-# }
-# now, build item tab !
-# the main difference is that datas are in lines and not in columns : thus, we build the <th> first, then the values...
-# loop through each tag
-# warning : we may have differents number of columns in each row. Thus, we first build a hash, complete it if necessary
-# then construct template.
-# my @fields = $record->fields();
-# my %witness; #---- stores the list of subfields used at least once, with the "meaning" of the code
-# my @big_array;
-# foreach my $field (@fields) {
-# next if ($field->tag()<10);
-# my @subf=$field->subfields;
-# my %this_row;
-# # loop through each subfield
-# for my $i (0..$#subf) {
-# next if ($tagslib->{$field->tag()}->{$subf[$i][0]}->{tab} ne 10);
-# $witness{$subf[$i][0]} = $tagslib->{$field->tag()}->{$subf[$i][0]}->{lib};
-# $this_row{$subf[$i][0]} =$subf[$i][1];
-# }
-# if (%this_row) {
-# push(@big_array, \%this_row);
-# }
-# }
-# #fill big_row with missing datas
-# foreach my $subfield_code (keys(%witness)) {
-# for (my $i=0;$i<=$#big_array;$i++) {
-# $big_array[$i]{$subfield_code}=" " unless ($big_array[$i]{$subfield_code});
-# }
-# }
-# # now, construct template !
-# my @item_value_loop;
-# my @header_value_loop;
-# for (my $i=0;$i<=$#big_array; $i++) {
-# my $items_data;
-# foreach my $subfield_code (keys(%witness)) {
-# $items_data .="<td>".$big_array[$i]{$subfield_code}."</td>";
-# }
-# my %row_data;
-# $row_data{item_value} = $items_data;
-# push(@item_value_loop,\%row_data);
-# }
-# foreach my $subfield_code (keys(%witness)) {
-# my %header_value;
-# $header_value{header_value} = $witness{$subfield_code};
-# push(@header_value_loop, \%header_value);
-# }