=head1 NAME

Bio::DB::GFF::Adaptor::dbi -- Database adaptor for DBI (SQL) databases

=head1 SYNOPSIS

See L<Bio::DB::GFF>

=head1 DESCRIPTION

This is the base class for DBI-based adaptors.  It does everything
except generating the text of the queries to be used.  See the section
QUERIES TO IMPLEMENT for the list of methods that must be implemented.

=cut

package Bio::DB::GFF::Adaptor::dbi;
$Bio::DB::GFF::Adaptor::dbi::VERSION = '1.7.4';
# base class for dbi-based implementations
use strict;

use DBI;
use Bio::DB::GFF::Util::Rearrange; # for rearrange()
use Bio::DB::GFF::Util::Binning;
use Bio::DB::GFF::Adaptor::dbi::iterator;
use Bio::DB::GFF::Adaptor::dbi::caching_handle;

use base qw(Bio::DB::GFF);

# constants for choosing

use constant MAX_SEGMENT => 1_000_000_000;  # the largest a segment can get

# this is the largest that any reference sequence can be (100 megabases)
use constant MAX_BIN    => 1_000_000_000;

# this is the smallest bin (1 K)
use constant MIN_BIN    => 1000;

# size of range over which it is faster to force the database to use the range for indexing
use constant STRAIGHT_JOIN_LIMIT => 200_000;

# this is the size to which DNA should be shredded
use constant DNA_CHUNK_SIZE  => 2000;

# size of summary bins for interval coverage statistics
use constant SUMMARY_BIN_SIZE => 1000;

# for debugging fbin optimization
use constant EPSILON  => 1e-7;  # set to zero if you trust mysql's floating point comparisons
use constant OPTIMIZE => 1;     # set to zero to turn off optimization completely

##############################################################################


=head2 new

 Title   : new
 Usage   : $db = Bio::DB::GFF->new(@args)
 Function: create a new adaptor
 Returns : a Bio::DB::GFF object
 Args    : see below
 Status  : Public

This is the constructor for the adaptor.  It is called automatically
by Bio::DB::GFF-E<gt>new.  In addition to arguments that are common among
all adaptors, the following class-specific arguments are recgonized:

  Argument       Description
  --------       -----------

  -dsn           the DBI data source, e.g. 'dbi:mysql:ens0040'

  -user          username for authentication

  -pass          the password for authentication

=cut

# Create a new Bio::DB::GFF::Adaptor::dbi object
sub new {
  my $class = shift;
  my ($features_db,$username,$auth,$other) = rearrange([
							[qw(FEATUREDB DB DSN)],
							[qw(USERNAME USER)],
							[qw(PASSWORD PASSWD PASS)],
						       ],@_);

  $features_db  || $class->throw("new(): Provide a data source or DBI database");

  if (!ref($features_db)) {
    my $dsn = $features_db;
    my @args;
    push @args,$username if defined $username;
    push @args,$auth     if defined $auth;
    $features_db = Bio::DB::GFF::Adaptor::dbi::caching_handle->new($dsn,@args)
      || $class->throw("new(): Failed to connect to $dsn: "
		       . Bio::DB::GFF::Adaptor::dbi::caching_handle->errstr);
  } else {
    $features_db->isa('DBI::db') 
      || $class->throw("new(): $features_db is not a DBI handle");
  }

  # fill in object
  return bless {
		features_db => $features_db
	       },$class;
}

sub debug {
  my $self = shift;
  $self->features_db->debug(@_);
  $self->SUPER::debug(@_);
}

=head2 features_db

 Title   : features_db
 Usage   : $dbh = $db->features_db
 Function: get database handle
 Returns : a DBI handle
 Args    : none
 Status  : Public

 Note: what is returned is not really a DBI::db handle, but a
 subclass of one.  This means that you cannot manipulate the
 handle's attributes directly.  Instead call the attribute
 method:

 my $dbh = $db->features_db;
 $dbh->attribute(AutoCommit=>0);

=cut

sub features_db { shift->{features_db} }
sub dbh         { shift->{features_db} }

=head2 get_dna

 Title   : get_dna
 Usage   : $string = $db->get_dna($name,$start,$stop,$class)
 Function: get DNA string
 Returns : a string
 Args    : name, class, start and stop of desired segment
 Status  : Public

This method performs the low-level fetch of a DNA substring given its
name, class and the desired range.  It is actually a front end to the
abstract method make_dna_query(), which it calls after some argument
consistency checking.

=cut

sub get_dna {
  my $self = shift;
  my ($ref,$start,$stop,$class) = @_;
  
  my ($offset_start,$offset_stop);

  my $has_start = defined $start;
  my $has_stop  = defined $stop;

  my $reversed;
  if ($has_start && $has_stop && $start > $stop) {
    $reversed++;
    ($start,$stop) = ($stop,$start);
  }

  # turn start and stop into 0-based offsets
  my $cs = $self->dna_chunk_size;
  $start -= 1;  $stop -= 1;
  $offset_start = int($start/$cs)*$cs;
  $offset_stop  = int($stop/$cs)*$cs;

  my $sth;
  # special case, get it all
  if (!($has_start || $has_stop)) {
    $sth = $self->dbh->do_query('select fdna,foffset from fdna where fref=? order by foffset',$ref);
  }

  elsif (!$has_stop) {
    $sth = $self->dbh->do_query('select fdna,foffset from fdna where fref=? and foffset>=? order by foffset',
				$ref,$offset_start);
  }

  else {  # both start and stop defined
    $sth = $self->dbh->do_query('select fdna,foffset from fdna where fref=? and foffset>=? and foffset<=? order by foffset',
				$ref,$offset_start,$offset_stop);
  }

  my $dna = '';
  while (my($frag,$offset) = $sth->fetchrow_array) {
      substr($frag,0,$start-$offset) = '' if $has_start && $start > $offset;
      $dna .= $frag;
  }
  substr($dna,$stop-$start+1) = '' if $has_stop && $stop-$start+1 < length($dna);
  if ($reversed) {
    $dna = reverse $dna;
    $dna =~ tr/gatcGATC/ctagCTAG/;
  }

  $sth->finish;
  $dna;
}


=head2 get_abscoords

 Title   : get_abscoords
 Usage   : ($refseq,$refclass,$start,$stop,$strand) = $db->get_abscoords($name,$class)
 Function: get absolute coordinates for landmark
 Returns : an array ref -- see below
 Args    : name and class of desired landmark
 Status  : Public

This method performs the low-level resolution of a landmark into a
reference sequence and position.

The result is an array ref, each element of which is a five-element
list containing reference sequence name, class, start, stop and strand.

=cut

sub get_abscoords {
  my $self = shift;
  my ($name,$class,$refseq)  = @_;

  my $sth = $self->make_abscoord_query($name,$class,$refseq);

  my @result;
  while (my @row = $sth->fetchrow_array) {
    push @result,\@row
  }
  $sth->finish;

  if (@result == 0) {
    #$self->error("$name not found in database");
    my $sth2 = $self->make_aliasabscoord_query($name,$class);

    while (my @row2 = $sth2->fetchrow_array) {
        push @result,\@row2
    }
    $sth->finish;

    if (@result == 0){
        $self->error("$name not found in database");
        return;
    }
  }
  return \@result;
}


=head2 get_features

 Title   : get_features
 Usage   : $db->get_features($search,$options,$callback)
 Function: retrieve features from the database
 Returns : number of features retrieved
 Args    : see below
 Status  : Public

This is the low-level method that is called to retrieve GFF lines from
the database.  It is responsible for retrieving features that satisfy
range and feature type criteria, and passing the GFF fields to a
callback subroutine.

See the manual page for Bio::DB::GFF for the interpretation of the
arguments and how the information retrieved by get_features is passed
to the callback for processing.

Internally, get_features() is a front end for range_query().  The
latter method constructs the query and executes it.  get_features()
calls fetchrow_array() to recover the fields and passes them to the
callback.

=cut

# Given sequence name, range, and optional filter, retrieve list of
# all features.  Passes features through callback.
sub get_features {
  my $self = shift;
  my ($search,$options,$callback) = @_;
  $callback || $self->throw('must provide a callback argument');

  my $sth = $self->range_query(@{$search}{qw(rangetype
					     refseq
					     refclass
					     start
					     stop
					     types) },
			       @{$options}{qw(
					      sparse
					      sort_by_group
					      ATTRIBUTES
					      BINSIZE)}) or return;

  my $count = 0;
  while (my @row = $sth->fetchrow_array) {
    $callback->(@row);
    $count++;
  }
  $sth->finish;
  return $count;
}

=head2 classes

 Title   : classes
 Usage   : $db->classes
 Function: return list of landmark classes in database
 Returns : a list of classes
 Args    : none
 Status  : public

This routine returns the list of reference classes known to the
database, or empty if classes are not used by the database.  Classes
are distinct from types, being essentially qualifiers on the reference
namespaces.

NOTE: In the current mysql-based schema, this query takes a while to
run due to the classes not being normalized.

=cut

sub classes {
  my $self = shift;
  my ($query,@args) = $self->make_classes_query or return;
  my $sth           = $self->dbh->do_query($query,@args);
  my @classes;
  while (my ($c) = $sth->fetchrow_array) {
     push @classes,$c;
  }
  @classes;
}

=head2 make_classes_query

 Title   : make_classes_query
 Usage   : ($query,@args) = $db->make_classes_query
 Function: return query fragment for generating list of reference classes
 Returns : a query and args
 Args    : none
 Status  : public

=cut

sub make_classes_query {
  my $self = shift;
  return;
}

=head2 _feature_by_name

 Title   : _feature_by_name
 Usage   : $db->get_features_by_name($name,$class,$callback)
 Function: get a list of features by name and class
 Returns : count of number of features retrieved
 Args    : name of feature, class of feature, and a callback
 Status  : protected

This method is used internally.  The callback arguments are those used
by make_feature().  Internally, it invokes the following abstract procedures:

 make_features_select_part
 make_features_from_part
 make_features_by_name_where_part
 make_features_by_alias_where_part  (for aliases)
 make_features_join_part

=cut

sub _feature_by_name {
  my $self = shift;
  my ($class,$name,$location,$callback) = @_;
  $callback || $self->throw('must provide a callback argument');

  my $select         = $self->make_features_select_part;
  my $from           = $self->make_features_from_part(undef,{sparse_groups=>1});
  my ($where,@args)  = $self->make_features_by_name_where_part($class,$name);
  my $join           = $self->make_features_join_part;
  my $range          = $self->make_features_by_range_where_part('overlaps',
								{refseq=>$location->[0],
								 class =>'',
								 start=>$location->[1],
								 stop =>$location->[2]}) if $location;
  # group query
  my $query1  = "SELECT $select FROM $from WHERE $where AND $join";
  $query1    .= " AND $range" if $range;

  # alias query
  $from  = $self->make_features_from_part(undef,{attributes=>1});
  ($where,@args) = $self->make_features_by_alias_where_part($class,$name);  # potential bug - @args1==@args2?

  my $query2  = "SELECT $select FROM $from WHERE $where AND $join";
  $query2    .= " AND $range" if $range;

  my $count = 0;

  for my $query ($query1,$query2) {
    my $sth    = $self->dbh->do_query($query,@args);
    while (my @row = $sth->fetchrow_array) {
      $callback->(@row);
      $count++;
    }
    $sth->finish;
  }

  return $count;
}

=head2 _feature_by_id

 Title   : _feature_by_id
 Usage   : $db->_feature_by_id($ids,$type,$callback)
 Function: get a list of features by ID
 Returns : count of number of features retrieved
 Args    : arrayref containing list of IDs to fetch and a callback
 Status  : protected

This method is used internally.  The $type selector is one of
"feature" or "group".  The callback arguments are those used by
make_feature().  Internally, it invokes the following abstract
procedures:

 make_features_select_part
 make_features_from_part
 make_features_by_id_where_part
 make_features_join_part

=cut

sub _feature_by_id {
  my $self = shift;
  my ($ids,$type,$callback) = @_;
  $callback || $self->throw('must provide a callback argument');

  my $select         = $self->make_features_select_part;
  my $from           = $self->make_features_from_part;
  my ($where,@args)  = $type eq 'feature' ? $self->make_features_by_id_where_part($ids)
                                          : $self->make_features_by_gid_where_part($ids);
  my $join           = $self->make_features_join_part;
  my $query          = "SELECT $select FROM $from WHERE $where AND $join";
  my $sth            = $self->dbh->do_query($query,@args);

  my $count = 0;
  while (my @row = $sth->fetchrow_array) {
    $callback->(@row);
    $count++;
  }
  $sth->finish;
  return $count;
}

sub _feature_by_attribute {
  my $self = shift;
  my ($attributes,$callback) = @_;
  $callback || $self->throw('must provide a callback argument');

  my $select         = $self->make_features_select_part;
  my $from           = $self->make_features_from_part(undef,{attributes=>$attributes});
  my ($where,@args)  = $self->make_features_by_range_where_part('',{attributes=>$attributes});
  my $join           = $self->make_features_join_part({attributes=>$attributes});
  my $query          = "SELECT $select FROM $from WHERE $where AND $join";
  my $sth            = $self->dbh->do_query($query,@args);

  my $count = 0;
  while (my @row = $sth->fetchrow_array) {
    $callback->(@row);
    $count++;
  }
  $sth->finish;
  return $count;
}

=head2 get_types

 Title   : get_types
 Usage   : $db->get_types($refseq,$refclass,$start,$stop,$count)
 Function: get list of types
 Returns : a list of Bio::DB::GFF::Typename objects
 Args    : see below
 Status  : Public

This method is responsible for fetching the list of feature type names
from the database.  The query may be limited to a particular range, in
which case the range is indicated by a landmark sequence name and
class and its subrange, if any.  These arguments may be undef if it is
desired to retrieve all feature types in the database (which may be a
slow operation in some implementations).

If the $count flag is false, the method returns a simple list of
vBio::DB::GFF::Typename objects.  If $count is true, the method returns
a list of $name=E<gt>$count pairs, where $count indicates the number of
times this feature occurs in the range.

Internally, this method calls upon the following functions to generate
the SQL and its bind variables:

  ($q1,@args) = make_types_select_part(@args);
  ($q2,@args) = make_types_from_part(@args);
  ($q3,@args) = make_types_where_part(@args);
  ($q4,@args) = make_types_join_part(@args);
  ($q5,@args) = make_types_group_part(@args);

The components are then combined as follows:

  $query = "SELECT $q1 FROM $q2 WHERE $q3 AND $q4 GROUP BY $q5";

If any of the query fragments contain the ? bind variable, then the
same number of bind arguments must be provided in @args.  The
fragment-generating functions are described below.

=cut

sub get_types {
  my $self = shift;
  my ($srcseq,$class,$start,$stop,$want_count,$typelist) = @_;
  my $straight      = $self->do_straight_join($srcseq,$start,$stop,[]) ? 'straight_join' : '';
  my ($select,@args1) = $self->make_types_select_part($srcseq,$start,$stop,$want_count,$typelist);
  my ($from,@args2)   = $self->make_types_from_part($srcseq,$start,$stop,$want_count,$typelist);
  my ($join,@args3)   = $self->make_types_join_part($srcseq,$start,$stop,$want_count,$typelist);
  my ($where,@args4)  = $self->make_types_where_part($srcseq,$start,$stop,$want_count,$typelist);
  my ($group,@args5)  = $self->make_types_group_part($srcseq,$start,$stop,$want_count,$typelist);

  my $query = "SELECT $straight $select FROM $from WHERE $join AND $where";
  $query   .= " GROUP BY $group" if $group;
  my @args  = (@args1,@args2,@args3,@args4,@args5);
  my $sth = $self->dbh->do_query($query,@args) or return;

  my (%result,%obj);
  while (my ($method,$source,$count) = $sth->fetchrow_array) {
    my $type = Bio::DB::GFF::Typename->new($method,$source);
    $result{$type} = $count;
    $obj{$type} = $type;
  }
  return $want_count ? %result : values %obj;
}

=head2 range_query

 Title   : range_query
 Usage   : $db->range_query($range_type,$refseq,$refclass,$start,$stop,$types,$order_by_group,$attributes,$binsize)
 Function: create statement handle for range/overlap queries
 Returns : a DBI statement handle
 Args    : see below
 Status  : Protected

This method constructs the statement handle for this module's central
query: given a range and/or a list of feature types, fetch their GFF
records.

The positional arguments are as follows:

  Argument               Description

  $isrange               A flag indicating that this is a range.
			 query.  Otherwise an overlap query is
			 assumed.

  $refseq		 The reference sequence name (undef if no range).

  $refclass		 The reference sequence class (undef if no range).

  $start		 The start of the range (undef if none).

  $stop                  The stop of the range (undef if none).

  $types                 Array ref containing zero or feature types in the
			 format [method,source].

  $order_by_group        A flag indicating that statement handler should group
                         the features by group id (handy for iterative fetches)

  $attributes            A hash containing select attributes.

  $binsize               A bin size for generating tables of feature density.

If successful, this method returns a statement handle.  The handle is
expected to return the fields described for get_features().

Internally, range_query() makes calls to the following methods,
each of which is expected to be overridden in subclasses:

  $select        = $self->make_features_select_part;
  $from          = $self->make_features_from_part;
  $join          = $self->make_features_join_part;
  ($where,@args) = $self->make_features_by_range_where_part($isrange,$srcseq,$class,
						           $start,$stop,$types,$class);

The query that is constructed looks like this:

  SELECT $select FROM $from WHERE $join AND $where

The arguments that are returned from make_features_by_range_where_part() are
passed to the statement handler's execute() method.

range_query() also calls a do_straight_join() method, described
below.  If this method returns true, then the keyword "straight_join"
is inserted right after SELECT.

=cut

sub range_query {
  my $self = shift;
  my($rangetype,$refseq,$class,$start,$stop,$types,$sparse,$order_by_group,$attributes,$bin) = @_;

  my $dbh = $self->features_db;

  # NOTE: straight_join is necessary in some database to force the right index to be used.
  my %a             = (refseq=>$refseq,class=>$class,start=>$start,stop=>$stop,types=>$types,attributes=>$attributes,bin_width=>$bin);
  my $straight      = $self->do_straight_join(\%a) ? 'straight_join' : '';
  my $select        = $self->make_features_select_part(\%a);
  my $from          = $self->make_features_from_part($sparse,\%a);
  my $join          = $self->make_features_join_part(\%a);
  my ($where,@args) = $self->make_features_by_range_where_part($rangetype,\%a);
  my ($group_by,@more_args) = $self->make_features_group_by_part(\%a);
  my $order_by      = $self->make_features_order_by_part(\%a) if $order_by_group;

  my $query         = "SELECT $straight $select FROM $from WHERE $join";
  $query           .= " AND $where" if $where;
  if ($group_by) {
    $query           .= " GROUP BY $group_by";
    push @args,@more_args;
  }
  $query           .= " ORDER BY $order_by" if $order_by;

  my $sth = $self->dbh->do_query($query,@args);
  $sth;
}

=head2 make_features_by_range_where_part

 Title   : make_features_by_range_where_part
 Usage   : ($string,@args) =
     $db->make_features_select_part($isrange,$refseq,$class,$start,$stop,$types)
 Function: make where part of the features query
 Returns : the list ($query,@bind_args)
 Args    : see below
 Status  : Protected

This method creates the part of the features query that immediately
follows the WHERE keyword and is ANDed with the string returned by
make_features_join_part().

The six positional arguments are a flag indicating whether to perform
a range search or an overlap search, the reference sequence, class,
start and stop, all of which define an optional range to search in,
and an array reference containing a list [$method,$souce] pairs.

The method result is a multi-element list containing the query string
and the list of runtime arguments to bind to it with the execute()
method.

This method's job is to clean up arguments and perform consistency
checking.  The real work is done by the following abstract methods:

  Method             Description

  refseq_query()     Return the query string needed to match the reference
		     sequence.

  range_query()	     Return the query string needed to find all features contained
		     within a range.

  overlap_query()    Return the query string needed to find all features that overlap
		     a range.

See Bio::DB::Adaptor::dbi::mysql for an example of how this works.

=cut

#'

sub make_features_by_range_where_part {
  my $self = shift;
  my ($rangetype,$options) = @_;
  $options ||= {};
  my ($refseq,$class,$start,$stop,$types,$attributes) =
    @{$options}{qw(refseq class start stop types attributes)};

  my (@query,@args);

  if ($refseq) {
    my ($q,@a) = $self->refseq_query($refseq,$class);
    push @query,$q;
    push @args,@a;
  }

  if (defined $start or defined $stop) {
    $start = 0               unless defined($start);
    $stop  = MAX_SEGMENT     unless defined($stop);

    my ($range_query,@range_args) =
           $rangetype eq 'overlaps'     ? $self->overlap_query($start,$stop)
	 : $rangetype eq 'contains'     ? $self->contains_query($start,$stop)
         : $rangetype eq 'contained_in' ? $self->contained_in_query($start,$stop)
         : ();

    push @query,$range_query;
    push @args,@range_args;
  }

  if (defined $types && @$types) {
    my ($type_query,@type_args) = $self->types_query($types);
    push @query,$type_query;
    push @args,@type_args;
  }

  if ($attributes) {
    my ($attribute_query,@attribute_args) = $self->make_features_by_attribute_where_part($attributes);
    push @query,"($attribute_query)";
    push @args,@attribute_args;
  }

  my $query = join "\n\tAND ",@query;
  return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
}

=head2 do_straight_join

 Title   : do_straight_join
 Usage   : $boolean = $db->do_straight_join($refseq,$class,$start,$stop,$types)
 Function: optimization flag
 Returns : a flag
 Args    : see range_query()
 Status  : Protected

This subroutine, called by range_query() returns a boolean flag.
If true, range_query() will perform a straight join, which can be
used to optimize certain SQL queries.  The four arguments correspond
to similarly-named arguments passed to range_query().

=cut

sub do_straight_join { 0 }  # false by default

=head2 string_match

 Title   : string_match
 Usage   : $string = $db->string_match($field,$value)
 Function: create a SQL fragment for performing exact or regexp string matching
 Returns : query string
 Args    : the table field and match value
 Status  : public

This method examines the passed value for meta characters.  If so it
produces a SQL fragment that performs a regular expression match.
Otherwise, it produces a fragment that performs an exact string match.

This method is not used in the module, but is available for use by
subclasses.

=cut

sub string_match {
  my $self           = shift;
  my ($field,$value) = @_;
  return qq($field = ?) if $value =~ /^[!@%&a-zA-Z0-9_\'\" ~-]+$/;
  return qq($field REGEXP ?);
}

=head2 exact_match

 Title   : exact_match
 Usage   : $string = $db->exact_match($field,$value)
 Function: create a SQL fragment for performing exact string matching
 Returns : query string
 Args    : the table field and match value
 Status  : public

This method produces the SQL fragment for matching a field name to a
constant string value.

=cut

sub exact_match {
  my $self           = shift;
  my ($field,$value) = @_;
  return qq($field = ?);
}

=head2 search_notes

 Title   : search_notes
 Usage   : @search_results = $db->search_notes("full text search string",$limit)
 Function: Search the notes for a text string, using mysql full-text search
 Returns : array of results
 Args    : full text search string, and an optional row limit
 Status  : public

This is a mysql-specific method.  Given a search string, it performs a
full-text search of the notes table and returns an array of results.
Each row of the returned array is a arrayref containing the following fields:

  column 1     A Bio::DB::GFF::Featname object, suitable for passing to segment()
  column 2     The text of the note
  column 3     A relevance score.
  column 4     A Bio::DB::GFF::Typename object

=cut

sub search_notes {
  my $self = shift;
  my ($search_string,$limit) = @_;

  $search_string =~ tr/*?//d; 

  my @words  = $search_string =~ /(\w+)/g;
  my $regex  = join '|',@words;
  my @searches = map {"fattribute_value LIKE '%${_}%'"} @words;
  my $search   = join(' OR ',@searches);

  my $query = <<END;
SELECT distinct gclass,gname,fattribute_value,fmethod,fsource
  FROM fgroup,fattribute_to_feature,fdata,ftype
  WHERE fgroup.gid=fdata.gid
     AND fdata.fid=fattribute_to_feature.fid
     AND fdata.ftypeid=ftype.ftypeid
     AND ($search)
END
;

  my $sth = $self->dbh->do_query($query);
  my @results;
  while (my ($class,$name,$note,$method,$source) = $sth->fetchrow_array) {
     next unless $class && $name;    # sorry, ignore NULL objects
     my @matches = $note =~ /($regex)/g;
     my $relevance = 10*@matches;
     my $featname = Bio::DB::GFF::Featname->new($class=>$name);
     my $type     = Bio::DB::GFF::Typename->new($method,$source);
     push @results,[$featname,$note,$relevance,$type];
     last if $limit && @results >= $limit;
  }
  @results;
}


=head2 meta

 Title   : meta
 Usage   : $value = $db->meta($name [,$newval])
 Function: get or set a meta variable
 Returns : a string
 Args    : meta variable name and optionally value
 Status  : public

Get or set a named metavariable for the database.  Metavariables can
be used for database-specific settings.  This method calls two
class-specific methods which must be implemented:

  make_meta_get_query()   Returns a sql fragment which given a meta
                          parameter name, returns its value.  One bind
                          variable.
  make_meta_set_query()   Returns a sql fragment which takes two bind
                          arguments, the parameter name and its value


Don't make changes unless you know what you're doing!  It will affect the
persistent database.

=cut

sub meta {
  my $self = shift;
  my $param_name = uc shift;

  # getting
  if (@_) {
    my $value = shift;
    my $sql = $self->make_meta_set_query() or return;
    my $sth = $self->dbh->prepare_delayed($sql) 
              or $self->error("Can't prepare $sql: ",$self->dbh->errstr), return;
    $sth->execute($param_name,$value)
              or $self->error("Can't execute $sql: ",$self->dbh->errstr), return;
    $sth->finish;
    return $self->{meta}{$param_name} = $value;
  }

  elsif (exists $self->{meta}{$param_name}) {
    return $self->{meta}{$param_name};
  }

  else {
    undef $self->{meta}{$param_name};  # so that we don't check again
    my $sql = $self->make_meta_get_query() or return;
    my $sth  = $self->dbh->prepare_delayed($sql)
            or $self->error("Can't prepare $sql: ",$self->dbh->errstr), return;
    $sth->execute($param_name)
            or $self->error("Can't execute $sql: ",$sth->errstr),return;
    my ($value) = $sth->fetchrow_array;
    $sth->finish;
    return $self->{meta}{$param_name} = $value;
  }

}

=head2 make_meta_get_query

 Title   : make_meta_get_query
 Usage   : $sql = $db->make_meta_get_query
 Function: return SQL fragment for getting a meta parameter
 Returns : SQL fragment
 Args    : none
 Status  : public

By default this does nothing; meta parameters are not stored or
retrieved.

=cut

sub make_meta_get_query {
   return 'SELECT fvalue FROM fmeta WHERE fname=?';
}


sub dna_chunk_size {
  my $self = shift;
  $self->meta('chunk_size') || DNA_CHUNK_SIZE;
}

=head2 make_meta_set_query

 Title   : make_meta_set_query
 Usage   : $sql = $db->make_meta_set_query
 Function: return SQL fragment for setting a meta parameter
 Returns : SQL fragment
 Args    : none
 Status  : public

By default this does nothing; meta parameters are not stored or
retrieved.

=cut

sub make_meta_set_query {
  return;
}

=head2 default_meta_values

 Title   : default_meta_values
 Usage   : %values = $db->default_meta_values
 Function: empty the database
 Returns : a list of tag=>value pairs
 Args    : none
 Status  : protected

This method returns a list of tag=E<gt>value pairs that contain default
meta information about the database.  It is invoked by initialize() to
write out the default meta values.  The base class version returns an
empty list.

For things to work properly, meta value names must be UPPERCASE.

=cut

sub default_meta_values {
  my $self = shift;
  my @values = $self->SUPER::default_meta_values;
  return (
	  @values,
	  max_bin             => MAX_BIN,
	  min_bin             => MIN_BIN,
	  straight_join_limit => STRAIGHT_JOIN_LIMIT,
          chunk_size          => DNA_CHUNK_SIZE,
	 );
}

sub min_bin {
  my $self = shift;
  return $self->meta('min_bin') || MIN_BIN;
}
sub max_bin {
  my $self = shift;
  return $self->meta('max_bin') || MAX_BIN;
}

sub straight_join_limit {
  my $self = shift;
  return $self->meta('straight_join_limit') || STRAIGHT_JOIN_LIMIT;
}

=head2 get_features_iterator

 Title   : get_features_iterator
 Usage   : $iterator = $db->get_features_iterator($search,$options,$callback)
 Function: create an iterator on a features() query
 Returns : A Bio::DB::GFF::Adaptor::dbi::iterator object
 Args    : see get_features()
 Status  : public

This method is similar to get_features(), except that it returns an
iterator across the query.  See
L<Bio::DB::GFF::Adaptor::dbi::iterator>.

=cut

sub get_features_iterator {
  my $self = shift;
  my ($search,$options,$callback) = @_;
  $callback || $self->throw('must provide a callback argument');
  my $sth = $self->range_query(@{$search}{qw(rangetype
					     refseq
					     refclass
					     start
					     stop
					     types)},
			       @{$options}{qw(
					      sparse
					      sort_by_group
					      ATTRIBUTES
					      BINSIZE)}) or return;
  return Bio::DB::GFF::Adaptor::dbi::iterator->new($sth,$callback);
}

########################## loading and initialization  #####################

=head2 do_initialize

 Title   : do_initialize
 Usage   : $success = $db->do_initialize($drop_all)
 Function: initialize the database
 Returns : a boolean indicating the success of the operation
 Args    : a boolean indicating whether to delete existing data
 Status  : protected

This method will load the schema into the database.  If $drop_all is
true, then any existing data in the tables known to the schema will be
deleted.

Internally, this method calls schema() to get the schema data.

=cut

# Create the schema from scratch.
# You will need create privileges for this.
sub do_initialize {
  #shift->throw("do_initialize(): must be implemented by subclass");
  my $self = shift;
  my $erase = shift;
  $self->drop_all if $erase;

  my $dbh = $self->features_db;
  my $schema = $self->schema;
  foreach my $table_name ($self->tables) {
    my $create_table_stmt = $schema->{$table_name}{table} ;
    $dbh->do($create_table_stmt) ||  warn $dbh->errstr;
    $self->create_other_schema_objects(\%{$schema->{$table_name}});
  }

  1;
}

=head2 finish_load

 Title   : finish_load
 Usage   : $db->finish_load
 Function: called after load_gff_line()
 Returns : number of records loaded
 Args    : none
 Status  : protected

This method performs schema-specific cleanup after loading a set of
GFF records.  It finishes each of the statement handlers prepared by
setup_load().

=cut

sub finish_load {
  my $self = shift;

  my $dbh = $self->features_db or return;
  $dbh->do('UNLOCK TABLES') if $self->lock_on_load;

  foreach (keys %{$self->{load_stuff}{sth}}) {
    $self->{load_stuff}{sth}{$_}->finish;
  }

  my $counter = $self->{load_stuff}{counter};
  delete $self->{load_stuff};
  return $counter;
}


=head2 create_other_schema_objects

 Title   : create_other_schema_objects
 Usage   : $self->create_other_schema_objects($table_name)
 Function: create other schema objects like : indexes, sequences, triggers
 Returns : 
 Args    : 
 Status  : Abstract

=cut

sub create_other_schema_objects{
  #shift->throw("create_other_schema_objects(): must be implemented by subclass");
  my $self = shift ;
  my $table_schema = shift ;
  my $dbh = $self->features_db;
  foreach my $object_type(keys %$table_schema){
    if ($object_type !~ /table/) {
      foreach my $object_name(keys %{$table_schema->{$object_type}}){
        my $create_object_stmt = $table_schema->{$object_type}{$object_name};
        $dbh->do($create_object_stmt) ||  warn $dbh->errstr;   
      }
    }
  }
  1;
}

=head2 drop_all

 Title   : drop_all
 Usage   : $db->drop_all
 Function: empty the database
 Returns : void
 Args    : none
 Status  : protected

This method drops the tables known to this module.  Internally it
calls the abstract tables() method.

=cut

# Drop all the GFF tables -- dangerous!
sub drop_all {
  #shift->throw("drop_all(): must be implemented by subclass");
  my $self = shift;
  my $dbh = $self->features_db;
  my $schema = $self->schema;

  local $dbh->{PrintError} = 0;
  foreach ($self->tables) {
    $dbh->do("drop table $_") || warn $dbh->errstr;

    #when dropping a table - the indexes and triggers are being dropped automatically
    # sequences needs to be dropped - if there are any (Oracle, PostgreSQL)
    if ($schema->{$_}{sequence}){
      foreach my $sequence_name(keys %{$schema->{$_}{sequence}}) {
	$dbh->do("drop sequence $sequence_name");
      }
    }

    #$self->drop_other_schema_objects($_);
    
  }
}

=head2 clone

The clone() method should be used when you want to pass the
Bio::DB::GFF object to a child process across a fork(). The child must
call clone() before making any queries.

This method does two things: (1) it sets the underlying database
handle's InactiveDestroy parameter to 1, thereby preventing the
database connection from being destroyed in the parent when the dbh's
destructor is called; (2) it replaces the dbh with the result of
dbh-E<gt>clone(), so that we now have an independent handle.

=cut

sub clone {
    my $self = shift;
    $self->features_db->clone;
}


=head1 QUERIES TO IMPLEMENT

The following astract methods either return DBI statement handles or
fragments of SQL.  They must be implemented by subclasses of this
module.  See Bio::DB::GFF::Adaptor::dbi::mysql for examples.




=head2 drop_other_schema_objects

 Title   : drop_other_schema_objects
 Usage   : $self->create_other_schema_objects($table_name)
 Function: create other schema objects like : indexes, sequences, triggers
 Returns : 
 Args    : 
 Status  : Abstract


=cut

sub drop_other_schema_objects{
  #shift->throw("drop_other_schema_objects(): must be implemented by subclass");
  
}

=head2 make_features_select_part

 Title   : make_features_select_part
 Usage   : $string = $db->make_features_select_part()
 Function: make select part of the features query
 Returns : a string
 Args    : none
 Status  : Abstract

This abstract method creates the part of the features query that
immediately follows the SELECT keyword.

=cut

sub make_features_select_part {
  shift->throw("make_features_select_part(): must be implemented by subclass");
}

=head2 tables

 Title   : tables
 Usage   : @tables = $db->tables
 Function: return list of tables that belong to this module
 Returns : list of tables
 Args    : none
 Status  : protected

This method lists the tables known to the module.

=cut

# return list of tables that "belong" to us. 
sub tables {
  my $schema = shift->schema;
  return keys %$schema;
}

=head2 schema

 Title   : schema
 Usage   : $schema = $db->schema
 Function: return the CREATE script for the schema
 Returns : a hashref
 Args    : none
 Status  : abstract

This method returns an array ref containing the various CREATE
statements needed to initialize the database tables.  The keys are the
table names, and the values are strings containing the appropriate
CREATE statement.

=cut

sub schema {
  shift->throw("The schema() method must be implemented by subclass");
}

=head2 DESTROY

 Title   : DESTROY
 Usage   : $db->DESTROY
 Function: disconnect database at destruct time
 Returns : void
 Args    : none
 Status  : protected

This is the destructor for the class.

=cut

sub DESTROY {
  my $self = shift;
  $self->features_db->disconnect if defined $self->features_db;
}

################## query cache ##################


#########################################  
## Moved from mysql.pm and mysqlopt.pm ##
#########################################

=head2 make_features_by_name_where_part

 Title   : make_features_by_name_where_part
 Usage   : $db->make_features_by_name_where_part
 Function: create the SQL fragment needed to select a feature by its group name & class
 Returns : a SQL fragment and bind arguments
 Args    : see below
 Status  : Protected

=cut

sub make_features_by_name_where_part {
  my $self = shift;
  my ($class,$name) = @_;
  if ($name =~ /\*/) {
    $name =~ s/%/\\%/g;
    $name =~ s/_/\\_/g;
    $name =~ tr/*/%/;
    return ("fgroup.gclass=? AND fgroup.gname LIKE ?",$class,$name);
  } else {
    return ("fgroup.gclass=? AND fgroup.gname=?",$class,$name);
  }
}

sub make_features_by_alias_where_part {
  my $self = shift;
  my ($class,$name) = @_;
  if ($name =~ /\*/) {
    $name =~ tr/*/%/;
    $name =~ s/_/\\_/g;
    return ("fgroup.gclass=? AND fattribute_to_feature.fattribute_value LIKE ? AND fgroup.gid=fdata.gid AND fattribute.fattribute_name in ('Alias','Name') AND fattribute_to_feature.fattribute_id=fattribute.fattribute_id AND fattribute_to_feature.fid=fdata.fid AND ftype.ftypeid=fdata.ftypeid",$class,$name)
  } else {
    return ("fgroup.gclass=? AND fattribute_to_feature.fattribute_value=? AND fgroup.gid=fdata.gid AND fattribute.fattribute_name in ('Alias','Name') AND fattribute_to_feature.fattribute_id=fattribute.fattribute_id AND fattribute_to_feature.fid=fdata.fid AND ftype.ftypeid=fdata.ftypeid",$class,$name);
  }

}

sub make_features_by_attribute_where_part {
  my $self = shift;
  my $attributes = shift;
  my @args;
  my @sql;
  foreach (keys %$attributes) {
     push @sql,"(fattribute.fattribute_name=? AND fattribute_to_feature.fattribute_value=?)";
     push @args,($_,$attributes->{$_});
  }
  return (join(' OR ',@sql),@args);
}

=head2 make_features_by_id_where_part

 Title   : make_features_by_id_where_part
 Usage   : $db->make_features_by_id_where_part($ids)
 Function: create the SQL fragment needed to select a set of features by their ids
 Returns : a SQL fragment and bind arguments
 Args    : arrayref of IDs
 Status  : Protected

=cut

sub make_features_by_id_where_part {
  my $self = shift;
  my $ids = shift;
  my $set = join ",",@$ids;
  return ("fdata.fid IN ($set)");
}

=head2 make_features_by_gid_where_part

 Title   : make_features_by_id_where_part
 Usage   : $db->make_features_by_gid_where_part($ids)
 Function: create the SQL fragment needed to select a set of features by their ids
 Returns : a SQL fragment and bind arguments
 Args    : arrayref of IDs
 Status  : Protected

=cut

sub make_features_by_gid_where_part {
  my $self = shift;
  my $ids = shift;
  my $set = join ",",@$ids;
  return ("fgroup.gid IN ($set)");
}


=head2 make_features_from_part

 Title   : make_features_from_part
 Usage   : $string = $db->make_features_from_part()
 Function: make from part of the features query
 Returns : a string
 Args    : none
 Status  : protected

This method creates the part of the features query that immediately
follows the FROM keyword.

=cut

sub make_features_from_part {
  my $self = shift;
  my $sparse = shift;
  my $options = shift || {};
  return $options->{attributes} ? "fdata,ftype,fgroup,fattribute,fattribute_to_feature\n"
                                : "fdata,ftype,fgroup\n";
}


=head2 make_features_join_part

 Title   : make_features_join_part
 Usage   : $string = $db->make_features_join_part()
 Function: make join part of the features query
 Returns : a string
 Args    : none
 Status  : protected

This method creates the part of the features query that immediately
follows the WHERE keyword.

=cut

sub make_features_join_part {
  my $self = shift;
  my $options = shift || {};
  return !$options->{attributes} ? <<END1 : <<END2;
  fgroup.gid = fdata.gid 
  AND ftype.ftypeid = fdata.ftypeid
END1
  fgroup.gid = fdata.gid 
  AND ftype.ftypeid = fdata.ftypeid
  AND fattribute.fattribute_id=fattribute_to_feature.fattribute_id
  AND fdata.fid=fattribute_to_feature.fid
END2
}

=head2 make_features_order_by_part

 Title   : make_features_order_by_part
 Usage   : ($query,@args) = $db->make_features_order_by_part()
 Function: make the ORDER BY part of the features() query
 Returns : a SQL fragment and bind arguments, if any
 Args    : none
 Status  : protected

This method creates the part of the features query that immediately
follows the ORDER BY part of the query issued by features() and
related methods.

=cut

sub make_features_order_by_part {
  my $self = shift;
  my $options = shift || {};
  return "fgroup.gname";
}

=head2 make_features_group_by_part

 Title   : make_features_group_by_part
 Usage   : ($query,@args) = $db->make_features_group_by_part()
 Function: make the GROUP BY part of the features() query
 Returns : a SQL fragment and bind arguments, if any
 Args    : none
 Status  : protected

This method creates the part of the features query that immediately
follows the GROUP BY part of the query issued by features() and
related methods.

=cut

sub make_features_group_by_part {
  my $self = shift;
  my $options = shift || {};
  if (my $att = $options->{attributes}) {
    my $key_count = keys %$att;
    return unless $key_count > 1;
    return ("fdata.fid,fref,fstart,fstop,fsource,
           fmethod,fscore,fstrand,fphase,gclass,gname,ftarget_start,
           ftarget_stop,fdata.gid
     HAVING count(fdata.fid) > ?",$key_count-1);
  }
  elsif (my $b = $options->{bin_width}) {
    return "fref,fstart,fdata.ftypeid";
  }

}

=head2 refseq_query

 Title   : refseq_query
 Usage   : ($query,@args) = $db->refseq_query($name,$class)
 Function: create SQL fragment that selects the desired reference sequence
 Returns : a list containing the query and bind arguments
 Args    : reference sequence name and class
 Status  : protected

This method is called by make_features_by_range_where_part() to
construct the part of the select WHERE section that selects a
particular reference sequence.  It returns a mult-element list in
which the first element is the SQL fragment and subsequent elements
are bind values.

For example:

  sub refseq_query {
     my ($name,$class) = @_;
     return ('gff.refseq=? AND gff.refclass=?',
	     $name,$class);
  }

The current schema does not distinguish among different classes of
reference sequence.

=cut

# IMPORTANT NOTE: THE MYSQL SCHEMA IGNORES THE SEQUENCE CLASS
# THIS SHOULD BE FIXED
sub refseq_query {
  my $self = shift;
  my ($refseq,$refclass) = @_;
  my $query = "fdata.fref=?";
  return wantarray ? ($query,$refseq) : $self->dbh->dbi_quote($query,$refseq);
}

=head2 attributes

 Title   : attributes
 Usage   : @attributes = $db->attributes($id,$name)
 Function: get the attributes on a particular feature
 Returns : an array of string
 Args    : feature ID
 Status  : public

Some GFF version 2 files use the groups column to store a series of
attribute/value pairs.  In this interpretation of GFF, the first such
pair is treated as the primary group for the feature; subsequent pairs
are treated as attributes.  Two attributes have special meaning:
"Note" is for backward compatibility and is used for unstructured text
remarks.  "Alias" is considered as a synonym for the feature name.

If no name is provided, then attributes() returns a flattened hash, of
attribute=E<gt>value pairs.  This lets you do:

  %attributes = $db->attributes($id);

Normally, attributes() will be called by the feature:

  @notes = $feature->attributes('Note');

=cut

sub do_attributes {
  my $self        = shift;
  my ($id,$tag)   = @_;
  my $sth;
  if ($id) {
    my $from   = 'fattribute_to_feature,fattribute';
    my $join   = 'fattribute.fattribute_id=fattribute_to_feature.fattribute_id';
    my $where1 = 'fid=? AND fattribute_name=?';
    my $where2 = 'fid=?';
    $sth = defined($tag) ? $self->dbh->do_query("SELECT fattribute_value FROM $from WHERE $where1 AND $join",$id,$tag)
                         : $self->dbh->do_query("SELECT fattribute_name,fattribute_value FROM $from WHERE $where2 AND $join",$id);
  }
  else {
    $sth = $self->dbh->do_query("SELECT fattribute_name FROM fattribute");
  }
  my @result;
  while (my @stuff = $sth->fetchrow_array) {
    push @result,@stuff;
  }
  $sth->finish;
  return @result;
}



=head2 overlap_query_nobin

 Title   : overlap_query
 Usage   : ($query,@args) = $db->overlap_query($start,$stop)
 Function: create SQL fragment that selects the desired features by range
 Returns : a list containing the query and bind arguments
 Args    : the start and stop of a range, inclusive
 Status  : protected

This method is called by make_features_byrange_where_part() to construct the
part of the select WHERE section that selects a set of features that
overlap a range. It returns a multi-element list in which the first
element is the SQL fragment and subsequent elements are bind values.


sub overlap_query_nobin {
     my ($start,$stop) = @_;
     return ('gff.stopE<gt>=? AND gff.startE<lt>=?',
	     $start,$stop);

=cut

# find features that overlap a given range
sub overlap_query_nobin {
  my $self = shift;
  my ($start,$stop) = @_;

  my $query    = qq(fdata.fstop>=? AND fdata.fstart<=?);
  return wantarray ? ($query,$start,$stop) : $self->dbh->dbi_quote($query,$start,$stop);
}

=head2 contains_query_nobin

 Title   : contains_query
 Usage   : ($query,@args) = $db->contains_query_nobin($start,$stop)
 Function: create SQL fragment that selects the desired features by range
 Returns : a list containing the query and bind arguments
 Args    : the start and stop of a range, inclusive
 Status  : protected

This method is called by make_features_byrange_where_part() to construct the
part of the select WHERE section that selects a set of features
entirely enclosed by a range. It returns a multi-element list in which
the first element is the SQL fragment and subsequent elements are bind
values. For example:

  sub contains_query_nobin {
     my ($start,$stop) = @_;
     return ('gff.start>=? AND gff.stop<=?',
	     $start,$stop);

=cut

# find features that are completely contained within a range
sub contains_query_nobin {
  my $self = shift;
  my ($start,$stop) = @_;
  my $query    = qq(fdata.fstart>=? AND fdata.fstop<=?);
  return wantarray ? ($query,$start,$stop) : $self->dbh->dbi_quote($query,$start,$stop);
}

=head2 contained_in_query_nobin

 Title   : contained_in_query_nobin
 Usage   : ($query,@args) = $db->contained_in_query($start,$stop)
 Function: create SQL fragment that selects the desired features by range
 Returns : a list containing the query and bind arguments
 Args    : the start and stop of a range, inclusive
 Status  : protected

This method is called by make_features_byrange_where_part() to construct the
part of the select WHERE section that selects a set of features
entirely enclosed by a range. It returns a multi-element list in which
the first element is the SQL fragment and subsequent elements are bind
values.For example:

  sub contained_in_query_nobin {
     my ($start,$stop) = @_;
     return ('gff.start<=? AND gff.stop>=?',
	     $start,$stop);
  }

=cut

# find features that are completely contained within a range
sub contained_in_query_nobin {
  my $self = shift;
  my ($start,$stop) = @_;
  my $query    = qq(fdata.fstart<=? AND fdata.fstop>=?);
  return wantarray ? ($query,$start,$stop) : $self->dbh->dbi_quote($query,$start,$stop);
}

=head2 types_query

 Title   : types_query
 Usage   : ($query,@args) = $db->types_query($types)
 Function: create SQL fragment that selects the desired features by type
 Returns : a list containing the query and bind arguments
 Args    : an array reference containing the types
 Status  : protected

This method is called by make_features_byrange_where_part() to construct the
part of the select WHERE section that selects a set of features based
on their type. It returns a multi-element list in which the first
element is the SQL fragment and subsequent elements are bind values.
The argument is an array reference containing zero or more
[$method,$source] pairs.

=cut

# generate the fragment of SQL responsible for searching for
# features with particular types and methods
sub types_query {
  my $self = shift;
  my $types = shift;

  my @method_queries;
  my @args;
  for my $type (@$types) {
    my ($method,$source) = @$type;
    my ($mlike, $slike) = (0, 0);
    if ($method && $method =~ m/\.\*/) {
      $method =~ s/%/\\%/g;
      $method =~ s/_/\\_/g;
      $method =~ s/\.\*\??/%/g;
      $mlike++;
    }
    if ($source && $source =~ m/\.\*/) {
      $source =~ s/%/\\%/g;
      $source =~ s/_/\\_/g;
      $source =~ s/\.\*\??/%/g;
      $slike++;
    }
    my @pair;
    if (defined $method && length $method) {
	push @pair, $mlike ? qq(fmethod LIKE ?) : qq(fmethod = ?);
	push @args, $method;
    }
    if (defined $source && length $source) {
	push @pair, $slike ? qq(fsource LIKE ?) : qq(fsource = ?);
	push @args, $source;
    }
    push @method_queries,"(" . join(' AND ',@pair) .")" if @pair;
}
  my $query = " (".join(' OR ',@method_queries).")\n" if @method_queries;
  return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
}

=head2 make_types_select_part

 Title   : make_types_select_part
 Usage   : ($string,@args) = $db->make_types_select_part(@args)
 Function: create the select portion of the SQL for fetching features type list
 Returns : query string and bind arguments
 Args    : see below
 Status  : protected

This method is called by get_types() to generate the query fragment
and bind arguments for the SELECT part of the query that retrieves
lists of feature types.  The four positional arguments are as follows:

 $refseq      reference sequence name
 $start       start of region
 $stop        end of region
 $want_count  true to return the count of this feature type

If $want_count is false, the SQL fragment returned must produce a list
of feature types in the format (method, source).

If $want_count is true, the returned fragment must produce a list of
feature types in the format (method, source, count).

=cut

#------------------------- support for the types() query ------------------------
sub make_types_select_part {
  my $self = shift;
  my ($srcseq,$start,$stop,$want_count) = @_;
  my $query = $want_count ? 'ftype.fmethod,ftype.fsource,count(fdata.ftypeid)'
                          : 'fmethod,fsource';
  return $query;
}

=head2 make_types_from_part

 Title   : make_types_from_part
 Usage   : ($string,@args) = $db->make_types_from_part(@args)
 Function: create the FROM portion of the SQL for fetching features type lists
 Returns : query string and bind arguments
 Args    : see below
 Status  : protected

This method is called by get_types() to generate the query fragment
and bind arguments for the FROM part of the query that retrieves lists
of feature types.  The four positional arguments are as follows:

 $refseq      reference sequence name
 $start       start of region
 $stop        end of region
 $want_count  true to return the count of this feature type

If $want_count is false, the SQL fragment returned must produce a list
of feature types in the format (method, source).

If $want_count is true, the returned fragment must produce a list of
feature types in the format (method, source, count).

=cut

sub make_types_from_part {
  my $self = shift;
  my ($srcseq,$start,$stop,$want_count) = @_;
  my $query = defined($srcseq) || $want_count ? 'fdata,ftype' : 'ftype';
  return $query;
}

=head2 make_types_join_part

 Title   : make_types_join_part
 Usage   : ($string,@args) = $db->make_types_join_part(@args)
 Function: create the JOIN portion of the SQL for fetching features type lists
 Returns : query string and bind arguments
 Args    : see below
 Status  : protected

This method is called by get_types() to generate the query fragment
and bind arguments for the JOIN part of the query that retrieves lists
of feature types.  The four positional arguments are as follows:

 $refseq      reference sequence name
 $start       start of region
 $stop        end of region
 $want_count  true to return the count of this feature type

=cut

sub make_types_join_part {
  my $self = shift;
  my ($srcseq,$start,$stop,$want_count) = @_;
  my $query = defined($srcseq) || $want_count ? 'fdata.ftypeid=ftype.ftypeid'
                                              : '';
  return $query || '1=1';
}

=head2 make_types_where_part

 Title   : make_types_where_part
 Usage   : ($string,@args) = $db->make_types_where_part(@args)
 Function: create the WHERE portion of the SQL for fetching features type lists
 Returns : query string and bind arguments
 Args    : see below
 Status  : protected

This method is called by get_types() to generate the query fragment
and bind arguments for the WHERE part of the query that retrieves
lists of feature types.  The four positional arguments are as follows:

 $refseq      reference sequence name
 $start       start of region
 $stop        end of region
 $want_count  true to return the count of this feature type

=cut

sub make_types_where_part {
  my $self = shift;
  my ($srcseq,$start,$stop,$want_count,$typelist) = @_;
  my (@query,@args);
  if (defined($srcseq)) {
    push @query,'fdata.fref=?';
    push @args,$srcseq;
    if (defined $start or defined $stop) {
      $start = 1           unless defined $start;
      $stop  = MAX_SEGMENT unless defined $stop;
      my ($q,@a) = $self->overlap_query($start,$stop);
      push @query,"($q)";
      push @args,@a;
    }
  }
  if (defined $typelist && @$typelist) {
    my ($q,@a) = $self->types_query($typelist);
    push @query,($q);
    push @args,@a;
  }
  my $query = @query ? join(' AND ',@query) : '1=1';
  return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
}

=head2 make_types_group_part

 Title   : make_types_group_part
 Usage   : ($string,@args) = $db->make_types_group_part(@args)
 Function: create the GROUP BY portion of the SQL for fetching features type lists
 Returns : query string and bind arguments
 Args    : see below
 Status  : protected

This method is called by get_types() to generate the query fragment
and bind arguments for the GROUP BY part of the query that retrieves
lists of feature types.  The four positional arguments are as follows:

 $refseq      reference sequence name
 $start       start of region
 $stop        end of region
 $want_count  true to return the count of this feature type

=cut

sub make_types_group_part {
  my $self = shift;
  my ($srcseq,$start,$stop,$want_count) = @_;
  return unless $srcseq or $want_count;
  return 'ftype.ftypeid,ftype.fmethod,ftype.fsource';
}


=head2 get_feature_id

 Title   : get_feature_id
 Usage   : $integer = $db->get_feature_id($ref,$start,$stop,$typeid,$groupid)
 Function: get the ID of a feature
 Returns : an integer ID or undef
 Args    : none
 Status  : private

This internal method is called by load_gff_line to look up the integer
ID of an existing feature.  It is ony needed when replacing a feature
with new information.

=cut

# this method is called when needed to look up a feature's ID
sub get_feature_id {
  my $self = shift;
  my ($ref,$start,$stop,$typeid,$groupid) = @_;
  my $s = $self->{load_stuff};
  unless ($s->{get_feature_id}) {
    my $dbh = $self->features_db;
    $s->{get_feature_id} =
      $dbh->prepare_delayed('SELECT fid FROM fdata WHERE fref=? AND fstart=? AND fstop=? AND ftypeid=? AND gid=?');
  }
  my $sth = $s->{get_feature_id} or return;
  $sth->execute($ref,$start,$stop,$typeid,$groupid) or return;
  my ($fid) = $sth->fetchrow_array;
  return $fid;
}



=head2 make_abscoord_query

 Title   : make_abscoord_query
 Usage   : $sth = $db->make_abscoord_query($name,$class);
 Function: create query that finds the reference sequence coordinates given a landmark & classa
 Returns : a DBI statement handle
 Args    : name and class of landmark
 Status  : protected

The statement handler should return rows containing five fields:

  1. reference sequence name
  2. reference sequence class
  3. start position
  4. stop position
  5. strand ("+" or "-")

This query always returns "Sequence" as the class of the reference
sequence.

=cut

# given sequence name, return (reference,start,stop,strand)
sub make_abscoord_query {
  my $self = shift;
  my ($name,$class,$refseq) = @_;
  #my $query = GETSEQCOORDS;
  my $query = $self->getseqcoords_query();
  my $getforcedseqcoords = $self->getforcedseqcoords_query() ;
  if ($name =~ /\*/) {
    $name =~ s/%/\\%/g;
    $name =~ s/_/\\_/g;
    $name =~ tr/*/%/;
    $query =~ s/gname=\?/gname LIKE ?/;
  }
  defined $refseq ? $self->dbh->do_query($getforcedseqcoords,$name,$class,$refseq) 
    : $self->dbh->do_query($query,$name,$class);
}

sub make_aliasabscoord_query {
  my $self = shift;
  my ($name,$class) = @_;
  #my $query = GETALIASCOORDS;
  my $query = $self->getaliascoords_query();
  if ($name =~ /\*/) {
    $name =~ s/%/\\%/g;
    $name =~ s/_/\\_/g;
    $name =~ tr/*/%/;
    $query =~ s/gname=\?/gname LIKE ?/;
  }
  $self->dbh->do_query($query,$name,$class);
}

sub getseqcoords_query {
  shift->throw("getseqcoords_query(): must be implemented by a subclass");
}

sub getaliascoords_query {
  shift->throw("getaliascoords_query(): must be implemented by a subclass");
}

sub bin_query {
  my $self = shift;
  my ($start,$stop,$minbin,$maxbin) = @_;
  if ($start && $start < 0 && $stop > 0) {  # split the queries
    my ($lower_query,@lower_args) = $self->_bin_query($start,0,$minbin,$maxbin);
    my ($upper_query,@upper_args) = $self->_bin_query(0,$stop,$minbin,$maxbin);
    my $query = "$lower_query\n\t OR $upper_query";
    my @args  = (@lower_args,@upper_args);
    return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
  } else {
    return $self->_bin_query($start,$stop,$minbin,$maxbin);
  }
}

sub _bin_query {
  my $self = shift;
  my ($start,$stop,$minbin,$maxbin) = @_;
  my ($query,@args);

  $start = 0               unless defined($start);
  $stop  = $self->meta('max_bin') unless defined($stop);

  my @bins;
  $minbin = defined $minbin ? $minbin : $self->min_bin;
  $maxbin = defined $maxbin ? $maxbin : $self->max_bin;
  my $tier = $maxbin;
  while ($tier >= $minbin) {
    my ($tier_start,$tier_stop) = (bin_bot($tier,$start)-EPSILON(),bin_top($tier,$stop)+EPSILON());
    ($tier_start,$tier_stop)    = ($tier_stop,$tier_start) if $tier_start > $tier_stop;  # can happen when working with negative coordinates
    if ($tier_start == $tier_stop) {
      push @bins,'fbin=?';
      push @args,$tier_start;
    } else {
      push @bins,'fbin between ? and ?';
      push @args,($tier_start,$tier_stop);
    }
    $tier /= 10;
  }
  $query = join("\n\t OR ",@bins);
  return wantarray ? ($query,@args)
                   : $self->dbh->dbi_quote($query,@args);
}

# find features that overlap a given range
sub overlap_query {
  my $self = shift;
  my ($start,$stop) = @_;

  my ($query,@args);
  my ($iq,@iargs)   = $self->overlap_query_nobin($start,$stop);
  if (OPTIMIZE) {
    my ($bq,@bargs)   = $self->bin_query($start,$stop);
    $query = "($bq)\n\tAND $iq";
    @args  = (@bargs,@iargs);
  }
  else {
    $query = $iq;
    @args  = @iargs;
  }

  return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
}

# find features that are completely contained within a ranged
sub contains_query {
  my $self = shift;
  my ($start,$stop) = @_;
  my ($bq,@bargs)   = $self->bin_query($start,$stop,undef,bin($start,$stop,$self->min_bin));
  my ($iq,@iargs)   = $self->contains_query_nobin($start,$stop);
  my $query = "($bq)\n\tAND $iq";
  my @args  = (@bargs,@iargs);
  return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
}

# find features that are completely contained within a range
sub contained_in_query {
  my $self = shift;
  my ($start,$stop) = @_;
  my ($bq,@bargs)   = $self->bin_query($start,$stop,abs($stop-$start)+1,undef);
  my ($iq,@iargs)   = $self->contained_in_query_nobin($start,$stop);
  my $query = "($bq)\n\tAND $iq";
  my @args  = (@bargs,@iargs);
  return wantarray ? ($query,@args) : $self->dbh->dbi_quote($query,@args);
}

# implement the _delete_fattribute_to_feature() method
sub _delete_fattribute_to_feature {
  my $self         = shift;
  my @feature_ids  = @_;
  my $dbh          = $self->features_db;
  my $fields       = join ',',map{$dbh->quote($_)} @feature_ids;

  my $query = "delete from fattribute_to_feature where fid in ($fields)";
  warn "$query\n" if $self->debug;
  my $result = $dbh->do($query);
  defined $result or $self->throw($dbh->errstr);
  $result;
}

# implement the _delete_features() method
sub _delete_features {
  my $self = shift;
  my @feature_ids = @_;
  my $dbh          = $self->features_db;
  my $fields       = join ',',map{$dbh->quote($_)} @feature_ids;

  # delete from fattribute_to_feature
  $self->_delete_fattribute_to_feature(@feature_ids);

  my $query = "delete from fdata where fid in ($fields)";
  warn "$query\n" if $self->debug;
  my $result = $dbh->do($query);
  defined $result or $self->throw($dbh->errstr);
  $result;
}

# implement the _delete_groups() method
sub _delete_groups {
  my $self = shift;
  my @group_ids    = @_;
  my $dbh          = $self->features_db;
  my $fields       = join ',',map{$dbh->quote($_)} @group_ids;

  foreach my $gid (@group_ids){
      my @features = $self->get_feature_by_gid($gid);
      $self->delete_features(@features);
  }

  my $query  = "delete from fgroup where gid in ($fields)";
  warn "$query\n" if $self->debug;
  my $result = $dbh->do($query);
  defined $result or $self->throw($dbh->errstr);
  $result;
}

# implement the _delete() method
sub _delete {
  my $self = shift;
  my $delete_spec = shift;
  my $ranges      = $delete_spec->{segments} || [];
  my $types       = $delete_spec->{types}    || [];
  my $force       = $delete_spec->{force};
  my $range_type  = $delete_spec->{range_type};
  my $dbh         = $self->features_db;

  my $query = 'delete from fdata';
  my @where;

  my @range_part;
  for my $segment (@$ranges) {
    my $ref   = $dbh->quote($segment->abs_ref);
    my $start = $segment->abs_start;
    my $stop  = $segment->abs_stop;
    my $range =  $range_type eq 'overlaps'     ? $self->overlap_query($start,$stop)
               : $range_type eq 'contains'     ? $self->contains_query($start,$stop)
	       : $range_type eq 'contained_in' ? $self->contained_in_query($start,$stop)
	       : $self->throw("Invalid range type '$range_type'");
    push @range_part,"(fref=$ref AND $range)";
  }
  push @where,'('. join(' OR ',@range_part).')' if @range_part;

  # get all the types
  if (@$types) {
    my $types_where = $self->types_query($types);
    my $types_query = "select ftypeid from ftype where $types_where";
    my $result      = $dbh->selectall_arrayref($types_query);
    my @typeids     = map {$_->[0]} @$result;
    my $typelist    = join ',',map{$dbh->quote($_)} @typeids;
    $typelist ||= "0"; # don't cause DBI to die with invalid SQL when
                       # unknown feature types were requested.
    push @where,"(ftypeid in ($typelist))";
  }
  $self->throw("This operation would delete all feature data and -force not specified")
    unless @where || $force;
  $query .= " where ".join(' and ',@where) if @where;
  warn "$query\n" if $self->debug;
  my $result = $dbh->do($query);

  defined $result or $self->throw($dbh->errstr);
  $result;
}


=head2 feature_summary

 Title   : feature_summary
 Usage   : $summary = $db->feature_summary(@args)
 Function: returns a coverage summary across indicated region/type
 Returns : a Bio::SeqFeatureI object containing the "coverage" tag
 Args    : see below
 Status  : public

This method is used to get coverage density information across a
region of interest. You provide it with a region of interest, optional
a list of feature types, and a count of the number of bins over which
you want to calculate the coverage density. An object is returned
corresponding to the requested region. It contains a tag called
"coverage" that will return an array ref of "bins" length. Each
element of the array describes the number of features that overlap the
bin at this position.

Arguments:

  Argument       Description
  --------       -----------

  -seq_id        Sequence ID for the region
  -start         Start of region
  -end           End of region
  -type/-types   Feature type of interest or array ref of types
  -bins          Number of bins across region. Defaults to 1000.
  -iterator      Return an iterator across the region

Note that this method uses an approximate algorithm that is only
accurate to 500 bp, so when dealing with bins that are smaller than
1000 bp, you may see some shifting of counts between adjacent bins.

Although an -iterator option is provided, the method only ever returns
a single feature, so this is fairly useless.

=cut


sub feature_summary {
    my $self = shift;
    my ($seq_name,$start,$end,$types,$bins,$iterator) = 
	rearrange([['SEQID','SEQ_ID','REF'],'START',['STOP','END'],
		   ['TYPES','TYPE','PRIMARY_TAG'],
		   'BINS',
		   'ITERATOR',
		  ],@_);
    my ($coverage,$tag) = $self->coverage_array(-seqid=> $seq_name,
						-start=> $start,
						-end  => $end,
						-type => $types,
						-bins => $bins) or return;
    my $score = 0;
    for (@$coverage) { $score += $_ }
    $score /= @$coverage;

    my $feature = Bio::SeqFeature::Lite->new(-seq_id => $seq_name,
					     -start  => $start,
					     -end    => $end,
					     -type   => $tag,
					     -score  => $score,
					     -attributes => 
					     { coverage => [$coverage] });
    return $iterator 
	   ? Bio::DB::GFF::FeatureIterator->new($feature) 
	   : $feature;
}

=head2 coverage_array

 Title   : coverage_array
 Usage   : $arrayref = $db->coverage_array(@args)
 Function: returns a coverage summary across indicated region/type
 Returns : an array reference
 Args    : see below
 Status  : public

This method is used to get coverage density information across a
region of interest. The arguments are identical to feature_summary,
except that instead of returning a Bio::SeqFeatureI object, it returns
an array reference of the desired number of bins. The value of each
element corresponds to the number of features in the bin.

Arguments:

  Argument       Description
  --------       -----------

  -seq_id        Sequence ID for the region
  -start         Start of region
  -end           End of region
  -type/-types   Feature type of interest or array ref of types
  -bins          Number of bins across region. Defaults to 1000.

Note that this method uses an approximate algorithm that is only
accurate to 500 bp, so when dealing with bins that are smaller than
1000 bp, you may see some shifting of counts between adjacent bins.

=cut

sub coverage_array {
    my $self = shift;
    my ($seq_name,$start,$end,$types,$bins) = 
	rearrange([['SEQID','SEQ_ID','REF'],'START',['STOP','END'],
		   ['TYPES','TYPE','PRIMARY_TAG'],'BINS'],@_);

    $types = $self->parse_types($types);
    my $dbh = $self->features_db;
    
    $bins  ||= 1000;
    $start ||= 1;
    unless ($end) {
	my $segment = $self->segment($seq_name) or $self->throw("unknown seq_id $seq_name");
	$end        = $segment->end;
    }

    my $binsize = ($end-$start+1)/$bins;
    my $seqid   = $seq_name;

    return [] unless $seqid;

    # where each bin starts
    my @his_bin_array = map {$start + $binsize * $_}       (0..$bins);
    my @sum_bin_array = map {int(($_-1)/SUMMARY_BIN_SIZE)} @his_bin_array;

    my $interval_stats_table    = 'finterval_stats';
    
    # pick up the type ids
    my ($type_from,@a) = $self->types_query($types);
    my $query          = "select ftypeid,fmethod,fsource from ftype where $type_from";
    my $sth            = $dbh->prepare_delayed($query);
    my (@t,$report_tag);
    $sth->execute(@a);
    while (my ($t,$method,$source) = $sth->fetchrow_array) {
	$report_tag ||= "$method:$source";
	push @t,$t;
    }


    my %bins;
    my $sql = <<END;
SELECT fbin,fcum_count
  FROM $interval_stats_table
  WHERE ftypeid=?
    AND fref=? AND fbin >= ?
  LIMIT 1
END
;
    $sth = $dbh->prepare_delayed($sql) or warn $dbh->errstr;
    eval {
	for my $typeid (@t) {

	    for (my $i=0;$i<@sum_bin_array;$i++) {
		
		my @args = ($typeid,$seqid,$sum_bin_array[$i]);
		$self->_print_query($sql,@args) if $self->debug;
		
		$sth->execute(@args) or $self->throw($sth->errstr);
		my ($bin,$cum_count) = $sth->fetchrow_array;
		push @{$bins{$typeid}},[$bin,$cum_count];
	    }
	}
    };
    return unless %bins;

    my @merged_bins;
    my $firstbin = int(($start-1)/$binsize);
    for my $type (keys %bins) {
	my $arry       = $bins{$type};
	my $last_count = $arry->[0][1];
	my $last_bin   = -1;
	my $i          = 0;
	my $delta;
	for my $b (@$arry) {
	    my ($bin,$count) = @$b;
	    $delta              = $count - $last_count if $bin > $last_bin;
	    $merged_bins[$i++]  = $delta;
	    $last_count         = $count;
	    $last_bin           = $bin;
	}
    }

    return wantarray ? (\@merged_bins,$report_tag) : \@merged_bins;
}


=head2 build_summary_statistics

 Title   : build_summary_statistics
 Usage   : $db->build_summary_statistics
 Function: prepares the table needed to call feature_summary()
 Returns : nothing
 Args    : none
 Status  : public

This method is used to build the summary statistics table that is used
by the feature_summary() and coverage_array() methods. It needs to be
called whenever the database is updated.

=cut

sub build_summary_statistics {
    my $self   = shift;
    my $interval_stats_table = 'finterval_stats';
    my $dbh    = $self->dbh;
    $dbh->begin_work;

    my $sbs = SUMMARY_BIN_SIZE;
    
    my $result = eval {
	$self->_add_interval_stats_table;
	$self->disable_keys($interval_stats_table);
	$dbh->do("DELETE FROM $interval_stats_table");

	my $insert = $dbh->prepare(<<END) or $self->throw($dbh->errstr);
INSERT INTO $interval_stats_table 
           (ftypeid,fref,fbin,fcum_count)
    VALUES (?,?,?,?)
END

;
	    
	my $sql    = 'select ftypeid,fref,fstart,fstop from fdata order by ftypeid,fref,fstart';
	my $select = $dbh->prepare($sql) or $self->throw($dbh->errstr);

	my $current_bin = -1;
	my ($current_type,$current_seqid,$count);
	my $cum_count = 0;
	my (%residuals,$last_bin);

	my $le = -t \*STDERR ? "\r" : "\n";

	$select->execute;

	while (my($typeid,$seqid,$start,$end) = $select->fetchrow_array) {
	    print STDERR $count," features processed$le" if ++$count % 1000 == 0;

	    my $bin = int($start/$sbs);
	    $current_type  ||= $typeid;
	    $current_seqid ||= $seqid;

            # because the input is sorted by start, no more features will contribute to the 
	    # current bin so we can dispose of it
	    if ($bin != $current_bin) {
		if ($seqid != $current_seqid or $typeid != $current_type) {
		    # load all bins left over
		    $self->_load_bins($insert,\%residuals,\$cum_count,$current_type,$current_seqid);
		    %residuals = () ;
		    $cum_count = 0;
		} else {
		    # load all up to current one
		    $self->_load_bins($insert,\%residuals,\$cum_count,$current_type,$current_seqid,$current_bin); 
		}
	    }

	    $last_bin = $current_bin;
	    ($current_seqid,$current_type,$current_bin) = ($seqid,$typeid,$bin);

	    # summarize across entire spanned region
	    my $last_bin = int(($end-1)/$sbs);
	    for (my $b=$bin;$b<=$last_bin;$b++) {
		$residuals{$b}++;
	    }
	}
	# handle tail case
        # load all bins left over	
	$self->_load_bins($insert,\%residuals,\$cum_count,$current_type,$current_seqid);
	$self->enable_keys($interval_stats_table);
	1;
    };
	
    if ($result) { $dbh->commit } else { warn "Can't build summary statistics: $@"; $dbh->rollback };
    print STDERR "\n";
}

sub _load_bins {
    my $self = shift;
    my ($insert,$residuals,$cum_count,$type,$seqid,$stop_after) = @_;
    for my $b (sort {$a<=>$b} keys %$residuals) {
	last if defined $stop_after and $b > $stop_after;
	$$cum_count += $residuals->{$b};
	my @args    = ($type,$seqid,$b,$$cum_count);
	$insert->execute(@args) or warn $insert->errstr;
	delete $residuals->{$b}; # no longer needed
    }
}

sub _add_interval_stats_table {
    my $self              = shift;
    my $schema            = $self->schema;
    my $create_table_stmt = $schema->{'finterval_stats'}{'table'};
    my $dbh               = $self->features_db;
    $dbh->do("drop table finterval_stats");
    $dbh->do($create_table_stmt) ||  warn $dbh->errstr;
}

sub disable_keys { } # noop
sub enable_keys  { } # noop

1;

__END__

=head1 BUGS

Schemas need work to support multiple hierarchical groups.

=head1 SEE ALSO

L<Bio::DB::GFF>, L<bioperl>

=head1 AUTHOR

Lincoln Stein E<lt>lstein@cshl.orgE<gt>.

Copyright (c) 2001 Cold Spring Harbor Laboratory.

This library is free software; you can redistribute it and/or modify
it under the same terms as Perl itself.

=cut