NAME

Bio::Phylo::PhyloWS::Service - Base class for phylogenetic web services

SYNOPSIS

 # inside a CGI script:
 use CGI;
 use Bio::Phylo::PhyloWS::Service::${child};

 my $service = Bio::Phylo::PhyloWS::Service::${child}->new( '-url' => $url );
 $service->handle_request(CGI->new);

DESCRIPTION

This is the base class for services that implement the PhyloWS (<http://evoinfo.nescent.org/PhyloWS>) recommendations. Such services should subclass this class and implement any relevant abstract methods. Examples of this are Bio::Phylo::PhyloWS::Service::Tolweb and Bio::Phylo::PhyloWS::Service::Ubio.

PhyloWS services are web services for phylogenetics that provide two types of functionality:

Record lookup

Services that implement record lookups are services that know how to process URL requests of the form "/phylows/$object_type/$authority:$identifier?format=$format", where $object_type is a string representing the type of object that is returned, e.g. 'tree', 'matrix', 'taxon', etc., $authority is a naming authority such as TB2 for TreeBASE2, $identifier is a local identifier for the object, for example an accession number, and $format is a serialization format such as 'nexml'.

In order to provide this functionality, subclasses of this class must implement a method called "get_record" which is passed at least a named "-guid" argument that provides the local identifier. The "get_record" method must return a Bio::Phylo::Project object, which is subsequently serialized in the requested format by the "handle_request" method provided here.

Record search

Services that implement record searches are services that know how to process URL requests of the form "/phylows/$object_type/find?query=$query&format=$format", where $object_type is a string representing the type of object to search for, $query is a CQL query (<http://www.loc.gov/standards/sru/specs/cql.html>), and $format is the serialization format in which the returned results are represented.

In order to provide this functionality, subclasses of this class must implement a method called "get_query_result", which is passed the $query parameter and which must return a Bio::Phylo::Project object that combines the search results (e.g. in a single taxa block for taxon searches).

CQL has different levels of support, services may only implement certain levels of support. The example services Bio::Phylo::PhyloWS::Service::Tolweb and Bio::Phylo::PhyloWS::Service::Ubio only proved Level 0, term-only support, meaning that $query is simply a term such as "Homo+sapiens".

Child classes that implement some or all of the functionality described above can subsequently be made operational on a web server by calling them from a simple CGI script as shown in the SYNOPSIS section, where $child must be substituted by the actual class name of the child class (e.g. "Tolweb"). The $url parameter that is passed to the constructor is usually simply the URI of the CGI script, i.e. the environment variable $ENV{'SCRIPT_URI'} under most standard HTTP servers.

REQUEST HANDLER

handle_request()

 Type    : Request handler
 Title   : handle_request
 Usage   : $service->handle_request($cgi);
 Function: Handles a service request
 Returns : prints out response and exits
 Args    : Required: a CGI.pm object

ACCESSORS

get_result()

Gets a phylows result, either a record lookup or a query result, depending on the internal state of the service object

 Type    : Accessor
 Title   : get_result
 Usage   : my $proj = $obj->get_result;
 Function: Gets a phylows result
 Returns : Bio::Phylo::Project
 Args    : None
 Comments:

get_record()

Gets a phylows record by its id

 Type    : Abstract Accessor
 Title   : get_record
 Usage   : my $record = $obj->get_record( -guid => $guid );
 Function: Gets a phylows record by its id
 Returns : Bio::Phylo::Project
 Args    : Required: -guid => $guid, 
           Optional: -format => $format
 Comments: This is an ABSTRACT method that needs to be implemented
           by a child class

get_query_result()

Gets a phylows cql query result

 Type    : Abstract Accessor
 Title   : get_query_result
 Usage   : my $result = $obj->get_query_result( $query );
 Function: Gets a query result 
 Returns : Bio::Phylo::Project
 Args    : Required: $query
 Comments: This is an ABSTRACT method that needs to be implemented
           by a child class

get_supported_formats()

Gets an array ref of supported formats

 Type    : Abstract Accessor
 Title   : get_supported_formats
 Usage   : my @formats = @{ $obj->get_supported_formats };
 Function: Gets an array ref of supported formats
 Returns : ARRAY
 Args    : NONE
 Comments: This is an ABSTRACT method that needs to be implemented
           by a child class

get_redirect()

Gets a redirect URL if relevant

 Type    : Accessor
 Title   : get_redirect
 Usage   : my $url = $obj->get_redirect;
 Function: Gets a redirect URL if relevant
 Returns : String
 Args    : $cgi
 Comments: This method is called by handle_request so that
           services can 303 redirect a record lookup to 
           another URL. By default, this method returns 
           undef (i.e. no redirect)

get_description()

Gets an RSS1.0/XML representation of a phylows record

 Type    : Accessor
 Title   : get_description
 Usage   : my $desc = $obj->get_description;
 Function: Gets an RSS1.0/XML representation of a phylows record
 Returns : String
 Args    : None
 Comments: This method creates a representation of a single record
           (i.e. the service's base url + the record's guid)
           that can be serialized in whichever formats are 
           supported

CITATION

If you use Bio::Phylo in published research, please cite it: