This script can be use to rebuild the zebra DB. It stores all koha MARC records in...

I tried on a 9900 DB, here are the results :

[paul@bureau migration_tools]$ ./rebuild_zebra.pl -c
9900
9903 MARC record done in 37.9104120731354 seconds

[paul@bureau zebra]$ zebraidx update biblios
<snip>
18:31:24-11/08 zebraidx(20348) [log] Iterations . . . 144575
18:31:24-11/08 zebraidx(20348) [log] Distinct words .  39891
18:31:24-11/08 zebraidx(20348) [log] Updates. . . . .     46
18:31:24-11/08 zebraidx(20348) [log] Deletions. . . .      2
18:31:24-11/08 zebraidx(20348) [log] Insertions . . .  39843
18:31:24-11/08 zebraidx(20348) [log] zebra_register_close p=0x8104cf8
18:31:25-11/08 zebraidx(20348) [log] Records:    9887 i/u/d 9881/6/0
18:31:25-11/08 zebraidx(20348) [log] user/system: 531/145
18:31:25-11/08 zebraidx(20348) [log] zebra_stop
18:31:25-11/08 zebraidx(20348) [log] zebraidx times: 11.33  5.31  1.45

diff --git a/misc/migration_tools/rebuild_zebra.pl b/misc/migration_tools/rebuild_zebra.pl
new file mode 100755 (executable)
index 0000000..3a15e37
--- /dev/null
+++ b/misc/migration_tools/rebuild_zebra.pl
@@ -0,0 +1,55 @@
+#!/usr/bin/perl
+# small script that import an iso2709 file into koha 2.0
+
+use strict;
+
+# Koha modules used
+use MARC::File::USMARC;
+use MARC::Record;
+use MARC::Batch;
+use C4::Context;
+use C4::Biblio;
+use Time::HiRes qw(gettimeofday);
+
+use Getopt::Long;
+my ( $input_marc_file, $number) = ('',0);
+my ($confirm);
+GetOptions(
+    'c' => \$confirm,
+);
+
+unless ($confirm) {
+       print <<EOF
+
+script to write files for zebra DB reindexing. Once it's done, run zebraidx update biblios
+
+run the script with -c to confirm the reindexing.
+
+EOF
+;#'
+die;
+}
+
+$|=1; # flushes output
+
+my $dbh = C4::Context->dbh;
+my $cgidir = C4::Context->intranetdir ."/cgi-bin";
+unless (opendir(DIR, "$cgidir")) {
+               $cgidir = C4::Context->intranetdir."/";
+} 
+
+my $starttime = gettimeofday;
+my $sth = $dbh->prepare("select biblionumber from biblio");
+$sth->execute;
+my $i=0;
+while ((my $biblionumber) = $sth->fetchrow) {
+       my $record = MARCgetbiblio($dbh,$biblionumber);
+       my $filename = $cgidir."/zebra/biblios/BIBLIO".$biblionumber."iso2709";
+       open F,"> $filename";
+       print F $record->as_usmarc();
+       close F;
+       $i++;
+       print "\r$i" unless ($i % 100);
+}
+my $timeneeded = gettimeofday - $starttime;
+print "\n$i MARC record done in $timeneeded seconds\n";