biofetch documentation.

Do not use this object directly, only access it through the
Bio::Biblio module:

  use Bio::Biblio;
  my $biblio = new Bio::Biblio (-access => 'biofetch');
  my $ref = $biblio->get_by_id('20063307'));

  my $ids = ['20063307', '98276153'];
  my $refio = $biblio->get_all($ids);
  while ($ref = $refio->next_bibref) { 
    print $ref->identifier, "\n";
  }

Title : get_all
Usage : $seq = $db->get_all($ref);
Function: Retrieves reference objects from the server 'en masse',
rather than one at a time. For large numbers of sequences,
this is far superior than get_by_id().
Example :
Returns : a stream of Bio::Biblio::Medline objects
Args : $ref : either an array reference, a filename, or a filehandle
from which to get the list of unique ids/accession numbers.

Title : get_by_id
Usage : $entry = $db->get__by_id('20063307')
Function: Gets a Bio::Biblio::RefI object by its name
Returns : a Bio::Biblio::Medline object
Args : the id (as a string) of the reference

Title : get_seq_stream
Usage : my $seqio = $self->get_seq_sream(%qualifiers)
Function: builds a url and queries a web db
Returns : a Bio::SeqIO stream capable of producing sequence
Args : %qualifiers = a hash qualifiers that the implementing class
will process to make a url suitable for web querying

Title : postprocess_data
Usage : $self->postprocess_data ( 'type' => 'string',
'location' => \$datastr);
Function: process downloaded data before loading into a Bio::SeqIO
Returns : void
Args : hash with two keys - 'type' can be 'string' or 'file'
- 'location' either file location or string
reference containing data

BEGIN {

    # set the version for version checking
    $VERSION = do { my @r = (q$$Revision: 1.5 $ =~ /\d+/g); sprintf "%d.%-02d", @r };
    $Revision = q$$Id: biofetch.pm,v 1.5 2002/10/22 07:45:14 lapp Exp $;

    # you can add your own here theoretically.
    %HOSTS = (
	       'dbfetch' => {
		   baseurl => 'http://%s/cgi-bin/dbfetch?db=medline&style=raw',
		   hosts   => {
		       'ebi'  => 'www.ebi.ac.uk'
		       }
	       }
	      );
    %FORMATMAP = ( 'default' => 'medlinexml'
		   );
    $DEFAULTFORMAT = 'default';

    $DEFAULT_SERVICE = 'http://www.ebi.ac.uk/cgi-bin/dbfetch';

}

sub get_seq_stream {

    my ($self, %qualifiers) = @_;
    my ($rformat, $ioformat) = $self->request_format();
    my $seen = 0;
    foreach my $key ( keys %qualifiers ) {
	if( $key =~ /format/i ) {
	    $rformat = $qualifiers{$key};
	    $seen = 1;
	}
    }
    $qualifiers{'-format'} = $rformat if( !$seen);
    ($rformat, $ioformat) = $self->request_format($rformat);
    
    my $request = $self->get_request(%qualifiers);
    my ($stream,$resp);
    if( $self->retrieval_type =~ /temp/i ) {
	my $dir = $self->io()->tempdir( CLEANUP => 1);
	my ( $fh, $tmpfile) = $self->io()->tempfile( DIR => $dir );
	close $fh;
	my ($resp) = $self->_request($request, $tmpfile);		
	if( ! -e $tmpfile || -z $tmpfile || ! $resp->is_success() ) {
            $self->throw("WebDBSeqI Error - check query sequences!\n");
	}
	$self->postprocess_data('type' => 'file',
				'location' => $tmpfile);	
	# this may get reset when requesting batch mode
	($rformat,$ioformat) = $self->request_format();
	if( $self->verbose > 0 ) {
	    open(ERR, "<$tmpfile");
	    while(<ERR>) { $self->debug($_);}
	} 
	$stream = new Bio::Biblio::IO('-format' => $ioformat,
				      '-file'   => $tmpfile);
    } elsif( $self->retrieval_type =~ /io_string/i ) {
	my ($resp) = $self->_request($request);
        my $content = $resp->content_ref;
	$self->debug( "content is $$content\n");
	if( ! $resp->is_success() || length(${$resp->content_ref()}) == 0 ) {
	    $self->throw("WebDBSeqI Error - check query sequences!\n");	
        }  
	($rformat,$ioformat) = $self->request_format();
	$self->postprocess_data('type'=> 'string',
				'location' => $content);
	$stream = new Bio::Biblio::IO('-format' => $ioformat,
#				      '-data'   => "<tag>". $$content. "</tag>");
				      '-data'   => $$content
				      );
    } else { 
	$self->throw("retrieval type " . $self->retrieval_type . 
		     " unsupported\n");
    }
    return $stream;

}

my ($class, @args ) = @_; my $self = $class->SUPER::new(@args); $self->{ '_hosts' } = {}; $self->{ '_formatmap' } = {}; $self->hosts(\%HOSTS); $self->formatmap(\%FORMATMAP); $self->{'_default_format'} = $DEFAULTFORMAT; return $self;

sub postprocess_data {

        my ($self, %args) = @_;
    my $data;
    my $type = uc $args{'type'};
    my $location = $args{'location'};
    if( !defined $type || $type eq '' || !defined $location) {
	return;
    } elsif( $type eq 'STRING' ) {
	$data = $$location; 
    } elsif ( $type eq 'FILE' ) {
	open(TMP, $location) or $self->throw("could not open file $location");
	my @in = <TMP>;
	close TMP;
	$data = join("", @in);
    }

    $data = "<tag>". $data. "</tag>";
    
    if( $type eq 'FILE'  ) {
	open(TMP, ">$location") or $self->throw("could overwrite file $location");
	print TMP $data;
	close TMP;
    } elsif ( $type eq 'STRING' ) {
	${$args{'location'}} = $data;
    }
    
    $self->debug("format is ". $self->request_format(). " data is $data\n");

}

User feedback is an integral part of the evolution of this and other
Bioperl modules. Send your comments and suggestions preferably to
the Bioperl mailing list. Your participation is much appreciated.

  bioperl-l@bioperl.org              - General discussion
  http://bioperl.org/MailList.shtml  - About the mailing lists

BEGIN		Code
get_all	Description	Code
get_by_id	Description	Code
get_seq_stream	Description	Code
new	No description	Code
postprocess_data	Description	Code