Bug 21908: Add DISTINCT biblionumber to rebuild_zebra.pl

Zebra indexing script misc/migration_tools/rebuild_zebra.pl as a table arg to allow filtering.
When using table=items we should use DISTINCT(biblionumber) to avoid indexing several times the same biblio record when it has several items.
This patch adds DISTINCT(biblionumber) in all cases it does not harm if its already unique.

Test plan :
1) Be sur you have a biblio record with biblionumber 1 with 3 items
2) Run misc/migration_tools/rebuild_zebra.pl -v -b --table items --where="biblionumber=1"
3) Without patch you see "Records exported: 3", with patch only one
4) Check indexing works well

Signed-off-by: Pierre-Marc Thibault <pierre-marc.thibault@inLibro.com>
Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>

Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
This commit is contained in:
Fridolin Somers 2018-11-29 10:50:06 +01:00 committed by Nick Clemens
parent c36597bc0c
commit a1b08269a7

View file

@ -476,7 +476,7 @@ sub select_all_authorities {
sub select_all_biblios {
$table = 'biblioitems'
unless grep { /^$table$/ } @tables_allowed_for_select;
my $strsth = qq{ SELECT biblionumber FROM $table };
my $strsth = qq{ SELECT DISTINCT(biblionumber) FROM $table };
$strsth.=qq{ WHERE $where } if ($where);
$strsth.=qq{ LIMIT $length } if ($length && !$offset);
$strsth.=qq{ LIMIT $offset,$length } if ($offset);