Bio::DB Registry
SummaryIncluded librariesPackage variablesSynopsisDescriptionGeneral documentationMethods
Toolbar
WebCvsRaw content
Summary
Bio::DB::Registry - Access to the Open Bio Database Access registry scheme
Package variables
Privates (from "my" definitions)
%implement = ( 'biocorba' => 'Bio::CorbaClient::SeqDB', 'flat' => 'Bio::DB::Flat', 'biosql' => 'Bio::DB::BioSQL::BioDatabaseAdaptor', 'biofetch' => 'Bio::DB::BioFetch' )
$fallbackRegistryURL = 'http://www.open-bio.org/registry/seqdatabase.ini'
Included modules
Bio::DB::Failover
Bio::Root::HTTPget
Bio::Root::Root
Inherit
Bio::Root::Root
Synopsis
    use Bio::DB::Registry();
$registry = new Bio::DB::Registry(); @available_services = $registry->services; $db = $registry->get_database('embl'); # $db is a Bio::DB::SeqI implementing class $seq = $db->get_Seq_by_acc("J02231");
Description
This module provides access to the Open Bio Database Access scheme,
which provides a cross language and cross platform specification of how
to get to databases.
If the user or system administrator has not installed the default init
file, seqdatabase.ini, in /etc/bioinformatics or ${HOME}/.bioinformatics
then creating the first Registry object copies the default settings from
the net. The Registry object will attempt to store these settings in
${HOME}/.bioinformatics/seqdatabase.ini.
Users can specify one or more custom locations for the init file by
setting $OBDA_SEARCH_PATH to those directories, where multiple
directories should be separated by ';'.
Methods
BEGIN Code
_load_registry
No description
Code
get_databaseDescriptionCode
new
No description
Code
servicesDescriptionCode
Methods description
get_databasecode    nextTop
 Title   : get_database
Usage : my $db = $registry->get_database($dbname);
Function: Retrieve a Database object which implements Bio::DB::SeqI interface
Returns : Bio::DB::SeqI object
Args : string describing the name of the database
servicescodeprevnextTop
 Title   : services
Usage : my @available = $registry->services();
Function: returns list of possible services
Returns : list of strings
Args : none
Methods code
BEGINTop
BEGIN {
    $OBDA_SPEC_VERSION = 1.0;
    if (defined $ENV{OBDA_SEARCH_PATH}) {
        $OBDA_SEARCH_PATH = $ENV{OBDA_SEARCH_PATH} || '';
}
_load_registrydescriptionprevnextTop
sub _load_registry {
    my ($self) = @_;

    my $home = (getpwuid($>))[7];
    my $f;

    if ( $OBDA_SEARCH_PATH ) {
        foreach ( split /;/,$OBDA_SEARCH_PATH ) {
            next unless -e $_;
            open(F,"$OBDA_SEARCH_PATH/seqdatabase.ini");
            $f =\* F;
            last;
        }
    }
    elsif( -e "$home/.bioinformatics/seqdatabase.ini" ) {
	open(F,"$home/.bioinformatics/seqdatabase.ini");
	$f =\* F;
    } elsif ( -e "/etc/bioinformatics/seqdatabase.ini" ) {
	open(F,"/etc/bioinformatics/seqdatabase.ini");
	$f =\* F;
    } else {
	# waiting for information
$self->warn("No seqdatabase.ini file found in ~/.bioinformatics/\n or in /etc/bioinformatics/.\nor in directory specified by $OBDA_SEARCH_PATH". "Using web to get database registry from\n $fallbackRegistryURL"); # Last gasp. Try to use HTTPget module to retrieve the registry from
# the web...
$f = Bio::Root::HTTPget::getFH($fallbackRegistryURL); # store the default registry file
mkdir "$home/.bioinformatics" unless -e "$home/.bioinformatics"; open(F,">$home/.bioinformatics/seqdatabase.ini"); print F while (<$f>); close F; $self->warn("Stored the default registry configuration into:\n". " $home/.bioinformatics/seqdatabase.ini"); open(F,"$home/.bioinformatics/seqdatabase.ini"); $f =\* F; } while( <$f> ) { /^VERSION=([\d\.]+)/; $self->throw("Do not know about this version [$1] > $OBDA_SPEC_VERSION") if $1 > $OBDA_SPEC_VERSION or !$1; last; } while( <$f> ) { if( /^#/ ) { next; } if( /^\s/ ) { next; } if( /\[(\w+)\]/ ) { my $db = $1; my $hash = {}; while( <$f> ) { chomp(); /^#/ && next; /^$/ && last; my ($tag,$value) = split('=',$_); $value =~ s/\s//g; $tag =~ s/\s//g; $hash->{"\L$tag"} = lc $value; } if( !exists $self->{'_dbs'}->{$db} ) { my $failover = Bio::DB::Failover->new(); $self->{'_dbs'}->{$db}=$failover; } my $class; if (defined $implement{$hash->{'protocol'}}) { $class = $implement{$hash->{'protocol'}}; } else { $self->warn("Registry does not support protocol ".$hash->{'protocol'}); next; } eval "require $class"; if ($@) { $self->verbose && $self->warn("Couldn't load $class"); next; } else { eval { my $randi = $class->new_from_registry(%$hash); $self->{'_dbs'}->{$db}->add_database($randi); }; if ($@) { $self->warn("Couldn't call new_from_registry on [$class]\n$@"); } } next; # back to main loop
} $self->warn("Uninterpretable line in registry, $_"); }
}
get_databasedescriptionprevnextTop
sub get_database {
    my ($self,$dbname) = @_;

    $dbname = lc $dbname;
    if( !defined $dbname ) {
	$self->warn("must get_database with a database name");
	return undef;
    }
    if( !exists $self->{'_dbs'}->{$dbname} ) {
	$self->warn("No database in with $dbname in registry");
	return undef;
    }
    return $self->{'_dbs'}->{$dbname};
}
newdescriptionprevnextTop
sub new {
    my ($class,@args) = shift;
    my $self = $class->SUPER::new(@args);

    # open files in order
$self->{'_dbs'} = {}; $self->_load_registry(); return $self;
}
servicesdescriptionprevnextTop
sub services {
    my ($self) = @_;
    return () unless ( defined $self->{'_dbs'} &&
		       ref( $self->{'_dbs'} ) =~ /HASH/i);
    return keys %{$self->{'_dbs'}};
}


## End of Package
1; __END__
}
General documentation
CONTACTTop
Ewan Birney originally wrote this class.
Reporting BugsTop
Report bugs to the Bioperl bug tracking system to help us keep track
the bugs and their resolution. Bug reports can be submitted via email
or the web:
    bioperl-bugs@bio.perl.org
http://bugzilla.bioperl.org/
APPENDIXTop
The rest of the documentation details each of the object
methods. Internal methods are usually preceded with a _