diff --git a/modules/Bio/EnsEMBL/DBSQL/BaseAdaptor.pm b/modules/Bio/EnsEMBL/DBSQL/BaseAdaptor.pm
index c0a61caf9aa81cbb239b85e78d7016d4ca7e96ed..45daae394222f0b810e442e1a5b2c75eca437d79 100755
--- a/modules/Bio/EnsEMBL/DBSQL/BaseAdaptor.pm
+++ b/modules/Bio/EnsEMBL/DBSQL/BaseAdaptor.pm
@@ -95,6 +95,7 @@ use Bio::EnsEMBL::Utils::Exception qw(throw);
use Bio::EnsEMBL::Utils::Scalar qw(assert_ref assert_integer wrap_array);
use DBI qw(:sql_types);
use Data::Dumper;
+use Scalar::Util qw/looks_like_number/;
@ISA = qw(Exporter);
@EXPORT = (@{$DBI::EXPORT_TAGS{'sql_types'}});
@@ -702,8 +703,7 @@ sub fetch_all_by_dbID_list {
# otherwise we'd constantly loop
sub _uncached_fetch_all_by_dbID_list {
my ( $self, $id_list_ref, $slice ) = @_;
-
- return $self->_uncached_fetch_all_by_id_list($id_list_ref, $slice, "dbID");
+ return $self->_uncached_fetch_all_by_id_list($id_list_ref, $slice, "dbID", 1);
} ## end sub fetch_all_by_dbID_list
=head2 _uncached_fetch_all_by_id_list
@@ -715,21 +715,43 @@ sub _uncached_fetch_all_by_dbID_list {
Valid values include dbID and stable_id. dbID is an alias for
the primary key, while other names map directly to table columns
of the Feature this adaptor manages.
+ Arg [4] : Boolean $numeric
+ Indicates if the incoming data is to be processed as a numeric
+ or as a String. If arg [3] was set to dbID then we default this to
+ be true. If arg [3] was set to stable_id then we default this to
+ be false.
+ When not using a standard arg[3] the IDs are assumed to be Strings.
+ Arg [5] : Integer $max_size
+ Control the maximum number of IDs sent to a database in a single
+ query. Defaults to 2K for Strings and 16K for integers. Only
+ provide if you know *exactly* why you need to edit it.
Example : $list_of_features = $adaptor->_uncached_fetch_all_by_id_list(
[qw(ENSG00000101321 ENSG00000101346 ENSG00000101367)],
undef,
- "stable_id");
+ "stable_id", 0); #using strings
+
+ # Numeric set to true because we are using numerics
+ $list_of_features = $adaptor->_uncached_fetch_all_by_id_list(
+ [1,2,3,4],
+ undef,
+ "dbID", 1);
+
+ # Numeric defaults to true because we are querying using dbID
+ $list_of_features = $adaptor->_uncached_fetch_all_by_id_list(
+ [1,2,3,4],
+ undef,
+ "dbID");
Description: This is a generic method used to fetch lists of features by IDs.
It avoids caches, meaning it is best suited for block fetching.
See fetch_all_by_dbID_list() for more info.
- Returntype : ListRef of Bio::EnsEMBL::Feature
+ Returntype : ArrayRef of Bio::EnsEMBL::Feature
Exceptions : Thrown if a list of IDs is not supplied.
Caller : BaseFeatureAdaptor, BaseAdaptor and derived classes.
=cut
sub _uncached_fetch_all_by_id_list {
- my ( $self, $id_list_ref, $slice, $id_type ) = @_;
+ my ( $self, $id_list_ref, $slice, $id_type, $numeric, $max_size ) = @_;
if ( !defined($id_list_ref) || ref($id_list_ref) ne 'ARRAY' ) {
throw("id_list list reference argument is required");
@@ -741,57 +763,74 @@ sub _uncached_fetch_all_by_id_list {
my @tabs = $self->_tables();
my ( $name, $syn ) = @{ $tabs[0] };
- # Ensure that we do not exceed MySQL's max_allowed_packet (defaults to
- # 1 MB) splitting large queries into smaller queries of at most 256 KB
- # (32768 8-bit characters). Assuming a (generous) average dbID string
- # length of 16, this means 2048 dbIDs in each query.
- my $max_size = 2048;
-
- # prepare column name for query
+ # prepare column name for query. If the id_type was
+ # set to dbID then we assume the column must be
+ # tablename_id e.g. gene_id. Otherwise we assume the id_type
+ # is the field/column name
my $field_name;
- if ($id_type eq "dbID") {
- $field_name = $name."_id";
- } else {
- $field_name = $id_type;
+ if($id_type eq 'dbID') {
+ $field_name = $name.'_id';
+ # If numeric was not set default it to 1 since this is an int
+ $numeric = 1 if ! defined $numeric;
+ }
+ elsif($id_type eq 'stable_id') {
+ # If numeric was not set default it to 0 since this is a string
+ $numeric = 0 if ! defined $numeric;
+ $field_name = $id_type;
+ }
+ else {
+ $field_name = $id_type;
+ }
+
+ my $sql_data_type;
+
+ # Ensuring we do not exceed MySQL's max_allowed_packet (defaults to 1MB)
+ # by splitting large queries into smaller queries of at most 256KB
+ # (262,144 8-bit characters)
+ # If we had a numeric then really we are talking about working with
+ # integers. Normal max ensembl id size is 12 plus 2 characters for
+ # commas in our IN statement comes to 14. Even bloating this to 16 gives
+ # a max number of 16,384 IDs (262114/16).
+ #
+ if($numeric) {
+ my $first_id = $id_list_ref->[0];
+ if(!looks_like_number($first_id)) {
+ throw "You specified that we are looking for numerics but $first_id is not a numeric";
+ }
+ $max_size = 16384 if ! defined $max_size;
+ $sql_data_type = SQL_INTEGER;
+ }
+ # However when dealing with Strings those can be very large (assuming
+ # 128 is the max length of a stable ID). 128 is 8x smaller than our
+ # previous max expected integer so we reduce the max ids by 8. This gives
+ # 2048 IDs (16384/8)
+ else {
+ $max_size = 2048 if ! defined $max_size;
+ $sql_data_type = SQL_VARCHAR;
}
# build up unique id list, also validate on the way by
my %id_list;
for (@{$id_list_ref}) {
- $id_list{$_}++;
- if ($id_type ne "stable_id") {
- assert_integer($_,"$field_name");
- }
+ $id_list{$_}++;
}
my @id_list = keys %id_list;
my @out;
-
+ my $inline = 1;
while (@id_list) {
my @ids;
my $id_str;
if ( scalar(@id_list) > $max_size ) {
@ids = splice( @id_list, 0, $max_size );
- } else {
+ }
+ else {
@ids = @id_list;
@id_list = ();
}
-
- if ( scalar(@ids) > 1 ) {
- # stable_ids are the only feature attribute which is expressed as a
- # varchar. These need to be quoted or the SQL will bounce
- if ($id_type eq "stable_id") {
- $id_str = " IN (" . join( ',', map qq("$_"), @ids ) . ")";
- } else {
- $id_str = " IN (" . join( ',', @ids). ")";
- }
- } else {
- $id_str = " = " . $ids[0];
- }
-
- my $constraint = "${syn}.${field_name} $id_str";
-
+ # Push off to our IN statement constructor for this work
+ my $constraint = $self->generate_in_constraint(\@ids, "${syn}.${field_name}", $sql_data_type, $inline);
push @out, @{ $self->generic_fetch($constraint, undef, $slice) };
}