blacklight_oai_provider 0.1.0 → 4.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +18 -0
- data/.rubocop.yml +39 -0
- data/.rubocop_todo.yml +130 -0
- data/.solr_wrapper +8 -0
- data/.travis.yml +16 -0
- data/Gemfile +34 -2
- data/README.md +131 -0
- data/Rakefile +25 -0
- data/VERSION +1 -1
- data/app/assets/{xsl → stylesheets/blacklight_oai_provider}/oai2.xsl +12 -5
- data/app/controllers/concerns/blacklight_oai_provider/controller.rb +38 -0
- data/app/models/concerns/blacklight_oai_provider/solr_document.rb +23 -0
- data/blacklight_oai_provider.gemspec +17 -10
- data/config/initializers/oai_patches.rb +20 -0
- data/lib/blacklight_oai_provider/engine.rb +15 -2
- data/lib/blacklight_oai_provider/exceptions.rb +9 -0
- data/lib/blacklight_oai_provider/resumption_token.rb +51 -0
- data/lib/blacklight_oai_provider/routes.rb +15 -0
- data/lib/blacklight_oai_provider/solr_document_provider.rb +2 -2
- data/lib/blacklight_oai_provider/solr_document_wrapper.rb +55 -27
- data/lib/blacklight_oai_provider/version.rb +1 -2
- data/lib/blacklight_oai_provider.rb +16 -14
- data/lib/generators/blacklight_oai_provider/install_generator.rb +27 -0
- data/lib/railties/blacklight_oai_provider.rake +14 -0
- data/solr/conf/_rest_managed.json +3 -0
- data/solr/conf/admin-extra.html +31 -0
- data/solr/conf/elevate.xml +36 -0
- data/solr/conf/mapping-ISOLatin1Accent.txt +246 -0
- data/solr/conf/protwords.txt +21 -0
- data/solr/conf/schema.xml +629 -0
- data/solr/conf/scripts.conf +24 -0
- data/solr/conf/solrconfig.xml +401 -0
- data/solr/conf/spellings.txt +2 -0
- data/solr/conf/stopwords.txt +58 -0
- data/solr/conf/stopwords_en.txt +58 -0
- data/solr/conf/synonyms.txt +31 -0
- data/solr/conf/xslt/example.xsl +132 -0
- data/solr/conf/xslt/example_atom.xsl +67 -0
- data/solr/conf/xslt/example_rss.xsl +66 -0
- data/solr/conf/xslt/luke.xsl +337 -0
- data/solr/sample_solr_documents.yml +2722 -0
- data/spec/controllers/catalog_controller_spec.rb +39 -0
- data/spec/features/html_rendering_spec.rb +24 -0
- data/spec/models/solr_document_spec.rb +43 -0
- data/spec/requests/get_record_spec.rb +47 -0
- data/spec/requests/identify_spec.rb +53 -0
- data/spec/requests/list_identifiers_spec.rb +80 -0
- data/spec/requests/list_metadata_formats_spec.rb +43 -0
- data/spec/requests/list_records_spec.rb +126 -0
- data/spec/spec_helper.rb +15 -38
- data/spec/test_app_templates/config/solr.yml +9 -0
- data/spec/test_app_templates/lib/generators/test_app_generator.rb +86 -0
- metadata +146 -87
- data/Gemfile.lock +0 -187
- data/README.rdoc +0 -74
- data/config/routes.rb +0 -5
- data/lib/blacklight_oai_provider/README.rdoc +0 -0
- data/lib/blacklight_oai_provider/controller_extension.rb +0 -29
- data/lib/blacklight_oai_provider/route_sets.rb +0 -13
- data/lib/blacklight_oai_provider/solr_document_extension.rb +0 -10
- data/lib/generators/blacklight_oai_provider/blacklight_oai_provider_generator.rb +0 -27
- data/spec/acceptance/blacklight_oai_provider_spec.rb +0 -49
- data/spec/integration/blacklight_stub_spec.rb +0 -10
- data/spec/internal/app/controllers/application_controller.rb +0 -4
- data/spec/internal/app/models/solr_document.rb +0 -3
- data/spec/internal/config/database.yml +0 -3
- data/spec/internal/config/routes.rb +0 -6
- data/spec/internal/config/solr.yml +0 -18
- data/spec/internal/db/combustion_test.sqlite +0 -0
- data/spec/internal/db/schema.rb +0 -53
- data/spec/internal/log/.gitignore +0 -1
- data/spec/internal/public/favicon.ico +0 -0
- data/spec/lib/solr_document_extension_spec.rb +0 -6
- data/spec/vcr_cassettes/solr.yml +0 -113
@@ -0,0 +1,15 @@
|
|
1
|
+
module BlacklightOaiProvider
|
2
|
+
module Routes
|
3
|
+
extend ActiveSupport::Concern
|
4
|
+
|
5
|
+
included do |klass|
|
6
|
+
klass.default_route_sets.insert(klass.default_route_sets.index(:catalog), :oai_routing)
|
7
|
+
end
|
8
|
+
|
9
|
+
def oai_routing
|
10
|
+
add_routes do
|
11
|
+
get "catalog/oai", to: "catalog#oai", as: 'oai_provider'
|
12
|
+
end
|
13
|
+
end
|
14
|
+
end
|
15
|
+
end
|
@@ -1,8 +1,8 @@
|
|
1
1
|
module BlacklightOaiProvider
|
2
2
|
class SolrDocumentProvider < ::OAI::Provider::Base
|
3
3
|
attr_accessor :options
|
4
|
-
def initialize controller, options = {}
|
5
4
|
|
5
|
+
def initialize(controller, options = {})
|
6
6
|
options[:provider] ||= {}
|
7
7
|
options[:document] ||= {}
|
8
8
|
|
@@ -11,7 +11,7 @@ module BlacklightOaiProvider
|
|
11
11
|
options[:repository_name] ||= controller.view_context.send(:application_name)
|
12
12
|
options[:repository_url] ||= controller.view_context.send(:oai_provider_url)
|
13
13
|
|
14
|
-
options[:provider].each do |k,v|
|
14
|
+
options[:provider].each do |k, v|
|
15
15
|
self.class.send k, v
|
16
16
|
end
|
17
17
|
end
|
@@ -1,57 +1,85 @@
|
|
1
1
|
module BlacklightOaiProvider
|
2
2
|
class SolrDocumentWrapper < ::OAI::Provider::Model
|
3
|
-
attr_reader :
|
4
|
-
attr_accessor :options
|
5
|
-
def initialize(controller, options = {})
|
6
|
-
@controller = controller
|
7
|
-
|
8
|
-
defaults = { :timestamp => 'timestamp', :limit => 15}
|
9
|
-
@options = defaults.merge options
|
3
|
+
attr_reader :document_model, :timestamp_field, :solr_timestamp, :limit
|
10
4
|
|
11
|
-
|
12
|
-
@
|
5
|
+
def initialize(controller, options = {})
|
6
|
+
@controller = controller
|
7
|
+
@document_model = options[:model] || ::SolrDocument
|
8
|
+
@solr_timestamp = document_model.timestamp_key
|
9
|
+
@timestamp_field = 'timestamp' # method name used by ruby-oai
|
10
|
+
@limit = options[:limit] || 15
|
13
11
|
end
|
14
12
|
|
15
|
-
def sets
|
16
|
-
end
|
13
|
+
def sets; end
|
17
14
|
|
18
15
|
def earliest
|
19
|
-
|
16
|
+
_response, records = @controller.get_search_results(@controller.params, { fl: solr_timestamp, sort: "#{solr_timestamp} asc", rows: 1 })
|
17
|
+
records.first.timestamp
|
20
18
|
end
|
21
19
|
|
22
20
|
def latest
|
23
|
-
|
21
|
+
_response, records = @controller.get_search_results(@controller.params, { fl: solr_timestamp, sort: "#{solr_timestamp} desc", rows: 1 })
|
22
|
+
records.first.timestamp
|
24
23
|
end
|
25
24
|
|
26
|
-
def find(selector, options={})
|
25
|
+
def find(selector, options = {})
|
27
26
|
return next_set(options[:resumption_token]) if options[:resumption_token]
|
28
27
|
|
29
|
-
if
|
30
|
-
response, records = @controller.get_search_results(@controller.params,
|
28
|
+
if selector == :all
|
29
|
+
response, records = @controller.get_search_results(@controller.params, conditions(options))
|
31
30
|
|
32
|
-
if
|
33
|
-
return select_partial(
|
31
|
+
if limit && response.total > limit
|
32
|
+
return select_partial(BlacklightOaiProvider::ResumptionToken.new(options.merge(last: 0), nil, response.total))
|
34
33
|
end
|
35
|
-
else
|
36
|
-
records = @controller.
|
34
|
+
else
|
35
|
+
_response, records = @controller.get_solr_response_for_doc_id selector.split('/', 2).last
|
37
36
|
end
|
38
37
|
records
|
39
38
|
end
|
40
39
|
|
41
|
-
def select_partial
|
42
|
-
records = @controller.get_search_results(@controller.params,
|
40
|
+
def select_partial(token)
|
41
|
+
_response, records = @controller.get_search_results(@controller.params, token_conditions(token))
|
43
42
|
|
44
|
-
raise ::OAI::ResumptionTokenException
|
43
|
+
raise ::OAI::ResumptionTokenException unless records
|
45
44
|
|
46
|
-
OAI::Provider::PartialResult.new(records, token.next(token.last
|
45
|
+
OAI::Provider::PartialResult.new(records, token.next(token.last + limit))
|
47
46
|
end
|
48
47
|
|
49
48
|
def next_set(token_string)
|
50
|
-
raise ::OAI::ResumptionTokenException
|
49
|
+
raise ::OAI::ResumptionTokenException unless limit
|
51
50
|
|
52
|
-
token =
|
51
|
+
token = BlacklightOaiProvider::ResumptionToken.parse(token_string)
|
53
52
|
select_partial(token)
|
54
53
|
end
|
54
|
+
|
55
|
+
private
|
56
|
+
|
57
|
+
def base_conditions
|
58
|
+
{ sort: "#{solr_timestamp} asc", rows: limit }
|
59
|
+
end
|
60
|
+
|
61
|
+
def token_conditions(token)
|
62
|
+
conditions(token.to_conditions_hash).merge(start: token.last)
|
63
|
+
end
|
64
|
+
|
65
|
+
def conditions(options) # conditions/query derived from options
|
66
|
+
if !(options[:from].blank? && options[:until].blank?)
|
67
|
+
base_conditions.merge(
|
68
|
+
fq: "#{solr_timestamp}:[#{solr_date(options[:from])} TO #{solr_date(options[:until]).gsub('Z', '.999Z')}]"
|
69
|
+
)
|
70
|
+
else
|
71
|
+
base_conditions
|
72
|
+
end
|
73
|
+
end
|
74
|
+
|
75
|
+
def solr_date(time)
|
76
|
+
if time.respond_to?(:xmlschema)
|
77
|
+
time.utc.xmlschema # Force UTC.
|
78
|
+
elsif time.blank?
|
79
|
+
'*'
|
80
|
+
else
|
81
|
+
time.to_s
|
82
|
+
end
|
83
|
+
end
|
55
84
|
end
|
56
85
|
end
|
57
|
-
|
@@ -1,33 +1,35 @@
|
|
1
|
-
# BlacklightOaiProvider
|
2
|
-
|
3
1
|
module BlacklightOaiProvider
|
4
|
-
|
5
|
-
autoload :ControllerExtension, 'blacklight_oai_provider/controller_extension'
|
6
|
-
autoload :SolrDocumentExtension, 'blacklight_oai_provider/solr_document_extension'
|
2
|
+
autoload :Exceptions, 'blacklight_oai_provider/exceptions'
|
7
3
|
autoload :SolrDocumentProvider, 'blacklight_oai_provider/solr_document_provider'
|
8
4
|
autoload :SolrDocumentWrapper, 'blacklight_oai_provider/solr_document_wrapper'
|
9
|
-
autoload :
|
5
|
+
autoload :ResumptionToken, 'blacklight_oai_provider/resumption_token'
|
6
|
+
autoload :Routes, 'blacklight_oai_provider/routes'
|
10
7
|
|
11
8
|
require 'oai'
|
12
9
|
require 'blacklight_oai_provider/version'
|
13
10
|
require 'blacklight_oai_provider/engine'
|
14
|
-
|
11
|
+
|
15
12
|
@omit_inject = {}
|
16
13
|
def self.omit_inject=(value)
|
17
14
|
value = Hash.new(true) if value == true
|
18
|
-
@omit_inject = value
|
15
|
+
@omit_inject = value
|
16
|
+
end
|
17
|
+
|
18
|
+
def self.omit_inject
|
19
|
+
@omit_inject
|
19
20
|
end
|
20
|
-
|
21
|
-
|
21
|
+
|
22
22
|
def self.inject!
|
23
|
-
unless BlacklightOaiProvider.omit_inject[:routes]
|
24
|
-
Blacklight::Routes.send(:include, BlacklightOaiProvider::RouteSets)
|
25
|
-
end
|
23
|
+
Blacklight::Routes.send(:include, BlacklightOaiProvider::Routes) unless BlacklightOaiProvider.omit_inject[:routes]
|
26
24
|
end
|
27
25
|
|
28
26
|
# Add element to array only if it's not already there
|
29
27
|
def self.safe_arr_add(array, element)
|
30
28
|
array << element unless array.include?(element)
|
31
29
|
end
|
32
|
-
|
30
|
+
|
31
|
+
# returns the full path the the blacklight plugin installation
|
32
|
+
def self.root
|
33
|
+
@root ||= File.expand_path(File.dirname(File.dirname(__FILE__)))
|
34
|
+
end
|
33
35
|
end
|
@@ -0,0 +1,27 @@
|
|
1
|
+
require 'rails/generators'
|
2
|
+
|
3
|
+
module BlacklightOaiProvider
|
4
|
+
class InstallGenerator < Rails::Generators::Base
|
5
|
+
argument :model_name, type: :string, default: "SolrDocument"
|
6
|
+
argument :controller_name, type: :string, default: "CatalogController"
|
7
|
+
|
8
|
+
def inject_solr_document_concern
|
9
|
+
file_path = "app/models/#{model_name.underscore}.rb"
|
10
|
+
|
11
|
+
if File.exist? file_path
|
12
|
+
inject_into_file file_path, after: "include Blacklight::Solr::Document" do
|
13
|
+
"\n include BlacklightOaiProvider::SolrDocument\n"
|
14
|
+
end
|
15
|
+
end
|
16
|
+
end
|
17
|
+
|
18
|
+
def inject_catalog_controller_concern
|
19
|
+
file_path = "app/controllers/#{controller_name.underscore}.rb"
|
20
|
+
if File.exist? file_path
|
21
|
+
inject_into_file file_path, after: "include Blacklight::Catalog" do
|
22
|
+
"\n include BlacklightOaiProvider::Controller\n"
|
23
|
+
end
|
24
|
+
end
|
25
|
+
end
|
26
|
+
end
|
27
|
+
end
|
@@ -0,0 +1,14 @@
|
|
1
|
+
namespace :blacklight_oai_provider do
|
2
|
+
# Copied index seeding task from newer versions of Blacklight
|
3
|
+
namespace :index do
|
4
|
+
desc "Put sample data into solr"
|
5
|
+
task seed: [:environment] do
|
6
|
+
require 'yaml'
|
7
|
+
|
8
|
+
docs = YAML.safe_load(File.open(File.join(BlacklightOaiProvider.root, 'solr', 'sample_solr_documents.yml')))
|
9
|
+
conn = RSolr.connect(Blacklight.solr_config)
|
10
|
+
conn.add docs
|
11
|
+
conn.commit
|
12
|
+
end
|
13
|
+
end
|
14
|
+
end
|
@@ -0,0 +1,31 @@
|
|
1
|
+
<!--
|
2
|
+
Licensed to the Apache Software Foundation (ASF) under one or more
|
3
|
+
contributor license agreements. See the NOTICE file distributed with
|
4
|
+
this work for additional information regarding copyright ownership.
|
5
|
+
The ASF licenses this file to You under the Apache License, Version 2.0
|
6
|
+
(the "License"); you may not use this file except in compliance with
|
7
|
+
the License. You may obtain a copy of the License at
|
8
|
+
|
9
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
10
|
+
|
11
|
+
Unless required by applicable law or agreed to in writing, software
|
12
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
13
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
14
|
+
See the License for the specific language governing permissions and
|
15
|
+
limitations under the License.
|
16
|
+
-->
|
17
|
+
|
18
|
+
<!-- The content of this page will be statically included into the top
|
19
|
+
of the admin page. Uncomment this as an example to see there the content
|
20
|
+
will show up.
|
21
|
+
|
22
|
+
<hr>
|
23
|
+
<i>This line will appear before the first table</i>
|
24
|
+
<tr>
|
25
|
+
<td colspan="2">
|
26
|
+
This row will be appended to the end of the first table
|
27
|
+
</td>
|
28
|
+
</tr>
|
29
|
+
<hr>
|
30
|
+
|
31
|
+
-->
|
@@ -0,0 +1,36 @@
|
|
1
|
+
<?xml version="1.0" encoding="UTF-8" ?>
|
2
|
+
<!--
|
3
|
+
Licensed to the Apache Software Foundation (ASF) under one or more
|
4
|
+
contributor license agreements. See the NOTICE file distributed with
|
5
|
+
this work for additional information regarding copyright ownership.
|
6
|
+
The ASF licenses this file to You under the Apache License, Version 2.0
|
7
|
+
(the "License"); you may not use this file except in compliance with
|
8
|
+
the License. You may obtain a copy of the License at
|
9
|
+
|
10
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
11
|
+
|
12
|
+
Unless required by applicable law or agreed to in writing, software
|
13
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
14
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
15
|
+
See the License for the specific language governing permissions and
|
16
|
+
limitations under the License.
|
17
|
+
-->
|
18
|
+
|
19
|
+
<!-- If this file is found in the config directory, it will only be
|
20
|
+
loaded once at startup. If it is found in Solr's data
|
21
|
+
directory, it will be re-loaded every commit.
|
22
|
+
-->
|
23
|
+
|
24
|
+
<elevate>
|
25
|
+
<query text="foo bar">
|
26
|
+
<doc id="1" />
|
27
|
+
<doc id="2" />
|
28
|
+
<doc id="3" />
|
29
|
+
</query>
|
30
|
+
|
31
|
+
<query text="ipod">
|
32
|
+
<doc id="MA147LL/A" /> <!-- put the actual ipod at the top -->
|
33
|
+
<doc id="IW-02" exclude="true" /> <!-- exclude this cable -->
|
34
|
+
</query>
|
35
|
+
|
36
|
+
</elevate>
|
@@ -0,0 +1,246 @@
|
|
1
|
+
# The ASF licenses this file to You under the Apache License, Version 2.0
|
2
|
+
# (the "License"); you may not use this file except in compliance with
|
3
|
+
# the License. You may obtain a copy of the License at
|
4
|
+
#
|
5
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
6
|
+
#
|
7
|
+
# Unless required by applicable law or agreed to in writing, software
|
8
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
9
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
10
|
+
# See the License for the specific language governing permissions and
|
11
|
+
# limitations under the License.
|
12
|
+
|
13
|
+
# Syntax:
|
14
|
+
# "source" => "target"
|
15
|
+
# "source".length() > 0 (source cannot be empty.)
|
16
|
+
# "target".length() >= 0 (target can be empty.)
|
17
|
+
|
18
|
+
# example:
|
19
|
+
# "??" => "A"
|
20
|
+
# "\u00C0" => "A"
|
21
|
+
# "\u00C0" => "\u0041"
|
22
|
+
# "??" => "ss"
|
23
|
+
# "\t" => " "
|
24
|
+
# "\n" => ""
|
25
|
+
|
26
|
+
# ?? => A
|
27
|
+
"\u00C0" => "A"
|
28
|
+
|
29
|
+
# ?? => A
|
30
|
+
"\u00C1" => "A"
|
31
|
+
|
32
|
+
# ?? => A
|
33
|
+
"\u00C2" => "A"
|
34
|
+
|
35
|
+
# ?? => A
|
36
|
+
"\u00C3" => "A"
|
37
|
+
|
38
|
+
# ?? => A
|
39
|
+
"\u00C4" => "A"
|
40
|
+
|
41
|
+
# ?? => A
|
42
|
+
"\u00C5" => "A"
|
43
|
+
|
44
|
+
# ?? => AE
|
45
|
+
"\u00C6" => "AE"
|
46
|
+
|
47
|
+
# ?? => C
|
48
|
+
"\u00C7" => "C"
|
49
|
+
|
50
|
+
# ?? => E
|
51
|
+
"\u00C8" => "E"
|
52
|
+
|
53
|
+
# ?? => E
|
54
|
+
"\u00C9" => "E"
|
55
|
+
|
56
|
+
# ?? => E
|
57
|
+
"\u00CA" => "E"
|
58
|
+
|
59
|
+
# ?? => E
|
60
|
+
"\u00CB" => "E"
|
61
|
+
|
62
|
+
# ?? => I
|
63
|
+
"\u00CC" => "I"
|
64
|
+
|
65
|
+
# ?? => I
|
66
|
+
"\u00CD" => "I"
|
67
|
+
|
68
|
+
# ?? => I
|
69
|
+
"\u00CE" => "I"
|
70
|
+
|
71
|
+
# ?? => I
|
72
|
+
"\u00CF" => "I"
|
73
|
+
|
74
|
+
# ?? => IJ
|
75
|
+
"\u0132" => "IJ"
|
76
|
+
|
77
|
+
# ?? => D
|
78
|
+
"\u00D0" => "D"
|
79
|
+
|
80
|
+
# ?? => N
|
81
|
+
"\u00D1" => "N"
|
82
|
+
|
83
|
+
# ?? => O
|
84
|
+
"\u00D2" => "O"
|
85
|
+
|
86
|
+
# ?? => O
|
87
|
+
"\u00D3" => "O"
|
88
|
+
|
89
|
+
# ?? => O
|
90
|
+
"\u00D4" => "O"
|
91
|
+
|
92
|
+
# ?? => O
|
93
|
+
"\u00D5" => "O"
|
94
|
+
|
95
|
+
# ?? => O
|
96
|
+
"\u00D6" => "O"
|
97
|
+
|
98
|
+
# ?? => O
|
99
|
+
"\u00D8" => "O"
|
100
|
+
|
101
|
+
# ?? => OE
|
102
|
+
"\u0152" => "OE"
|
103
|
+
|
104
|
+
# ??
|
105
|
+
"\u00DE" => "TH"
|
106
|
+
|
107
|
+
# ?? => U
|
108
|
+
"\u00D9" => "U"
|
109
|
+
|
110
|
+
# ?? => U
|
111
|
+
"\u00DA" => "U"
|
112
|
+
|
113
|
+
# ?? => U
|
114
|
+
"\u00DB" => "U"
|
115
|
+
|
116
|
+
# ?? => U
|
117
|
+
"\u00DC" => "U"
|
118
|
+
|
119
|
+
# ?? => Y
|
120
|
+
"\u00DD" => "Y"
|
121
|
+
|
122
|
+
# ?? => Y
|
123
|
+
"\u0178" => "Y"
|
124
|
+
|
125
|
+
# ?? => a
|
126
|
+
"\u00E0" => "a"
|
127
|
+
|
128
|
+
# ?? => a
|
129
|
+
"\u00E1" => "a"
|
130
|
+
|
131
|
+
# ?? => a
|
132
|
+
"\u00E2" => "a"
|
133
|
+
|
134
|
+
# ?? => a
|
135
|
+
"\u00E3" => "a"
|
136
|
+
|
137
|
+
# ?? => a
|
138
|
+
"\u00E4" => "a"
|
139
|
+
|
140
|
+
# ?? => a
|
141
|
+
"\u00E5" => "a"
|
142
|
+
|
143
|
+
# ?? => ae
|
144
|
+
"\u00E6" => "ae"
|
145
|
+
|
146
|
+
# ?? => c
|
147
|
+
"\u00E7" => "c"
|
148
|
+
|
149
|
+
# ?? => e
|
150
|
+
"\u00E8" => "e"
|
151
|
+
|
152
|
+
# ?? => e
|
153
|
+
"\u00E9" => "e"
|
154
|
+
|
155
|
+
# ?? => e
|
156
|
+
"\u00EA" => "e"
|
157
|
+
|
158
|
+
# ?? => e
|
159
|
+
"\u00EB" => "e"
|
160
|
+
|
161
|
+
# ?? => i
|
162
|
+
"\u00EC" => "i"
|
163
|
+
|
164
|
+
# ?? => i
|
165
|
+
"\u00ED" => "i"
|
166
|
+
|
167
|
+
# ?? => i
|
168
|
+
"\u00EE" => "i"
|
169
|
+
|
170
|
+
# ?? => i
|
171
|
+
"\u00EF" => "i"
|
172
|
+
|
173
|
+
# ?? => ij
|
174
|
+
"\u0133" => "ij"
|
175
|
+
|
176
|
+
# ?? => d
|
177
|
+
"\u00F0" => "d"
|
178
|
+
|
179
|
+
# ?? => n
|
180
|
+
"\u00F1" => "n"
|
181
|
+
|
182
|
+
# ?? => o
|
183
|
+
"\u00F2" => "o"
|
184
|
+
|
185
|
+
# ?? => o
|
186
|
+
"\u00F3" => "o"
|
187
|
+
|
188
|
+
# ?? => o
|
189
|
+
"\u00F4" => "o"
|
190
|
+
|
191
|
+
# ?? => o
|
192
|
+
"\u00F5" => "o"
|
193
|
+
|
194
|
+
# ?? => o
|
195
|
+
"\u00F6" => "o"
|
196
|
+
|
197
|
+
# ?? => o
|
198
|
+
"\u00F8" => "o"
|
199
|
+
|
200
|
+
# ?? => oe
|
201
|
+
"\u0153" => "oe"
|
202
|
+
|
203
|
+
# ?? => ss
|
204
|
+
"\u00DF" => "ss"
|
205
|
+
|
206
|
+
# ?? => th
|
207
|
+
"\u00FE" => "th"
|
208
|
+
|
209
|
+
# ?? => u
|
210
|
+
"\u00F9" => "u"
|
211
|
+
|
212
|
+
# ?? => u
|
213
|
+
"\u00FA" => "u"
|
214
|
+
|
215
|
+
# ?? => u
|
216
|
+
"\u00FB" => "u"
|
217
|
+
|
218
|
+
# ?? => u
|
219
|
+
"\u00FC" => "u"
|
220
|
+
|
221
|
+
# ?? => y
|
222
|
+
"\u00FD" => "y"
|
223
|
+
|
224
|
+
# ?? => y
|
225
|
+
"\u00FF" => "y"
|
226
|
+
|
227
|
+
# ??? => ff
|
228
|
+
"\uFB00" => "ff"
|
229
|
+
|
230
|
+
# ??? => fi
|
231
|
+
"\uFB01" => "fi"
|
232
|
+
|
233
|
+
# ??? => fl
|
234
|
+
"\uFB02" => "fl"
|
235
|
+
|
236
|
+
# ??? => ffi
|
237
|
+
"\uFB03" => "ffi"
|
238
|
+
|
239
|
+
# ??? => ffl
|
240
|
+
"\uFB04" => "ffl"
|
241
|
+
|
242
|
+
# ??? => ft
|
243
|
+
"\uFB05" => "ft"
|
244
|
+
|
245
|
+
# ??? => st
|
246
|
+
"\uFB06" => "st"
|
@@ -0,0 +1,21 @@
|
|
1
|
+
# The ASF licenses this file to You under the Apache License, Version 2.0
|
2
|
+
# (the "License"); you may not use this file except in compliance with
|
3
|
+
# the License. You may obtain a copy of the License at
|
4
|
+
#
|
5
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
6
|
+
#
|
7
|
+
# Unless required by applicable law or agreed to in writing, software
|
8
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
9
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
10
|
+
# See the License for the specific language governing permissions and
|
11
|
+
# limitations under the License.
|
12
|
+
|
13
|
+
#-----------------------------------------------------------------------
|
14
|
+
# Use a protected word file to protect against the stemmer reducing two
|
15
|
+
# unrelated words to the same base word.
|
16
|
+
|
17
|
+
# Some non-words that normally won't be encountered,
|
18
|
+
# just to test that they won't be stemmed.
|
19
|
+
dontstems
|
20
|
+
zwhacky
|
21
|
+
|