dir_revhash is now an importable Util subroutine that is used by both Worker and JobFactory

modules/Bio/EnsEMBL/Hive/RunnableDB/JobFactory.pm

@@ -40,6 +40,8 @@ package Bio::EnsEMBL::Hive::RunnableDB::JobFactory;
 
 use strict;
 use DBI;
+use Bio::EnsEMBL::Hive::Utils ('dir_revhash');  # import dir_revhash
+
 use base ('Bio::EnsEMBL::Hive::ProcessWithParams');
 
 =head2 fetch_input
@@ -67,6 +69,9 @@ sub fetch_input {
     param('key_column'): If every line of your input is a list (it happens, for example, when your SQL returns multiple columns or you have set the 'delimiter' in file/cmd mode)
                          this is the way to say which column is undergoing 'ranging'
 
+    param('hashed_column_number'): if defined, turns 'hashed_column_number' into a dir_revhash and appends it to the list of fields.
+
+
         # The following 4 parameters are mutually exclusive and define the source of ids for the jobs:
 
     param('inputlist');  The list is explicitly given in the parameters, can be abbreviated: 'inputlist' => ['a'..'z']
@@ -89,6 +94,8 @@ sub run {
     my $key_column      = $self->param('key_column')    || 0;
     my $delimiter       = $self->param('delimiter');
 
+    my $hashed_column_number   = $self->param('hashed_column_number');   # skip this step if undefined
+
     my $inputlist       = $self->param('inputlist');
     my $inputfile       = $self->param('inputfile');
     my $inputquery      = $self->param('inputquery');
@@ -104,6 +111,17 @@ sub run {
         _fisher_yates_shuffle_in_place($list);
     }
 
+    if(defined($hashed_column_number) and scalar(@$list)) {
+
+        if(!ref($list->[0])) {
+            $list = [ map { [$_] } @$list ];    # create the second dimension if it was missing
+        }
+
+        foreach my $row (@$list) {
+            push @$row, dir_revhash($row->[$hashed_column_number]);
+        }
+    }
+
     my $output_ids = $self->_split_list_into_ranges($template_hash, $list, $step, $key_column);
     $self->param('output_ids', $output_ids);
 }

modules/Bio/EnsEMBL/Hive/Utils.pm

@@ -44,7 +44,7 @@ use warnings;
 use Data::Dumper;
 
 use Exporter 'import';
-our @EXPORT_OK = qw( stringify destringify );
+our @EXPORT_OK = qw( stringify destringify dir_revhash );
 
 
 =head2 stringify
@@ -96,5 +96,24 @@ sub destringify {
     return $value;
 }
 
+=head2 dir_revhash
+
+    Description: This function takes in a string (which is usually a numeric id) and turns its reverse into a multilevel directory hash.
+                 Please note that no directory is created at this step - it is purely a string conversion function.
+
+    Callers    : Bio::EnsEMBL::Hive::Worker                 # hashing of the worker output directories
+                 Bio::EnsEMBL::Hive::RunnableDB::JobFactory # hashing of an arbitrary id
+
+=cut
+
+sub dir_revhash {
+    my $id = pop @_;
+
+    my @dirs = reverse(split(//, $id));
+    pop @dirs;  # do not use the first digit for hashing
+
+    return join('/', @dirs);
+}
+
 1;
 

modules/Bio/EnsEMBL/Hive/Worker.pm

@@ -78,6 +78,8 @@ use Bio::EnsEMBL::Hive::DBSQL::DataflowRuleAdaptor;
 use Bio::EnsEMBL::Hive::Extensions;
 use Bio::EnsEMBL::Hive::Process;
 
+use Bio::EnsEMBL::Hive::Utils ('dir_revhash');  # import dir_revhash
+
 ## Minimum amount of time in msec that a worker should run before reporting
 ## back to the hive. This is used when setting the batch_size automatically.
 ## 120000 msec = 2 minutes
@@ -324,11 +326,8 @@ sub worker_output_dir {
         } elsif( my $hive_output_dir = $self->hive_output_dir ) {
 
             my $worker_id = $self->worker_id();
-            my @dirs = ( $hive_output_dir, reverse(split(//, $worker_id)) );
-            pop @dirs;  # do not use the first digit for hashing
-            push @dirs, "worker_id_${worker_id}";
 
-            $worker_output_dir = join('/', @dirs);
+            $worker_output_dir = join('/', $hive_output_dir, dir_revhash($worker_id), 'worker_id_'.$worker_id );
         }
 
         if($worker_output_dir) { # will not attempt to create if set to false