As Robin points out, the 2.0.59 version of Zebra has another problem as well. When using ICU, words with hyphens aren't tokenized correctly. For instance, "Mont-Royal" would be truncated down to "Mont" both at indexing and search time. So a query for "Mont-Royal" would indeed return the records for "Mont-Royal", but it would also return the records for "Mont-Liban". A query for "Royal" would also return no records for "Mont-Royal" as well. Dreadful. However, IndexData have fixed this problem with Zebra 2.0.60. So... that's a thing. At the moment, the easiest most short-term fix to me is to point to IndexData's Debian repositories, but I don't think that'll be acceptable as a long-term solution. We mostly use OpenSuse, and I managed to get them to add Zebra 2.0.60 to their stable repositories, so we're OK for now. Now that Jessie has been released, perhaps there's a higher chance of the fix getting into Debian stable as well? David Cook Systems Librarian Prosentient Systems 72/330 Wattle St, Ultimo, NSW 2007
-----Original Message----- From: koha-devel-bounces@lists.koha-community.org [mailto:koha-devel- bounces@lists.koha-community.org] On Behalf Of Robin Sheat Sent: Friday, 1 May 2015 9:03 AM To: koha-devel@lists.koha-community.org Subject: Re: [Koha-devel] Problems with Indexdata idzebra-2.0 build (Zebra 2.0.59)
Tomas Cohen Arazi schreef op do 30-04-2015 om 10:23 [-0300]:
It seems (from jenkins output) that search is broken on GRS-1 setups on Jessie. Which we don't really care as we dropped support for GRS-1 on 3.18.
There is something else wrong with the jessie version, too.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=777515
-- Robin Sheat Catalyst IT Ltd. ✆ +64 4 803 2204 GPG: 5FA7 4B49 1E4D CAA4 4C38 8505 77F5 B724 F871 3BDF