Simon' ]
means in category
(Jo AND ( Fritz OR Laurie ) AND (NOT Simon))
=item author
Entry author
The service returns entries where the author name and/or email address
match your query string.
=item updated-min
=item updated-max
Bounds on the entry publication date.
Use DateTime objects or the RFC 3339 timestamp format. For example:
2005-08-09T10:57:00-08:00.
The lower bound is inclusive, whereas the upper bound is exclusive.
=item start-min
=item start-max
Respectively, the earliest event start time to match (If not specified,
default is 1970-01-01) and the latest event start time to match (If
not specified, default is 2031-01-01).
Use DateTime objects or the RFC 3339 timestamp format. For example:
2005-08-09T10:57:00-08:00.
The lower bound is inclusive, whereas the upper bound is exclusive.
=item start-index
1-based index of the first result to be retrieved
Note that this isn't a general cursoring mechanism. If you first send a
query with
start-index => 1,
max-results => 10
and then send another query with
start-index => 11,
max-results => 10
the service cannot guarantee that the results are equivalent to
start-index => 1
max-results => 20
because insertions and deletions could have taken place in between the
two queries.
=item max-results
Maximum number of results to be retrieved.
For any service that has a default max-results value (to limit default
feed size), you can specify a very large number if you want to receive
the entire feed.
=item entryID
ID of a specific entry to be retrieved.
If you specify an entry ID, you can't specify any other parameters.
=back
=cut
sub get_events {
my ($self, %opts) = @_;
# check for DateTime objects and convert them to RFC 3339
for (keys %opts) {
next unless UNIVERSAL::isa($opts{$_}, 'DateTime');
# maybe we should chuck an error if it's a Ref and *not* a DateTime
#next unless $opts{$_}->isa('DateTime');
$opts{$_} = $opts{$_}->iso8601 . 'Z';
}
my $url = URI->new($self->{url});
# special handling for single entryID lookup
if (exists $opts{entryID}) {
if (scalar(keys %opts)>1) {
$@ = "You can't specify entryID and anything else";
return undef;
}
my $path = $url->path;
$url->path("$path/".$opts{entryID});
return $self->_get_entry("$url", "Net::Google::Calendar::Entry");
}
if (exists $opts{category} && 'ARRAY' eq ref($opts{category})) {
my $path = $url->path."/".join("/", ( '-', @{delete $opts{category}}));
$url->path("$path");
}
$url->query_form(%opts);
$self->_get("$url", "Net::Google::Calendar::Entry");
}
=head2 add_entry <Net::Google::Calendar::Entry>
Create a new entry.
Returns the new entry with extra data provided by Google but will
also modify the entry in place unless the C<no_event_modification>
option is passed to C<new()>.
Returns undef on failure.
=cut
sub add_entry {
my ($self, $entry) = @_;
# TODO for neatness' sake we could make calendar_id = 'default' when calendar_id = user
my $url = $self->_get_protocol()."://www.google.com/calendar/feeds/$self->{calendar_id}/private/full";
push @_, ($url, 'POST');
goto $self->can('_do');
}
=head2 delete_entry <Net::Google::Calendar::Entry>
Delete a given entry.
Returns undef on failure or the old entry on success.
=cut
sub delete_entry {
my ($self, $entry) = @_;
my $force = (scalar(@_)>2)? pop @_ : 0;
my $url = $entry->edit_url($force) || return undef;
push @_, ($url, 'DELETE');
goto $self->can('_do');
}
=head2 update_entry <Net::Google::Calendar::Entry>
Update a given entry.
Returns the updated entry with extra data provided by Google but will
also modify the entry in place unless the C<no_event_modification>
option is passed to C<new()>.
Returns undef on failure.
=cut
sub update_entry {
my ($self, $entry) = @_;
my $url = $entry->edit_url || return undef;
push @_, ($url, 'PUT');
goto $self->can('_do');
}
=head2 get_calendars <owned>
Get a list of all of a user's Calendars as C<Net::Google::Calendar::Calendar> objects.
If C<owned> is true then only get the ones a user owns.
=cut
sub get_calendars {
my $self = shift;
my $owned = shift || 0;
my $which = ($owned)? "owncalendars" : "allcalendars";
my $url = $self->_get_protocol()."://www.google.com/calendar/feeds/default/$which/full";
return $self->_get("$url", "Net::Google::Calendar::Calendar");
}
sub _get {
my ($self, $url, $class, %opts) = @_;
my $feed = $self->get_feed(URI->new("$url"), %opts);
return map { bless $_, $class; $_->_initialize(); $_ } $feed->entries;
}
=head2 get_feed [feed] [opt[s]]
If C<feed> is a C<URI> object then feed is fetch remotely.
Otherwise it is assumed to be XML data and is parsed.
Returns an C<XML::Atom::Feed> object.
=cut
sub get_feed {
my ($self, $feed, %opts) = @_;
if (ref($feed)){
return $feed if $feed->isa('XML::Atom::Feed');
if ($feed->isa('URI')) {
my %params = ($self->{_auth}->auth_params('GET', $feed), %opts);
my $r = $self->{_ua}->get("$feed", %params);
my $redirect_tries = 0;
while ($r->code == 302 || $r->code == 301) {
my $location = $r->header('location');
%params = ($self->{_auth}->auth_params('GET', $location), %opts);
$r = $self->{_ua}->get($location, %params);
$redirect_tries++;
die "Too many redirects ($redirect_tries)"
if $redirect_tries > $REDIRECT_MAX;
}
die $r->status_line unless $r->is_success;
$feed = $r->content;
}
}
return XML::Atom::Feed->new(\$feed);
}
=head2 update_feed <feed>
Take an C<XML::Atom::Feed> object with a C<http://schemas.google.com/g/2005#post> link and post it.
=cut
sub update_feed {
my ($self, $feed) = @_;
#my $uri = Net::Google::Calendar::Base::_generic_url($feed, 'http://schemas.google.com/g/2005#post') || die("Couldn't get url");
my $uri = Net::Google::Calendar::Base::_generic_url($feed, 'edit') || die("Couldn't get url");
push @_, ($uri, 'POST');
goto $self->can('_do');
}
# TODO collapse this with _get somehow
sub _get_entry {
my ($self, $url, $class) = @_;
my %params = ($self->{_auth}->auth_params);
my $r = $self->{_ua}->get("$url", %params);
if (!$r->is_success) {
if ($r->code == 404) {
$@ = "EntryID not found";
} else {
$@ = $r->status_line;
}
return;
}
my $atom = $r->content;
my $entry = XML::Atom::Entry->new(\$atom);
$entry = bless $entry, $class;
$entry->_initialize();
return $entry;
}
=head2 set_calendar <Net::Google::Calendar::Calendar>
Set the current calendar to use.
=cut
sub set_calendar {
my $self = shift;
my $cal = shift;
($self->{calendar_id}) = (uri_unescape($cal->id) =~ m!([^/]+)$!);
$self->{url} = $self->_get_protocol()."://www.google.com/calendar/feeds/$self->{calendar_id}/private/full";
}
=head2 add_calendar <Net::Google::Calendar::Calendar>
Create a new calendar
Returns the new calendar with extra data provided by Google but will
also modify the entry in place unless the C<no_event_modification>
option is passed to C<new()>.
Returns undef on failure.
=cut
sub add_calendar {
my ($self, $entry) = @_;
my $url = $self->_get_protocol()."://www.google.com/calendar/feeds/$self->{calendar_id}/owncalendars/full";
push @_, ($url, 'POST');
goto $self->can('_do');
}
=head2 update_calendar <Net::Google::Calendar::Calendar>
Update a calendar.
Returns the updated calendar with extra data provided by Google but will
also modify the entry in place unless the C<no_event_modification>
option is passed to C<new()>.
Returns undef on failure.
=cut
sub update_calendar {
my $self = shift;
$self->update_entry(@_);
}
=head2 delete_calendar <Net::Google::Calendar::Calendar> [force]
Delete a given calendar.
Returns undef on failure or the old entry on success.
Note that, at the moment, only C<Calendar> objects returned
by C<get_calendars> with the C<owned> parameter set to C<true>
can be deleted (unlike editing - I don't know if this is a Google
bug or not).
However, you can pass in an optional true C<force> parameter to this
method that will allow C<Calendar> objects returned by C<get_calendars>
where no positive C<owned> paramemter was passed to be deleted. It uses
an egregious hack though and might suddenly stop working if Google change
things or I suddenly decide to remove it.
=cut
sub delete_calendar {
my $self = shift;
$self->delete_entry(@_);
}
sub _do {
my ($self, $entry, $url, $method) = @_;
unless (defined $self->{_auth}) {
$@ = "You must log in to do a $method\n";
return undef;
}
my $class = ref($entry);
my $xml = eval { $entry->as_xml };
confess($@) if $@;
_utf8_off($xml);
my %params = $self->{_auth}->auth_params;
$params{Content_Type} = 'application/atom+xml; charset=UTF-8';
$params{Content} = $xml;
$params{'X-HTTP-Method-Override'} = $method unless "POST" eq $method;
if (defined $self->{_session_id} && !$self->{_force_no_session_id}) {
my $tmp = URI->new($url);
$tmp->query_form({ gsessionid => $self->{_session_id} });
$url = "$tmp";
}
REQUEST: while (1) {
my $rq = POST "$url", %params;
$self->{_cookie_jar}->add_cookie_header($rq);
#my $h = HTTP::Headers->new(%params);
#my $rq = HTTP::Request->new($method => $url, $h);
my $r = $self->{_ua}->request( $rq );
$self->{_cookie_jar}->extract_cookies($r);
my $redirect_tries = 0;
while (302 == $r->code || 301 == $r->code) {
$url = $r->header('location');
my %args = URI->new($url)->query_form;
$self->{_session_id} = $args{gsessionid};
$redirect_tries++;
die "Too many redirects ($redirect_tries)"
if $redirect_tries > $REDIRECT_MAX;
next REQUEST;
}
#print $rq->as_string unless $params{'X-HTTP-Method-Override'} ;
if (!$r->is_success) {
$@ = $r->status_line." - ".$r->content." - $url";
return undef;
}
my $c = $r->content;
if (defined $c && length($c)) {
my $tmp = $class->new(Stream => \$c);
$_[1] = $tmp unless $self->{no_event_modification};
return $tmp;
} else {
# in the case of DELETE should we return 1 instead?
return $entry;
}
}
}
sub _utf8_off {
if ($] >= 5.008) {
require Encode;
return Encode::_utf8_off($_[0]);
}
}
=head1 WARNING
This is ALPHA level software.
Don't use it. Ever. Or something.
=head1 TODO
Abstract this out to Net::Google::Data
=head1 LATEST VERSION
The latest version can always be obtained from my
Subversion repository.
http://svn.unixbeard.net/simon/Net-Google-Calendar
=head1 AUTHOR
Simon Wistow <simon@thegestalt.org>
=head1 COPYRIGHT
Copyright Simon Wistow, 2006
Distributed under the same terms as Perl itself.
=head1 SEE ALSO
http://code.google.com/apis/gdata/calendar.html
=cut
1;