rsolr 1.0.8 → 1.1.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +7 -0
- data/.travis.yml +21 -0
- data/CHANGES.txt +20 -1
- data/Gemfile +3 -9
- data/README.rdoc +34 -31
- data/Rakefile +1 -10
- data/lib/rsolr.rb +14 -10
- data/lib/rsolr/char.rb +4 -1
- data/lib/rsolr/client.rb +133 -54
- data/lib/rsolr/connection.rb +21 -17
- data/lib/rsolr/error.rb +39 -20
- data/lib/rsolr/response.rb +57 -13
- data/lib/rsolr/uri.rb +44 -27
- data/lib/rsolr/version.rb +7 -0
- data/lib/rsolr/xml.rb +69 -15
- data/rsolr.gemspec +14 -7
- data/spec/api/char_spec.rb +8 -3
- data/spec/api/client_spec.rb +221 -87
- data/spec/api/connection_spec.rb +93 -29
- data/spec/api/error_spec.rb +22 -32
- data/spec/api/pagination_spec.rb +12 -5
- data/spec/api/rsolr_spec.rb +28 -8
- data/spec/api/uri_spec.rb +108 -50
- data/spec/api/xml_spec.rb +196 -96
- data/spec/fixtures/basic_configs/_rest_managed.json +1 -0
- data/spec/fixtures/basic_configs/currency.xml +67 -0
- data/spec/fixtures/basic_configs/lang/stopwords_en.txt +54 -0
- data/spec/fixtures/basic_configs/protwords.txt +21 -0
- data/spec/fixtures/basic_configs/schema.xml +530 -0
- data/spec/fixtures/basic_configs/solrconfig.xml +572 -0
- data/spec/fixtures/basic_configs/stopwords.txt +14 -0
- data/spec/fixtures/basic_configs/synonyms.txt +29 -0
- data/spec/integration/solr5_spec.rb +26 -0
- data/spec/spec_helper.rb +8 -1
- data/tasks/spec.rake +1 -38
- metadata +98 -34
- data/tasks/rcov.rake +0 -25
data/spec/api/xml_spec.rb
CHANGED
@@ -1,121 +1,221 @@
|
|
1
1
|
require 'spec_helper'
|
2
|
+
require 'builder'
|
3
|
+
require 'nokogiri'
|
2
4
|
describe "RSolr::Xml" do
|
3
5
|
|
4
6
|
let(:generator){ RSolr::Xml::Generator.new }
|
7
|
+
|
8
|
+
builder_engines = {
|
9
|
+
:builder => { :val => false, :class => Builder::XmlMarkup, :engine => Builder::XmlMarkup.new(:indent => 0, :margin => 0, :encoding => 'UTF-8') },
|
10
|
+
:nokogiri => { :val => true, :class => Nokogiri::XML::Builder, :engine => Nokogiri::XML::Builder.new }
|
11
|
+
}
|
12
|
+
|
13
|
+
[:builder,:nokogiri].each do |engine_name|
|
14
|
+
describe engine_name do
|
15
|
+
before :all do
|
16
|
+
@engine = builder_engines[engine_name]
|
17
|
+
@old_ng_setting = RSolr::Xml::Generator.use_nokogiri
|
18
|
+
RSolr::Xml::Generator.use_nokogiri = @engine[:val]
|
19
|
+
end
|
20
|
+
|
21
|
+
after :all do
|
22
|
+
RSolr::Xml::Generator.use_nokogiri = @old_ng_setting
|
23
|
+
end
|
5
24
|
|
6
|
-
|
7
|
-
|
8
|
-
|
9
|
-
|
10
|
-
|
11
|
-
|
12
|
-
|
13
|
-
|
14
|
-
|
25
|
+
before :each do
|
26
|
+
builder_engines.each_pair do |name,spec|
|
27
|
+
expect(spec[:class]).not_to receive(:new) unless name == engine_name
|
28
|
+
end
|
29
|
+
end
|
30
|
+
|
31
|
+
context :xml_engine do
|
32
|
+
it "should use #{engine_name}" do
|
33
|
+
expect(@engine[:class]).to receive(:new).and_return(@engine[:engine])
|
34
|
+
generator.send(:commit)
|
35
|
+
end
|
36
|
+
end
|
37
|
+
|
38
|
+
# call all of the simple methods...
|
39
|
+
# make sure the xml string is valid
|
40
|
+
# ensure the class is actually Solr::XML
|
41
|
+
[:optimize, :rollback, :commit].each do |meth|
|
42
|
+
it "#{meth} should generator xml" do
|
43
|
+
result = generator.send(meth)
|
44
|
+
expect(result).to eq("<?xml version=\"1.0\" encoding=\"UTF-8\"?><#{meth}/>")
|
45
|
+
end
|
46
|
+
end
|
47
|
+
|
48
|
+
context :add do
|
49
|
+
|
50
|
+
it 'should yield a Message::Document object when #add is called with a block' do
|
51
|
+
documents = [{:id=>1, :name=>'sam', :cat=>['cat 1', 'cat 2']}]
|
52
|
+
add_attrs = {:boost=>200.00}
|
53
|
+
result = generator.add(documents, add_attrs) do |doc|
|
54
|
+
doc.field_by_name(:name).attrs[:boost] = 10
|
55
|
+
expect(doc.fields.size).to eq(4)
|
56
|
+
expect(doc.fields_by_name(:cat).size).to eq(2)
|
57
|
+
end
|
58
|
+
expect(result).to match(%r(name="cat">cat 1</field>))
|
59
|
+
expect(result).to match(%r(name="cat">cat 2</field>))
|
60
|
+
expect(result).to match(%r(<add boost="200.0">))
|
61
|
+
expect(result).to match(%r(boost="10"))
|
62
|
+
expect(result).to match(%r(<field name="id">1</field>))
|
63
|
+
end
|
64
|
+
|
65
|
+
it 'should work for values that yield enumerators' do
|
66
|
+
documents = [{id: 1, cat: ['cat 1', 'cat 2'].to_enum}]
|
67
|
+
result = generator.add(documents)
|
68
|
+
|
69
|
+
expect(result).to match(%r(name="cat">cat 1</field>))
|
70
|
+
expect(result).to match(%r(name="cat">cat 2</field>))
|
71
|
+
end
|
72
|
+
|
73
|
+
# add a single hash ("doc")
|
74
|
+
it 'should create an add from a hash' do
|
75
|
+
data = {
|
76
|
+
:id=>1,
|
77
|
+
:name=>'matt'
|
78
|
+
}
|
79
|
+
result = generator.add(data)
|
80
|
+
expect(result).to match(/<field name="name">matt<\/field>/)
|
81
|
+
expect(result).to match(/<field name="id">1<\/field>/)
|
82
|
+
end
|
83
|
+
|
84
|
+
# add an array of hashes
|
85
|
+
it 'should create many adds from an array of hashes' do
|
86
|
+
data = [
|
87
|
+
{
|
88
|
+
:id=>1,
|
89
|
+
:name=>'matt'
|
90
|
+
},
|
91
|
+
{
|
92
|
+
:id=>2,
|
93
|
+
:name=>'sam'
|
94
|
+
}
|
95
|
+
]
|
96
|
+
message = generator.add(data)
|
97
|
+
expected = "<?xml version=\"1.0\" encoding=\"UTF-8\"?><add><doc><field name=\"id\">1</field><field name=\"name\">matt</field></doc><doc><field name=\"id\">2</field><field name=\"name\">sam</field></doc></add>"
|
98
|
+
expect(message).to match(/<field name="name">matt<\/field>/)
|
99
|
+
expect(message).to match(/<field name="name">sam<\/field>/)
|
100
|
+
end
|
101
|
+
|
102
|
+
# multiValue field support test, thanks to Fouad Mardini!
|
103
|
+
it 'should create multiple fields from array values' do
|
104
|
+
data = {
|
105
|
+
:id => 1,
|
106
|
+
:name => ['matt1', 'matt2']
|
107
|
+
}
|
108
|
+
result = generator.add(data)
|
109
|
+
expect(result).to match(/<field name="name">matt1<\/field>/)
|
110
|
+
expect(result).to match(/<field name="name">matt2<\/field>/)
|
111
|
+
end
|
112
|
+
|
113
|
+
it 'should allow for objects which can be casted to an array' do
|
114
|
+
name = double("name", to_ary: ['matt1', 'matt2'])
|
115
|
+
data = {
|
116
|
+
:id => 1,
|
117
|
+
:name => name
|
118
|
+
}
|
119
|
+
result = generator.add(data)
|
120
|
+
expect(result).to match(/<field name="name">matt1<\/field>/)
|
121
|
+
expect(result).to match(/<field name="name">matt2<\/field>/)
|
122
|
+
end
|
123
|
+
|
124
|
+
it 'should create an add from a single Message::Document' do
|
125
|
+
document = RSolr::Xml::Document.new
|
126
|
+
document.add_field('id', 1)
|
127
|
+
document.add_field('name', 'matt', :boost => 2.0)
|
128
|
+
result = generator.add(document)
|
129
|
+
expect(result).to match(Regexp.escape('<?xml version="1.0" encoding="UTF-8"?>'))
|
130
|
+
expect(result).to match(/<field name="id">1<\/field>/)
|
131
|
+
expect(result).to match Regexp.escape('boost="2.0"')
|
132
|
+
expect(result).to match Regexp.escape('name="name"')
|
133
|
+
expect(result).to match Regexp.escape('matt</field>')
|
134
|
+
end
|
135
|
+
|
136
|
+
it 'should create adds from multiple Message::Documents' do
|
137
|
+
documents = (1..2).map do |i|
|
138
|
+
doc = RSolr::Xml::Document.new
|
139
|
+
doc.add_field('id', i)
|
140
|
+
doc.add_field('name', "matt#{i}")
|
141
|
+
doc
|
142
|
+
end
|
143
|
+
result = generator.add(documents)
|
144
|
+
expect(result).to match(/<field name="name">matt1<\/field>/)
|
145
|
+
expect(result).to match(/<field name="name">matt2<\/field>/)
|
146
|
+
end
|
147
|
+
|
148
|
+
end
|
15
149
|
|
16
|
-
|
150
|
+
context :delete_by_id do
|
151
|
+
|
152
|
+
it 'should create a doc id delete' do
|
153
|
+
expect(generator.delete_by_id(10)).to eq("<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><id>10</id></delete>")
|
154
|
+
end
|
155
|
+
|
156
|
+
it 'should create many doc id deletes' do
|
157
|
+
expect(generator.delete_by_id([1, 2, 3])).to eq("<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><id>1</id><id>2</id><id>3</id></delete>")
|
158
|
+
end
|
17
159
|
|
18
|
-
it 'should yield a Message::Document object when #add is called with a block' do
|
19
|
-
documents = [{:id=>1, :name=>'sam', :cat=>['cat 1', 'cat 2']}]
|
20
|
-
add_attrs = {:boost=>200.00}
|
21
|
-
result = generator.add(documents, add_attrs) do |doc|
|
22
|
-
doc.field_by_name(:name).attrs[:boost] = 10
|
23
|
-
doc.fields.size.should == 4
|
24
|
-
doc.fields_by_name(:cat).size.should == 2
|
25
160
|
end
|
26
|
-
|
27
|
-
|
28
|
-
|
29
|
-
|
30
|
-
|
31
|
-
end
|
161
|
+
|
162
|
+
context :delete_by_query do
|
163
|
+
it 'should create a query delete' do
|
164
|
+
expect(generator.delete_by_query('status:"LOST"')).to eq("<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><query>status:\"LOST\"</query></delete>")
|
165
|
+
end
|
32
166
|
|
33
|
-
|
34
|
-
|
167
|
+
it 'should create many query deletes' do
|
168
|
+
expect(generator.delete_by_query(['status:"LOST"', 'quantity:0'])).to eq("<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><query>status:\"LOST\"</query><query>quantity:0</query></delete>")
|
169
|
+
end
|
170
|
+
end
|
171
|
+
|
172
|
+
end
|
173
|
+
end
|
174
|
+
|
175
|
+
context :formatting do
|
176
|
+
it 'should format date objects into ISO 8601' do
|
35
177
|
data = {
|
36
|
-
:
|
37
|
-
:name=>'matt'
|
178
|
+
dt: Date.new(1992, 03, 15)
|
38
179
|
}
|
39
180
|
result = generator.add(data)
|
40
|
-
result.
|
41
|
-
result.should match(/<field name="id">1<\/field>/)
|
181
|
+
expect(result).to match(/<field name="dt">1992-03-15T00:00:00Z<\/field>/)
|
42
182
|
end
|
43
183
|
|
44
|
-
|
45
|
-
|
46
|
-
|
47
|
-
|
48
|
-
|
49
|
-
|
50
|
-
},
|
51
|
-
{
|
52
|
-
:id=>2,
|
53
|
-
:name=>'sam'
|
54
|
-
}
|
55
|
-
]
|
56
|
-
message = generator.add(data)
|
57
|
-
expected = "<?xml version=\"1.0\" encoding=\"UTF-8\"?><add><doc><field name=\"id\">1</field><field name=\"name\">matt</field></doc><doc><field name=\"id\">2</field><field name=\"name\">sam</field></doc></add>"
|
58
|
-
message.should match(/<field name="name">matt<\/field>/)
|
59
|
-
message.should match(/<field name="name">sam<\/field>/)
|
184
|
+
it 'should format time objects into ISO 8601' do
|
185
|
+
data = {
|
186
|
+
dt: Time.new(1992, 03, 15, 16, 23, 55, 3600)
|
187
|
+
}
|
188
|
+
result = generator.add(data)
|
189
|
+
expect(result).to match(/<field name="dt">1992-03-15T15:23:55Z<\/field>/)
|
60
190
|
end
|
61
191
|
|
62
|
-
|
63
|
-
it 'should create multiple fields from array values' do
|
192
|
+
it 'should format datetime objects into ISO 8601' do
|
64
193
|
data = {
|
65
|
-
:
|
66
|
-
:name => ['matt1', 'matt2']
|
194
|
+
dt: DateTime.new(1992, 03, 15, 16, 23, 55, '+1')
|
67
195
|
}
|
68
196
|
result = generator.add(data)
|
69
|
-
result.
|
70
|
-
result.should match(/<field name="name">matt2<\/field>/)
|
197
|
+
expect(result).to match(/<field name="dt">1992-03-15T15:23:55Z<\/field>/)
|
71
198
|
end
|
72
199
|
|
73
|
-
it '
|
74
|
-
|
75
|
-
|
76
|
-
|
77
|
-
|
78
|
-
result
|
79
|
-
result.
|
80
|
-
result.should match Regexp.escape('boost="2.0"')
|
81
|
-
result.should match Regexp.escape('name="name"')
|
82
|
-
result.should match Regexp.escape('matt</field>')
|
200
|
+
it 'passes through other values' do
|
201
|
+
data = {
|
202
|
+
whatever: 'some string'
|
203
|
+
}
|
204
|
+
|
205
|
+
result = generator.add(data)
|
206
|
+
expect(result).to match(/<field name="whatever">some string<\/field>/)
|
83
207
|
end
|
84
|
-
|
85
|
-
|
86
|
-
|
87
|
-
|
88
|
-
|
89
|
-
|
90
|
-
|
208
|
+
|
209
|
+
# rails monkey-patches String to add a #to_time casting..
|
210
|
+
context 'with rails monkey patching' do
|
211
|
+
it 'passes through string values' do
|
212
|
+
data = {
|
213
|
+
whatever: double(to_s: 'some string', to_time: nil)
|
214
|
+
}
|
215
|
+
|
216
|
+
result = generator.add(data)
|
217
|
+
expect(result).to match(/<field name="whatever">some string<\/field>/)
|
91
218
|
end
|
92
|
-
result = generator.add(documents)
|
93
|
-
result.should match(/<field name="name">matt1<\/field>/)
|
94
|
-
result.should match(/<field name="name">matt2<\/field>/)
|
95
219
|
end
|
96
|
-
|
97
|
-
end
|
98
|
-
|
99
|
-
context :delete_by_id do
|
100
|
-
|
101
|
-
it 'should create a doc id delete' do
|
102
|
-
generator.delete_by_id(10).should == "<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><id>10</id></delete>"
|
103
|
-
end
|
104
|
-
|
105
|
-
it 'should create many doc id deletes' do
|
106
|
-
generator.delete_by_id([1, 2, 3]).should == "<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><id>1</id><id>2</id><id>3</id></delete>"
|
107
|
-
end
|
108
|
-
|
109
220
|
end
|
110
|
-
|
111
|
-
context :delete_by_query do
|
112
|
-
it 'should create a query delete' do
|
113
|
-
generator.delete_by_query('status:"LOST"').should == "<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><query>status:\"LOST\"</query></delete>"
|
114
|
-
end
|
115
|
-
|
116
|
-
it 'should create many query deletes' do
|
117
|
-
generator.delete_by_query(['status:"LOST"', 'quantity:0']).should == "<?xml version=\"1.0\" encoding=\"UTF-8\"?><delete><query>status:\"LOST\"</query><query>quantity:0</query></delete>"
|
118
|
-
end
|
119
|
-
end
|
120
|
-
|
121
|
-
end
|
221
|
+
end
|
@@ -0,0 +1 @@
|
|
1
|
+
{"initArgs":{},"managedList":[]}
|
@@ -0,0 +1,67 @@
|
|
1
|
+
<?xml version="1.0" ?>
|
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
|
+
<!-- Example exchange rates file for CurrencyField type named "currency" in example schema -->
|
20
|
+
|
21
|
+
<currencyConfig version="1.0">
|
22
|
+
<rates>
|
23
|
+
<!-- Updated from http://www.exchangerate.com/ at 2011-09-27 -->
|
24
|
+
<rate from="USD" to="ARS" rate="4.333871" comment="ARGENTINA Peso" />
|
25
|
+
<rate from="USD" to="AUD" rate="1.025768" comment="AUSTRALIA Dollar" />
|
26
|
+
<rate from="USD" to="EUR" rate="0.743676" comment="European Euro" />
|
27
|
+
<rate from="USD" to="BRL" rate="1.881093" comment="BRAZIL Real" />
|
28
|
+
<rate from="USD" to="CAD" rate="1.030815" comment="CANADA Dollar" />
|
29
|
+
<rate from="USD" to="CLP" rate="519.0996" comment="CHILE Peso" />
|
30
|
+
<rate from="USD" to="CNY" rate="6.387310" comment="CHINA Yuan" />
|
31
|
+
<rate from="USD" to="CZK" rate="18.47134" comment="CZECH REP. Koruna" />
|
32
|
+
<rate from="USD" to="DKK" rate="5.515436" comment="DENMARK Krone" />
|
33
|
+
<rate from="USD" to="HKD" rate="7.801922" comment="HONG KONG Dollar" />
|
34
|
+
<rate from="USD" to="HUF" rate="215.6169" comment="HUNGARY Forint" />
|
35
|
+
<rate from="USD" to="ISK" rate="118.1280" comment="ICELAND Krona" />
|
36
|
+
<rate from="USD" to="INR" rate="49.49088" comment="INDIA Rupee" />
|
37
|
+
<rate from="USD" to="XDR" rate="0.641358" comment="INTNL MON. FUND SDR" />
|
38
|
+
<rate from="USD" to="ILS" rate="3.709739" comment="ISRAEL Sheqel" />
|
39
|
+
<rate from="USD" to="JPY" rate="76.32419" comment="JAPAN Yen" />
|
40
|
+
<rate from="USD" to="KRW" rate="1169.173" comment="KOREA (SOUTH) Won" />
|
41
|
+
<rate from="USD" to="KWD" rate="0.275142" comment="KUWAIT Dinar" />
|
42
|
+
<rate from="USD" to="MXN" rate="13.85895" comment="MEXICO Peso" />
|
43
|
+
<rate from="USD" to="NZD" rate="1.285159" comment="NEW ZEALAND Dollar" />
|
44
|
+
<rate from="USD" to="NOK" rate="5.859035" comment="NORWAY Krone" />
|
45
|
+
<rate from="USD" to="PKR" rate="87.57007" comment="PAKISTAN Rupee" />
|
46
|
+
<rate from="USD" to="PEN" rate="2.730683" comment="PERU Sol" />
|
47
|
+
<rate from="USD" to="PHP" rate="43.62039" comment="PHILIPPINES Peso" />
|
48
|
+
<rate from="USD" to="PLN" rate="3.310139" comment="POLAND Zloty" />
|
49
|
+
<rate from="USD" to="RON" rate="3.100932" comment="ROMANIA Leu" />
|
50
|
+
<rate from="USD" to="RUB" rate="32.14663" comment="RUSSIA Ruble" />
|
51
|
+
<rate from="USD" to="SAR" rate="3.750465" comment="SAUDI ARABIA Riyal" />
|
52
|
+
<rate from="USD" to="SGD" rate="1.299352" comment="SINGAPORE Dollar" />
|
53
|
+
<rate from="USD" to="ZAR" rate="8.329761" comment="SOUTH AFRICA Rand" />
|
54
|
+
<rate from="USD" to="SEK" rate="6.883442" comment="SWEDEN Krona" />
|
55
|
+
<rate from="USD" to="CHF" rate="0.906035" comment="SWITZERLAND Franc" />
|
56
|
+
<rate from="USD" to="TWD" rate="30.40283" comment="TAIWAN Dollar" />
|
57
|
+
<rate from="USD" to="THB" rate="30.89487" comment="THAILAND Baht" />
|
58
|
+
<rate from="USD" to="AED" rate="3.672955" comment="U.A.E. Dirham" />
|
59
|
+
<rate from="USD" to="UAH" rate="7.988582" comment="UKRAINE Hryvnia" />
|
60
|
+
<rate from="USD" to="GBP" rate="0.647910" comment="UNITED KINGDOM Pound" />
|
61
|
+
|
62
|
+
<!-- Cross-rates for some common currencies -->
|
63
|
+
<rate from="EUR" to="GBP" rate="0.869914" />
|
64
|
+
<rate from="EUR" to="NOK" rate="7.800095" />
|
65
|
+
<rate from="GBP" to="NOK" rate="8.966508" />
|
66
|
+
</rates>
|
67
|
+
</currencyConfig>
|
@@ -0,0 +1,54 @@
|
|
1
|
+
# Licensed to the Apache Software Foundation (ASF) under one or more
|
2
|
+
# contributor license agreements. See the NOTICE file distributed with
|
3
|
+
# this work for additional information regarding copyright ownership.
|
4
|
+
# The ASF licenses this file to You under the Apache License, Version 2.0
|
5
|
+
# (the "License"); you may not use this file except in compliance with
|
6
|
+
# the License. You may obtain a copy of the License at
|
7
|
+
#
|
8
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
9
|
+
#
|
10
|
+
# Unless required by applicable law or agreed to in writing, software
|
11
|
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
12
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
13
|
+
# See the License for the specific language governing permissions and
|
14
|
+
# limitations under the License.
|
15
|
+
|
16
|
+
# a couple of test stopwords to test that the words are really being
|
17
|
+
# configured from this file:
|
18
|
+
stopworda
|
19
|
+
stopwordb
|
20
|
+
|
21
|
+
# Standard english stop words taken from Lucene's StopAnalyzer
|
22
|
+
a
|
23
|
+
an
|
24
|
+
and
|
25
|
+
are
|
26
|
+
as
|
27
|
+
at
|
28
|
+
be
|
29
|
+
but
|
30
|
+
by
|
31
|
+
for
|
32
|
+
if
|
33
|
+
in
|
34
|
+
into
|
35
|
+
is
|
36
|
+
it
|
37
|
+
no
|
38
|
+
not
|
39
|
+
of
|
40
|
+
on
|
41
|
+
or
|
42
|
+
such
|
43
|
+
that
|
44
|
+
the
|
45
|
+
their
|
46
|
+
then
|
47
|
+
there
|
48
|
+
these
|
49
|
+
they
|
50
|
+
this
|
51
|
+
to
|
52
|
+
was
|
53
|
+
will
|
54
|
+
with
|
@@ -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
|
+
|