View | Details | Raw Unified | Return to bug 3216
Collapse All | Expand All

(-)a/Makefile.PL (-3 / +19 lines)
Lines 409-414 my %config_defaults = ( Link Here
409
  'AUTH_INDEX_MODE'   => 'dom',
409
  'AUTH_INDEX_MODE'   => 'dom',
410
  'ZEBRA_MARC_FORMAT' => 'marc21',
410
  'ZEBRA_MARC_FORMAT' => 'marc21',
411
  'ZEBRA_LANGUAGE'    => 'en',
411
  'ZEBRA_LANGUAGE'    => 'en',
412
  'ZEBRA_TOKENIZER'   => 'chr',
412
  'ZEBRA_USER'        => 'kohauser',
413
  'ZEBRA_USER'        => 'kohauser',
413
  'ZEBRA_PASS'        => 'zebrastripes',
414
  'ZEBRA_PASS'        => 'zebrastripes',
414
  'ZEBRA_SRU_HOST'    => 'localhost',
415
  'ZEBRA_SRU_HOST'    => 'localhost',
Lines 460-465 my %valid_config_values = ( Link Here
460
  'AUTH_INDEX_MODE' => { 'grs1' => 1, 'dom' => 1 },
461
  'AUTH_INDEX_MODE' => { 'grs1' => 1, 'dom' => 1 },
461
  'ZEBRA_MARC_FORMAT' => { 'marc21' => 1, 'normarc' => 1, 'unimarc' => 1 }, # FIXME should generate from contents of distributation
462
  'ZEBRA_MARC_FORMAT' => { 'marc21' => 1, 'normarc' => 1, 'unimarc' => 1 }, # FIXME should generate from contents of distributation
462
  'ZEBRA_LANGUAGE'    => { 'en' => 1, 'fr' => 1, 'nb' => 1 }, # FIXME should generate from contents of distribution
463
  'ZEBRA_LANGUAGE'    => { 'en' => 1, 'fr' => 1, 'nb' => 1 }, # FIXME should generate from contents of distribution
464
  'ZEBRA_TOKENIZER' => { chr => 1, icu => 1 },
463
  'RUN_DATABASE_TESTS' => { 'yes' => 1, 'no' => 1 },
465
  'RUN_DATABASE_TESTS' => { 'yes' => 1, 'no' => 1 },
464
  'USE_MEMCACHED'      => { 'yes' => 1, 'no' => 1 },
466
  'USE_MEMCACHED'      => { 'yes' => 1, 'no' => 1 },
465
);
467
);
Lines 486-496 my $pl_files = { Link Here
486
      'rewrite-config.PL' => [
488
      'rewrite-config.PL' => [
487
         'blib/KOHA_CONF_DIR/koha-conf.xml',
489
         'blib/KOHA_CONF_DIR/koha-conf.xml',
488
         'blib/KOHA_CONF_DIR/koha-httpd.conf',
490
         'blib/KOHA_CONF_DIR/koha-httpd.conf',
491
         'blib/ZEBRA_CONF_DIR/etc/default.idx',
489
         'blib/MISC_DIR/koha-install-log'
492
         'blib/MISC_DIR/koha-install-log'
490
         ],
493
         ],
491
	  'fix-perl-path.PL' => [	# this script ensures the correct shebang line for the platform installed on...
494
	  'fix-perl-path.PL' => [	# this script ensures the correct shebang line for the platform installed on...
492
		 'blib'
495
		 'blib'
493
		 ]
496
		 ],
494
};
497
};
495
498
496
if ($config{'INSTALL_ZEBRA'} eq "yes") {
499
if ($config{'INSTALL_ZEBRA'} eq "yes") {
Lines 527-532 if ($config{'INSTALL_MODE'} ne "dev") { Link Here
527
    );
530
    );
528
}
531
}
529
532
533
$config{ZEBRA_TOKENIZER_STMT} = $config{ZEBRA_TOKENIZER} eq 'icu'
534
    ? 'icuchain words-icu.xml'
535
    : 'charmap word-phrase-utf.chr';
536
530
my %test_suite_override_dirs = (
537
my %test_suite_override_dirs = (
531
    KOHA_CONF_DIR  => ['etc'],
538
    KOHA_CONF_DIR  => ['etc'],
532
    ZEBRA_CONF_DIR => ['etc', 'zebradb'],
539
    ZEBRA_CONF_DIR => ['etc', 'zebradb'],
Lines 917-924 Since you've chosen to use Zebra with Koha, Link Here
917
you must specify the primary MARC format of the
924
you must specify the primary MARC format of the
918
records to be indexed by Zebra.
925
records to be indexed by Zebra.
919
926
920
Koha provides Zebra configuration files for MARC 21
927
Koha provides Zebra configuration files for MARC21,
921
and UNIMARC.
928
NORMARC and UNIMARC.
922
929
923
MARC format for Zebra indexing);
930
MARC format for Zebra indexing);
924
        $msg .= _add_valid_values_disp('ZEBRA_MARC_FORMAT', $valid_values);
931
        $msg .= _add_valid_values_disp('ZEBRA_MARC_FORMAT', $valid_values);
Lines 946-951 Authorities indexing mode); Link Here
946
        $config{'AUTH_INDEX_MODE'} = _get_value('AUTH_INDEX_MODE', $msg, $defaults->{'AUTH_INDEX_MODE'}, $valid_values, $install_log_values);
953
        $config{'AUTH_INDEX_MODE'} = _get_value('AUTH_INDEX_MODE', $msg, $defaults->{'AUTH_INDEX_MODE'}, $valid_values, $install_log_values);
947
954
948
        $msg = q(
955
        $msg = q(
956
Zebra has two methods to perform records tokenization
957
and characters normalization: CHR and ICU. ICU is
958
recommended for catalogs containing non-Latin
959
characters.);
960
961
        $msg .= _add_valid_values_disp('ZEBRA_TOKENIZER', $valid_values);
962
        $config{'ZEBRA_TOKENIZER'} = _get_value('ZEBRA_TOKENIZER', $msg, $defaults->{'ZEBRA_TOKENIZER'}, $valid_values, $install_log_values);
963
964
        $msg = q(
949
Please specify Zebra database user);
965
Please specify Zebra database user);
950
        $config{'ZEBRA_USER'} = _get_value('ZEBRA_USER', $msg, $defaults->{'ZEBRA_USER'}, $valid_values, $install_log_values);
966
        $config{'ZEBRA_USER'} = _get_value('ZEBRA_USER', $msg, $defaults->{'ZEBRA_USER'}, $valid_values, $install_log_values);
951
967
(-)a/etc/zebradb/etc/default.idx (-3 / +2 lines)
Lines 10-25 completeness 0 Link Here
10
position 1
10
position 1
11
alwaysmatches 1
11
alwaysmatches 1
12
firstinfield 1
12
firstinfield 1
13
charmap word-phrase-utf.chr
13
__ZEBRA_TOKENIZER_STMT__
14
#firstinfield 1
15
14
16
# Phrase index
15
# Phrase index
17
# Used if completeness is 'complete {sub}field' (@attr 6=2, @attr 6=1)
16
# Used if completeness is 'complete {sub}field' (@attr 6=2, @attr 6=1)
18
# and structure is word/phrase/word-list/free-form-text/document-text
17
# and structure is word/phrase/word-list/free-form-text/document-text
19
index p
18
index p
20
completeness 1
19
completeness 1
21
charmap word-phrase-utf.chr
22
firstinfield 1
20
firstinfield 1
21
__ZEBRA_TOKENIZER_STMT__
23
22
24
# URX (URL) index
23
# URX (URL) index
25
# Used if structure=urx (@attr 4=104)
24
# Used if structure=urx (@attr 4=104)
(-)a/etc/zebradb/etc/words-icu.xml (+12 lines)
Line 0 Link Here
1
<icu_chain locale="">
2
  <transliterate rule="\'>\ "/>
3
  <transliterate rule="[:Number:] { '-' > '' "/>
4
  <transform rule="[:Control:] Any-Remove"/>
5
  <tokenize rule="l"/>
6
  <transform rule="[[:WhiteSpace:][:Punctuation:]] Remove"/>
7
  <transform rule="NFD"/>
8
  <transform rule="[:Nonspacing Mark:] Remove"/>
9
  <transform rule="NFC"/>
10
  <display/>
11
  <casemap rule="l"/>
12
</icu_chain>
(-)a/rewrite-config.PL (-1 / +1 lines)
Lines 120-125 $prefix = $ENV{'INSTALL_BASE'} || "/usr"; Link Here
120
  '__ZEBRA_RUN_DIR__' => "$prefix/var/run/zebradb",
120
  '__ZEBRA_RUN_DIR__' => "$prefix/var/run/zebradb",
121
  '__ZEBRA_MARC_FORMAT__' => 'marc21',
121
  '__ZEBRA_MARC_FORMAT__' => 'marc21',
122
  '__ZEBRA_LANGUAGE__' => 'en',
122
  '__ZEBRA_LANGUAGE__' => 'en',
123
  '__ZEBRA_TOKENIZER_STMT__' => 'charmap word-phrase-utf.chr',
123
  '__ZEBRA_AUTH_CFG__' => 'zebra-authorities.cfg',
124
  '__ZEBRA_AUTH_CFG__' => 'zebra-authorities.cfg',
124
  '__AUTH_RETRIEVAL_CFG__' => 'retrieval-info-auth-grs1.xml',
125
  '__AUTH_RETRIEVAL_CFG__' => 'retrieval-info-auth-grs1.xml',
125
  "__MERGE_SERVER_HOST__" => $myhost,
126
  "__MERGE_SERVER_HOST__" => $myhost,
126
- 

Return to bug 3216