[Koha-patches] [PATCH 1/6] bug 7818: update installer for biblio DOM indexing
Galen Charlton
gmc at esilibrary.com
Mon Mar 26 07:44:42 CEST 2012
Adds the necessary bits to enable DOM indexing for bib
records as an option during installation from source.
Signed-off-by: Galen Charlton <gmc at esilibrary.com>
---
Makefile.PL | 22 +++++++++++
etc/koha-conf.xml | 11 ++++-
etc/zebradb/biblios/etc/dom-config-marc.xml | 54 ++++++++++++++++++++++++++
etc/zebradb/biblios/etc/dom-config.xml | 54 ++++++++++++++++++++++++++
etc/zebradb/retrieval-info-bib-dom.xml | 22 +++++++++++
etc/zebradb/retrieval-info-bib-grs1.xml | 40 +++++++++++++++++++
etc/zebradb/zebra-biblios-dom.cfg | 55 +++++++++++++++++++++++++++
misc/koha-install-log | 4 +-
rewrite-config.PL | 3 +
9 files changed, 261 insertions(+), 4 deletions(-)
create mode 100644 etc/zebradb/biblios/etc/dom-config-marc.xml
create mode 100644 etc/zebradb/biblios/etc/dom-config.xml
create mode 100644 etc/zebradb/retrieval-info-bib-dom.xml
create mode 100644 etc/zebradb/retrieval-info-bib-grs1.xml
create mode 100644 etc/zebradb/zebra-biblios-dom.cfg
diff --git a/Makefile.PL b/Makefile.PL
index 2e414d5..c3644f8 100644
--- a/Makefile.PL
+++ b/Makefile.PL
@@ -408,6 +408,7 @@ my %config_defaults = (
'INSTALL_SRU' => 'yes',
'INSTALL_PAZPAR2' => 'no',
'AUTH_INDEX_MODE' => 'dom',
+ 'BIB_INDEX_MODE' => 'dom',
'ZEBRA_MARC_FORMAT' => 'marc21',
'ZEBRA_LANGUAGE' => 'en',
'ZEBRA_TOKENIZER' => 'chr',
@@ -460,6 +461,7 @@ my %valid_config_values = (
'INSTALL_ZEBRA' => { 'yes' => 1, 'no' => 1 },
'INSTALL_SRU' => { 'yes' => 1, 'no' => 1 },
'AUTH_INDEX_MODE' => { 'grs1' => 1, 'dom' => 1 },
+ 'BIB_INDEX_MODE' => { 'grs1' => 1, 'dom' => 1 },
'ZEBRA_MARC_FORMAT' => { 'marc21' => 1, 'normarc' => 1, 'unimarc' => 1 }, # FIXME should generate from contents of distributation
'ZEBRA_LANGUAGE' => { 'en' => 1, 'fr' => 1, 'nb' => 1 }, # FIXME should generate from contents of distribution
'ZEBRA_TOKENIZER' => { chr => 1, icu => 1 },
@@ -501,12 +503,15 @@ if ($config{'INSTALL_ZEBRA'} eq "yes") {
push @{ $pl_files->{'rewrite-config.PL'} }, (
'blib/ZEBRA_CONF_DIR/etc/passwd',
'blib/ZEBRA_CONF_DIR/zebra-biblios.cfg',
+ 'blib/ZEBRA_CONF_DIR/zebra-biblios-dom.cfg',
'blib/ZEBRA_CONF_DIR/zebra-authorities.cfg',
'blib/ZEBRA_CONF_DIR/zebra-authorities-dom.cfg',
'blib/ZEBRA_CONF_DIR/explain-authorities.xml',
'blib/ZEBRA_CONF_DIR/explain-biblios.xml',
'blib/ZEBRA_CONF_DIR/retrieval-info-auth-grs1.xml',
'blib/ZEBRA_CONF_DIR/retrieval-info-auth-dom.xml',
+ 'blib/ZEBRA_CONF_DIR/retrieval-info-bib-grs1.xml',
+ 'blib/ZEBRA_CONF_DIR/retrieval-info-bib-dom.xml',
);
push @{ $pl_files->{'rewrite-config.PL'} }, (
'blib/SCRIPT_DIR/koha-zebra-ctl.sh',
@@ -520,8 +525,11 @@ if ($config{'INSTALL_ZEBRA'} eq "yes") {
);
}
$config{'ZEBRA_AUTH_CFG'} = $config{'AUTH_INDEX_MODE'} eq 'dom' ? 'zebra-authorities-dom.cfg' : 'zebra-authorities.cfg';
+ $config{'ZEBRA_BIB_CFG'} = $config{'BIB_INDEX_MODE'} eq 'dom' ? 'zebra-biblios-dom.cfg' : 'zebra-biblios.cfg';
$config{'AUTH_RETRIEVAL_CFG'} =
$config{'AUTH_INDEX_MODE'} eq 'dom' ? 'retrieval-info-auth-dom.xml' : 'retrieval-info-auth-grs1.xml';
+ $config{'BIB_RETRIEVAL_CFG'} =
+ $config{'BIB_INDEX_MODE'} eq 'dom' ? 'retrieval-info-bib-dom.xml' : 'retrieval-info-bib-grs1.xml';
}
if ($config{'INSTALL_MODE'} ne "dev") {
@@ -942,6 +950,20 @@ Primary language for Zebra indexing);
$msg = q(
Koha can use one of two different indexing modes
+for the MARC bibliographic records:
+
+grs1 - uses the Zebra GRS-1 filter, available
+ for legacy support
+dom - uses the DOM XML filter; offers improved
+ functionality.
+
+Bibliographic indexing mode);
+ $msg .= _add_valid_values_disp('BIB_INDEX_MODE', $valid_values);
+ $config{'BIB_INDEX_MODE'} = _get_value('BIB_INDEX_MODE', $msg, $defaults->{'BIB_INDEX_MODE'}, $valid_values, $install_log_values);
+
+
+ $msg = q(
+Koha can use one of two different indexing modes
for the MARC authorities records:
grs1 - uses the Zebra GRS-1 filter, available
diff --git a/etc/koha-conf.xml b/etc/koha-conf.xml
index f5e2c0f..2cb4e52 100644
--- a/etc/koha-conf.xml
+++ b/etc/koha-conf.xml
@@ -27,7 +27,7 @@ __PAZPAR2_TOGGLE_XML_PRE__
<listen id="mergeserver">tcp:@:__MERGE_SERVER_PORT__</listen>
<server id="mergeserver" listenref="mergeserver">
<directory>__ZEBRA_DATA_DIR__/biblios</directory>
- <config>__ZEBRA_CONF_DIR__/zebra-biblios.cfg</config>
+ <config>__ZEBRA_CONF_DIR__/__ZEBRA_BIB_CFG__</config>
<cql2rpn>__ZEBRA_CONF_DIR__/pqf.properties</cql2rpn>
</server>
__PAZPAR2_TOGGLE_XML_POST__
@@ -35,7 +35,7 @@ __PAZPAR2_TOGGLE_XML_POST__
<!-- BIBLIOGRAPHIC RECORDS -->
<server id="biblioserver" listenref="biblioserver">
<directory>__ZEBRA_DATA_DIR__/biblios</directory>
- <config>__ZEBRA_CONF_DIR__/zebra-biblios.cfg</config>
+ <config>__ZEBRA_CONF_DIR__/__ZEBRA_BIB_CFG__</config>
<cql2rpn>__ZEBRA_CONF_DIR__/pqf.properties</cql2rpn>
<!-- <docpath>xsl</docpath> -->
<!-- <stylesheet>xsl/default.xsl</stylesheet> -->
@@ -182,8 +182,11 @@ __PAZPAR2_TOGGLE_XML_POST__
<!--
<server id="publicserver" listenref="publicserver">
<directory>__ZEBRA_DATA_DIR__/biblios</directory>
- <config>__ZEBRA_CONF_DIR__/zebra-biblios.cfg</config>
+ <config>__ZEBRA_CONF_DIR__/__ZEBRA_BIB_CFG__</config>
<cql2rpn>__ZEBRA_CONF_DIR__/pqf.properties</cql2rpn>
+ <xi:include href="__KOHA_CONF_DIR__/zebradb/__AUTH_RETRIEVAL_CFG__"
+ xmlns:xi="http://www.w3.org/2001/XInclude">
+ <xi:fallback>
<retrievalinfo>
<retrieval syntax="usmarc" name="F"/>
<retrieval syntax="usmarc" name="B"/>
@@ -232,6 +235,8 @@ __PAZPAR2_TOGGLE_XML_POST__
</backend>
</retrieval>
</retrievalinfo>
+ </xi:fallback>
+ </xi:include>
<xi:include href="__KOHA_CONF_DIR__/zebradb/explain-biblios.xml"
xmlns:xi="http://www.w3.org/2001/XInclude">
<xi:fallback>
diff --git a/etc/zebradb/biblios/etc/dom-config-marc.xml b/etc/zebradb/biblios/etc/dom-config-marc.xml
new file mode 100644
index 0000000..b53c68e
--- /dev/null
+++ b/etc/zebradb/biblios/etc/dom-config-marc.xml
@@ -0,0 +1,54 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<!--
+$Id: dom-config.xml,v 1.1 2007-12-13 17:42:28 adam Exp $
+ Copyright (C) 1995-2006
+ Index Data ApS
+
+This file is part of the Zebra server.
+
+Zebra is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 2, or (at your option) any later
+version.
+
+Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with Zebra; see the file LICENSE.zebra. If not, write to the
+Free Software Foundation, 51 Franklin Street, Fifth Floor, Boston,
+MA 02110-1301 USA.
+-->
+
+<!--
+Bibliographic DOM config for MARC ISO2709 input
+-->
+<dom>
+ <extract name="index">
+ <xslt stylesheet="biblio-zebra-indexdefs.xsl"/>
+ </extract>
+ <retrieve name="F">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="usmarc">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="marc">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="marcxml">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="zebra">
+ <xslt stylesheet="zebra.xsl"/>
+ </retrieve>
+ <retrieve name="index">
+ <xslt stylesheet="biblio-zebra-indexdefs.xsl"/>
+ </retrieve>
+ <input>
+ <marc inputcharset="utf-8"/>
+ </input>
+</dom>
diff --git a/etc/zebradb/biblios/etc/dom-config.xml b/etc/zebradb/biblios/etc/dom-config.xml
new file mode 100644
index 0000000..8b3b858
--- /dev/null
+++ b/etc/zebradb/biblios/etc/dom-config.xml
@@ -0,0 +1,54 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<!--
+$Id: dom-config.xml,v 1.1 2007-12-13 17:42:28 adam Exp $
+ Copyright (C) 1995-2006
+ Index Data ApS
+
+This file is part of the Zebra server.
+
+Zebra is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 2, or (at your option) any later
+version.
+
+Zebra is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with Zebra; see the file LICENSE.zebra. If not, write to the
+Free Software Foundation, 51 Franklin Street, Fifth Floor, Boston,
+MA 02110-1301 USA.
+-->
+
+<!--
+Bibliographic DOM config for MARC XML input
+-->
+<dom>
+ <extract name="index">
+ <xslt stylesheet="biblio-zebra-indexdefs.xsl"/>
+ </extract>
+ <retrieve name="F">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="usmarc">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="marc">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="marcxml">
+ <xslt stylesheet="identity.xsl" />
+ </retrieve>
+ <retrieve name="zebra">
+ <xslt stylesheet="zebra.xsl"/>
+ </retrieve>
+ <retrieve name="index">
+ <xslt stylesheet="biblio-zebra-indexdefs.xsl"/>
+ </retrieve>
+ <input>
+ <xmlreader level="0"/>
+ </input>
+</dom>
diff --git a/etc/zebradb/retrieval-info-bib-dom.xml b/etc/zebradb/retrieval-info-bib-dom.xml
new file mode 100644
index 0000000..3093c72
--- /dev/null
+++ b/etc/zebradb/retrieval-info-bib-dom.xml
@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<retrievalinfo xmlns="http://indexdata.com/yaz">
+ <retrieval syntax="usmarc" name="F">
+ <backend syntax="xml" name="marc">
+ <marc inputformat="xml" outputformat="marc"
+ inputcharset="utf-8"
+ outputcharset="utf-8"/>
+ </backend>
+ </retrieval>
+ <retrieval syntax="usmarc" name="B">
+ <backend syntax="xml" name="marc">
+ <marc inputformat="xml" outputformat="marc"
+ inputcharset="utf-8"
+ outputcharset="utf-8"/>
+ </backend>
+ </retrieval>
+ <retrieval syntax="xml" name="index"/> <!-- allow viewing index entries -->
+ <retrieval syntax="xml" name="marc"
+ identifier="info:srw/schema/1/marcxml-v1.1"/>
+ <retrieval syntax="xml" name="marcxml"
+ identifier="info:srw/schema/1/marcxml-v1.1"/>
+</retrievalinfo>
diff --git a/etc/zebradb/retrieval-info-bib-grs1.xml b/etc/zebradb/retrieval-info-bib-grs1.xml
new file mode 100644
index 0000000..58f319d
--- /dev/null
+++ b/etc/zebradb/retrieval-info-bib-grs1.xml
@@ -0,0 +1,40 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<retrievalinfo xmlns="http://indexdata.com/yaz">
+ <retrieval syntax="usmarc" name="F"/>
+ <retrieval syntax="usmarc" name="B"/>
+ <retrieval syntax="xml" name="marcxml"
+ identifier="info:srw/schema/1/marcxml-v1.1">
+ <backend syntax="usmarc" name="F">
+ <marc inputformat="marc" outputformat="marcxml"
+ inputcharset="utf-8"/>
+ </backend>
+ </retrieval>
+ <retrieval syntax="xml" name="dc">
+ <backend syntax="usmarc" name="F">
+ <marc inputformat="marc" outputformat="marcxml"
+ inputcharset="utf-8"/>
+ <xslt stylesheet="__INTRANET_TMPL_DIR__/prog/en/xslt/MARC21slim2DC.xsl"/>
+ </backend>
+ </retrieval>
+ <retrieval syntax="xml" name="mods">
+ <backend syntax="usmarc" name="F">
+ <marc inputformat="marc" outputformat="marcxml"
+ inputcharset="utf-8"/>
+ <xslt stylesheet="__INTRANET_TMPL_DIR__/prog/en/xslt/MARC21slim2MODS.xsl"/>
+ </backend>
+ </retrieval>
+ <retrieval syntax="xml" name="rdfdc">
+ <backend syntax="usmarc" name="F">
+ <marc inputformat="marc" outputformat="marcxml"
+ inputcharset="utf-8"/>
+ <xslt stylesheet="__INTRANET_TMPL_DIR__/prog/en/xslt/MARC21slim2RDFDC.xsl"/>
+ </backend>
+ </retrieval>
+ <retrieval syntax="xml" name="utils">
+ <backend syntax="usmarc" name="F">
+ <marc inputformat="marc" outputformat="marcxml"
+ inputcharset="utf-8"/>
+ <xslt stylesheet="__INTRANET_TMPL_DIR__/prog/en/xslt/MARC21slimUtils.xsl"/>
+ </backend>
+ </retrieval>
+</retrievalinfo>
diff --git a/etc/zebradb/zebra-biblios-dom.cfg b/etc/zebradb/zebra-biblios-dom.cfg
new file mode 100644
index 0000000..a748ddd
--- /dev/null
+++ b/etc/zebradb/zebra-biblios-dom.cfg
@@ -0,0 +1,55 @@
+# Simple Zebra configuration file that defines
+# a database with MARCXML records.
+# $Id: zebra.cfg,v 1.1.2.2 2006/05/09 12:03:16 rangi Exp $
+#
+# Where are the config files located?
+profilePath:__ZEBRA_CONF_DIR__/biblios/etc:__ZEBRA_CONF_DIR__/etc:__ZEBRA_CONF_DIR__/marc_defs/__ZEBRA_MARC_FORMAT__/biblios:__ZEBRA_CONF_DIR__/lang_defs/__ZEBRA_LANGUAGE__:__ZEBRA_CONF_DIR__/xsl
+# modulePath - where to look for loadable zebra modules
+modulePath: /usr/lib/idzebra-2.0/modules
+
+encoding: UTF-8
+# Files that describe the attribute sets supported.
+attset: bib1.att
+attset: explain.att
+attset: gils.att
+
+# systag sysno rank
+
+# Specify record type
+# group .recordType[ .name ]: type
+# type is split into fundamental type. file-read-type . argument
+# http://www.indexdata.dk/zebra/doc/zebra-cfg.tkl
+# http://www.indexdata.dk/zebra/doc/grs.tkl
+
+recordtype: dom.__ZEBRA_CONF_DIR__/biblios/etc/dom-config.xml
+marcxml.recordtype: dom.__ZEBRA_CONF_DIR__/biblios/etc/dom-config.xml
+iso2709.recordtype: dom.__ZEBRA_CONF_DIR__/biblios/etc/dom-config-marc.xml
+
+recordId: (bib1,Local-number)
+storeKeys:1
+storeData:1
+
+
+# Lock File Area
+lockDir: __ZEBRA_LOCK_DIR__/biblios
+perm.anonymous:ar
+perm.__ZEBRA_USER__:rw
+passwd: __ZEBRA_CONF_DIR__/etc/passwd
+register: __ZEBRA_DATA_DIR__/biblios/register:20G
+shadow: __ZEBRA_DATA_DIR__/biblios/shadow:20G
+
+# Temp File area for result sets
+setTmpDir: __ZEBRA_DATA_DIR__/biblios/tmp
+
+# Temp File area for index program
+keyTmpDir: __ZEBRA_DATA_DIR__/biblios/key
+
+# Approx. Memory usage during indexing
+memMax: 50M
+rank:rank-1
+truncmax: 1000000000
+
+# Specifies the maximum number of records that will be sorted in a result set.
+# If the result set contains more than that limit, the records after the limit
+# will not be sorted. If omitted, the default value is 1,000.
+sortmax: 1000
diff --git a/misc/koha-install-log b/misc/koha-install-log
index 6f6a0eb..d842484 100644
--- a/misc/koha-install-log
+++ b/misc/koha-install-log
@@ -44,7 +44,9 @@ ZEBRA_RUN_DIR=__ZEBRA_RUN_DIR__
ZEBRA_MARC_FORMAT=__ZEBRA_MARC_FORMAT__
ZEBRA_LANGUAGE=__ZEBRA_LANGUAGE__
ZEBRA_AUTH_CFG=__ZEBRA_AUTH_CFG__
+ZEBRA_BIB_CFG=__ZEBRA_BIB_CFG__
AUTH_RETRIEVAL_CFG=__AUTH_RETRIEVAL_CFG__
+BIB_RETRIEVAL_CFG=__BIB_RETRIEVAL_CFG__
MERGE_SERVER_HOST=__MERGE_SERVER_HOST__
MERGE_SERVER_PORT=__MERGE_SERVER_PORT__
PAZPAR2_HOST=__PAZPAR2_HOST__
@@ -54,7 +56,7 @@ INSTALL_BASE=__INSTALL_BASE__
INSTALL_ZEBRA=__INSTALL_ZEBRA__
INSTALL_SRU=__INSTALL_SRU__
INSTALL_PAZPAR2=__INSTALL_PAZPAR2__
-AUTH_INDEX_MODE=__AUTH_INDEX_MODE__
+BIB_INDEX_MODE=__BIB_INDEX_MODE__
RUN_DATABASE_TESTS=__RUN_DATABASE_TESTS__
PATH_TO_ZEBRA=__PATH_TO_ZEBRA__
USE_MEMCACHED=__USE_MEMCACHED__
diff --git a/rewrite-config.PL b/rewrite-config.PL
index 967a546..083b3a8 100644
--- a/rewrite-config.PL
+++ b/rewrite-config.PL
@@ -122,7 +122,9 @@ $prefix = $ENV{'INSTALL_BASE'} || "/usr";
'__ZEBRA_LANGUAGE__' => 'en',
'__ZEBRA_TOKENIZER_STMT__' => 'charmap word-phrase-utf.chr',
'__ZEBRA_AUTH_CFG__' => 'zebra-authorities.cfg',
+ '__ZEBRA_BIB_CFG__' => 'zebra-biblios.cfg',
'__AUTH_RETRIEVAL_CFG__' => 'retrieval-info-auth-grs1.xml',
+ '__BIB_RETRIEVAL_CFG__' => 'retrieval-info-bib-grs1.xml',
"__MERGE_SERVER_HOST__" => $myhost,
"__MERGE_SERVER_PORT__" => '11001',
"__PAZPAR2_HOST__" => $myhost,
@@ -135,6 +137,7 @@ $prefix = $ENV{'INSTALL_BASE'} || "/usr";
"__PAZPAR2_TOGGLE_XML_PRE__" => '<!--',
"__PAZPAR2_TOGGLE_XML_POST__" => '-->',
"__AUTH_INDEX_MODE__" => 'grs1',
+ "__BIB_INDEX_MODE__" => 'grs1',
"__RUN_DATABASE_TESTS__" => 'no',
"__PATH_TO_ZEBRA__" => "",
"__USE_MEMCACHED__" => 'no',
--
1.7.2.5
More information about the Koha-patches
mailing list