polipus-elasticsearch 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.gitignore +35 -0
- data/Gemfile +2 -0
- data/LICENSE.txt +22 -0
- data/README.md +29 -0
- data/Rakefile +2 -0
- data/lib/polipus-elasticsearch/index/page.rb +168 -0
- data/lib/polipus-elasticsearch/storage/elasticsearch_store.rb +100 -0
- data/lib/polipus-elasticsearch.rb +3 -0
- data/polipus-elasticsearch.gemspec +27 -0
- data/spec/polipus-elasticsearch/storage/elasticsearch_store_spec.rb +175 -0
- data/spec/spec_helper.rb +51 -0
- metadata +215 -0
data/.gitignore
ADDED
@@ -0,0 +1,35 @@
|
|
1
|
+
*.gem
|
2
|
+
*.rbc
|
3
|
+
/.config
|
4
|
+
/coverage/
|
5
|
+
/InstalledFiles
|
6
|
+
/pkg/
|
7
|
+
/spec/reports/
|
8
|
+
/test/tmp/
|
9
|
+
/test/version_tmp/
|
10
|
+
/tmp/
|
11
|
+
|
12
|
+
## Specific to RubyMotion:
|
13
|
+
.dat*
|
14
|
+
.repl_history
|
15
|
+
build/
|
16
|
+
|
17
|
+
## Documentation cache and generated files:
|
18
|
+
/.yardoc/
|
19
|
+
/_yardoc/
|
20
|
+
/doc/
|
21
|
+
/rdoc/
|
22
|
+
|
23
|
+
## Environment normalisation:
|
24
|
+
/.bundle/
|
25
|
+
/vendor/bundle
|
26
|
+
/lib/bundler/man/
|
27
|
+
|
28
|
+
# for a library or gem, you might want to ignore these files since the code is
|
29
|
+
# intended to run in multiple environments; otherwise, check them in:
|
30
|
+
Gemfile.lock
|
31
|
+
# .ruby-version
|
32
|
+
# .ruby-gemset
|
33
|
+
|
34
|
+
# unless supporting rvm < 1.11.0 or doing something fancy, ignore this:
|
35
|
+
.rvmrc
|
data/Gemfile
ADDED
data/LICENSE.txt
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
Copyright (c) 2015 Stefano Fontanelli
|
2
|
+
|
3
|
+
MIT License
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
6
|
+
a copy of this software and associated documentation files (the
|
7
|
+
"Software"), to deal in the Software without restriction, including
|
8
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
9
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
10
|
+
permit persons to whom the Software is furnished to do so, subject to
|
11
|
+
the following conditions:
|
12
|
+
|
13
|
+
The above copyright notice and this permission notice shall be
|
14
|
+
included in all copies or substantial portions of the Software.
|
15
|
+
|
16
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
17
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
18
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
19
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
20
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
21
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
22
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,29 @@
|
|
1
|
+
# Polipus: addons for ElasticSearch
|
2
|
+
|
3
|
+
TODO: Write a gem description
|
4
|
+
|
5
|
+
## Installation
|
6
|
+
|
7
|
+
Add this line to your application's Gemfile:
|
8
|
+
|
9
|
+
gem 'polipus-elasticsearch'
|
10
|
+
|
11
|
+
And then execute:
|
12
|
+
|
13
|
+
$ bundle
|
14
|
+
|
15
|
+
Or install it yourself as:
|
16
|
+
|
17
|
+
$ gem install polipus-elasticsearch
|
18
|
+
|
19
|
+
## Usage
|
20
|
+
|
21
|
+
TODO: Write usage instructions here
|
22
|
+
|
23
|
+
## Contributing
|
24
|
+
|
25
|
+
1. Fork it ( http://github.com/<my-github-username>/polipus-elasticsearch/fork )
|
26
|
+
2. Create your feature branch (`git checkout -b my-new-feature`)
|
27
|
+
3. Commit your changes (`git commit -am 'Add some feature'`)
|
28
|
+
4. Push to the branch (`git push origin my-new-feature`)
|
29
|
+
5. Create new Pull Request
|
data/Rakefile
ADDED
@@ -0,0 +1,168 @@
|
|
1
|
+
require 'elasticsearch/model'
|
2
|
+
|
3
|
+
ENV['POLIPUS_ELASTICSEACH_INDEX_SHARDS'] ||= '1'
|
4
|
+
ENV['POLIPUS_ELASTICSEACH_INDEX_REPLICAS'] ||= '0'
|
5
|
+
|
6
|
+
module Polipus
|
7
|
+
module ElasticSearch
|
8
|
+
class Page
|
9
|
+
include Elasticsearch::Model
|
10
|
+
|
11
|
+
DEFAULT_INDEX_NAME = 'polipus-pages'
|
12
|
+
document_type 'polipus_page'
|
13
|
+
index_name DEFAULT_INDEX_NAME
|
14
|
+
|
15
|
+
settings(
|
16
|
+
index: {
|
17
|
+
number_of_shards: ENV['POLIPUS_ELASTICSEACH_INDEX_SHARDS'].to_i,
|
18
|
+
number_of_replicas: ENV['POLIPUS_ELASTICSEACH_INDEX_REPLICAS'].to_i
|
19
|
+
}
|
20
|
+
)
|
21
|
+
mapping(_all: { enabled: false }) do
|
22
|
+
indexes(
|
23
|
+
:id,
|
24
|
+
index: :not_analyzed
|
25
|
+
)
|
26
|
+
indexes(
|
27
|
+
:body,
|
28
|
+
type: :string
|
29
|
+
)
|
30
|
+
indexes(
|
31
|
+
:code,
|
32
|
+
type: :integer
|
33
|
+
)
|
34
|
+
indexes(
|
35
|
+
:depth,
|
36
|
+
type: :integer
|
37
|
+
)
|
38
|
+
indexes(
|
39
|
+
:error,
|
40
|
+
type: :string
|
41
|
+
)
|
42
|
+
indexes(
|
43
|
+
:fetched,
|
44
|
+
type: :boolean
|
45
|
+
)
|
46
|
+
indexes(
|
47
|
+
:fetched_at,
|
48
|
+
type: :integer
|
49
|
+
)
|
50
|
+
indexes(
|
51
|
+
:headers,
|
52
|
+
type: :string
|
53
|
+
)
|
54
|
+
indexes(
|
55
|
+
:links,
|
56
|
+
type: :string
|
57
|
+
)
|
58
|
+
indexes(
|
59
|
+
:redirect_to,
|
60
|
+
type: :string
|
61
|
+
)
|
62
|
+
indexes(
|
63
|
+
:referer,
|
64
|
+
type: :string
|
65
|
+
)
|
66
|
+
indexes(
|
67
|
+
:response_time,
|
68
|
+
type: :integer
|
69
|
+
)
|
70
|
+
indexes(
|
71
|
+
:url,
|
72
|
+
type: :string
|
73
|
+
)
|
74
|
+
indexes(
|
75
|
+
:user_data,
|
76
|
+
type: :string
|
77
|
+
)
|
78
|
+
end
|
79
|
+
|
80
|
+
def self.client
|
81
|
+
__elasticsearch__.client
|
82
|
+
end
|
83
|
+
|
84
|
+
def self.count
|
85
|
+
client.count(index: index_name, type: document_type)['count'].to_i
|
86
|
+
end
|
87
|
+
|
88
|
+
def self.create_index!(name)
|
89
|
+
index_name(name) unless name.nil?
|
90
|
+
__elasticsearch__.create_index!(index: index_name)
|
91
|
+
end
|
92
|
+
|
93
|
+
def self.clear_index!
|
94
|
+
client.delete_by_query(
|
95
|
+
index: index_name,
|
96
|
+
body: { query: { match_all: {} } }
|
97
|
+
)
|
98
|
+
end
|
99
|
+
|
100
|
+
def self.delete_index!
|
101
|
+
client.indices.delete(index: index_name)
|
102
|
+
end
|
103
|
+
|
104
|
+
def self.exists?(id)
|
105
|
+
client.exists?(
|
106
|
+
index: index_name,
|
107
|
+
type: document_type,
|
108
|
+
id: id
|
109
|
+
)
|
110
|
+
end
|
111
|
+
|
112
|
+
def self.get(id)
|
113
|
+
return unless exists?(id)
|
114
|
+
client.get_source(
|
115
|
+
index: index_name,
|
116
|
+
type: document_type,
|
117
|
+
id: id
|
118
|
+
)
|
119
|
+
end
|
120
|
+
|
121
|
+
def self.index_exists?
|
122
|
+
client.indices.exists?(index: index_name)
|
123
|
+
end
|
124
|
+
|
125
|
+
def self.process_document(obj)
|
126
|
+
doc = { '_type' => document_type }
|
127
|
+
properties.each do |p|
|
128
|
+
doc[p.to_s] = obj.respond_to?(p.to_s) ? obj.send(p.to_s) : obj[p.to_s]
|
129
|
+
end
|
130
|
+
doc.reject { |_, value| value.nil? }
|
131
|
+
end
|
132
|
+
|
133
|
+
def self.properties
|
134
|
+
mapping.to_hash[document_type.to_sym][:properties].keys.map { |k| k.to_s }
|
135
|
+
end
|
136
|
+
|
137
|
+
def self.remove(id, refresh = false)
|
138
|
+
return unless exists?(id)
|
139
|
+
client.delete(
|
140
|
+
index: index_name,
|
141
|
+
type: document_type,
|
142
|
+
id: id,
|
143
|
+
refresh: refresh,
|
144
|
+
version: Time.now.to_i,
|
145
|
+
version_type: :external
|
146
|
+
)
|
147
|
+
end
|
148
|
+
|
149
|
+
def self.setup(client_)
|
150
|
+
__elasticsearch__.client = client_
|
151
|
+
end
|
152
|
+
|
153
|
+
def self.store(document, refresh = false)
|
154
|
+
document = process_document(document)
|
155
|
+
client.index(
|
156
|
+
index: index_name,
|
157
|
+
type: document_type,
|
158
|
+
id: document['id'],
|
159
|
+
body: document,
|
160
|
+
refresh: refresh,
|
161
|
+
version: document['fetched_at'].to_i,
|
162
|
+
version_type: :external
|
163
|
+
)
|
164
|
+
document['id']
|
165
|
+
end
|
166
|
+
end
|
167
|
+
end
|
168
|
+
end
|
@@ -0,0 +1,100 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
require 'base64'
|
3
|
+
require 'multi_json'
|
4
|
+
require 'polipus'
|
5
|
+
require 'polipus-elasticsearch'
|
6
|
+
|
7
|
+
module Polipus
|
8
|
+
module Storage
|
9
|
+
class ElasticSearchStore < Base
|
10
|
+
BINARY_FIELDS = %w(body headers user_data)
|
11
|
+
DEFAULT_INDEX = Polipus::ElasticSearch::Page
|
12
|
+
|
13
|
+
attr_accessor :index, :index_name, :except, :compress, :semaphore, :refresh
|
14
|
+
|
15
|
+
def initialize(client, options = {})
|
16
|
+
@index = options[:index] || options['index'] || DEFAULT_INDEX
|
17
|
+
@index_name = options[:index_name] || options['index_name']
|
18
|
+
@except = options[:except] || options['except'] || []
|
19
|
+
@compress = options[:compress] || options['compress']
|
20
|
+
@semaphore = Mutex.new
|
21
|
+
@refresh = options[:refresh] || options['refresh'] || true
|
22
|
+
index.setup(client)
|
23
|
+
index.create_index!(index_name) unless index.index_exists?
|
24
|
+
end
|
25
|
+
|
26
|
+
def add(page)
|
27
|
+
semaphore.synchronize do
|
28
|
+
obj = page.to_hash
|
29
|
+
Array(except).each { |field| obj.delete(field.to_s) }
|
30
|
+
BINARY_FIELDS.each do |field|
|
31
|
+
next if obj[field.to_s].nil? || obj[field.to_s].empty?
|
32
|
+
obj[field.to_s] = MultiJson.encode(obj[field.to_s]) if field.to_s == 'user_data'
|
33
|
+
obj[field.to_s] = Base64.encode64(obj[field.to_s])
|
34
|
+
end
|
35
|
+
obj['id'] = uuid(page)
|
36
|
+
obj['fetched_at'] = obj['fetched_at'].to_i
|
37
|
+
index.store(obj, refresh)
|
38
|
+
end
|
39
|
+
end
|
40
|
+
|
41
|
+
def clear
|
42
|
+
index.clear_index! if index.index_exists?
|
43
|
+
end
|
44
|
+
|
45
|
+
def count
|
46
|
+
index.count
|
47
|
+
end
|
48
|
+
|
49
|
+
def drop
|
50
|
+
index.delete_index! if index.index_exists?
|
51
|
+
end
|
52
|
+
|
53
|
+
def each
|
54
|
+
# This method is implemented only for testing purposes
|
55
|
+
response = index.client.search(
|
56
|
+
index: index_name,
|
57
|
+
body: {
|
58
|
+
query: { match_all: {} },
|
59
|
+
from: 0,
|
60
|
+
size: 25
|
61
|
+
}
|
62
|
+
)
|
63
|
+
response['hits']['hits'].each do |data|
|
64
|
+
page = load_page(data['_source'])
|
65
|
+
yield uuid(page), page
|
66
|
+
end
|
67
|
+
end
|
68
|
+
|
69
|
+
def exists?(page)
|
70
|
+
@semaphore.synchronize do
|
71
|
+
index.exists?(uuid(page))
|
72
|
+
end
|
73
|
+
end
|
74
|
+
|
75
|
+
def get(page)
|
76
|
+
@semaphore.synchronize do
|
77
|
+
load_page(index.get(uuid(page)))
|
78
|
+
end
|
79
|
+
end
|
80
|
+
|
81
|
+
def remove(page)
|
82
|
+
@semaphore.synchronize do
|
83
|
+
index.remove(uuid(page), refresh)
|
84
|
+
end
|
85
|
+
end
|
86
|
+
|
87
|
+
def load_page(data)
|
88
|
+
return nil if data.nil?
|
89
|
+
BINARY_FIELDS.each do |field|
|
90
|
+
next if data[field.to_s].nil? || data[field.to_s].empty?
|
91
|
+
data[field.to_s] = Base64.decode64(data[field.to_s])
|
92
|
+
data[field.to_s] = MultiJson.decode(data[field.to_s]) if field.to_s == 'user_data'
|
93
|
+
end
|
94
|
+
page = Page.from_hash(data)
|
95
|
+
page.fetched_at ||= 0
|
96
|
+
page
|
97
|
+
end
|
98
|
+
end
|
99
|
+
end
|
100
|
+
end
|
@@ -0,0 +1,27 @@
|
|
1
|
+
# coding: utf-8
|
2
|
+
lib = File.expand_path('../lib', __FILE__)
|
3
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
+
|
5
|
+
Gem::Specification.new do |spec|
|
6
|
+
spec.name = 'polipus-elasticsearch'
|
7
|
+
spec.version = '0.0.1'
|
8
|
+
spec.authors = ['Stefano Fontanelli']
|
9
|
+
spec.email = ['s.fontanelli@gmail.com']
|
10
|
+
spec.summary = 'Add support for ElasticSearch in Polipus crawler'
|
11
|
+
spec.description = 'Add support for ElasticSearch in Polipus crawler'
|
12
|
+
spec.homepage = 'https://github.com/stefanofontanelli/polipus-elasticsearch'
|
13
|
+
spec.license = 'MIT'
|
14
|
+
spec.files = `git ls-files -z`.split("\x0")
|
15
|
+
spec.executables = spec.files.grep(/^bin\//) { |f| File.basename(f) }
|
16
|
+
spec.test_files = spec.files.grep(/^(test|spec|features)\//)
|
17
|
+
spec.require_paths = ['lib']
|
18
|
+
spec.add_runtime_dependency 'elasticsearch', '~> 1.0.4'
|
19
|
+
spec.add_runtime_dependency 'elasticsearch-model', '~> 0.1.4'
|
20
|
+
spec.add_runtime_dependency 'polipus', '~> 0.3', '>= 0.3.0'
|
21
|
+
spec.add_development_dependency 'rake', '~> 10.3'
|
22
|
+
spec.add_development_dependency 'rspec', '~> 3.1.0'
|
23
|
+
spec.add_development_dependency 'flexmock', '~> 1.3'
|
24
|
+
spec.add_development_dependency 'vcr', '~> 2.9.0'
|
25
|
+
spec.add_development_dependency 'webmock', '~> 1.20.0'
|
26
|
+
spec.add_development_dependency 'coveralls'
|
27
|
+
end
|
@@ -0,0 +1,175 @@
|
|
1
|
+
# encoding: UTF-8
|
2
|
+
require 'logger'
|
3
|
+
require 'polipus-elasticsearch'
|
4
|
+
require 'spec_helper'
|
5
|
+
|
6
|
+
describe Polipus::Storage::ElasticSearchStore do
|
7
|
+
before(:each)do
|
8
|
+
@logger = Logger.new(STDOUT)
|
9
|
+
@client = Elasticsearch::Client.new(host: '127.0.0.1', logger: @logger)
|
10
|
+
@client.transport.logger.level = Logger::INFO
|
11
|
+
@index_name = 'polipus_elasticsearch_test'
|
12
|
+
@storage = Polipus::Storage::ElasticSearchStore.new(
|
13
|
+
@client,
|
14
|
+
index_name: @index_name,
|
15
|
+
refresh: true
|
16
|
+
)
|
17
|
+
@storage_without_code_and_body = Polipus::Storage::ElasticSearchStore.new(
|
18
|
+
@client,
|
19
|
+
index_name: @index_name,
|
20
|
+
except: ['code', 'body'],
|
21
|
+
refresh: true
|
22
|
+
)
|
23
|
+
end
|
24
|
+
|
25
|
+
after(:each) do
|
26
|
+
@storage.drop
|
27
|
+
end
|
28
|
+
|
29
|
+
it 'should store a page' do
|
30
|
+
p = page_factory 'http://www.google.com'
|
31
|
+
uuid = @storage.add(p)
|
32
|
+
expect(uuid).to eq('ed646a3334ca891fd3467db131372140')
|
33
|
+
p = @storage.get(p)
|
34
|
+
expect(p).not_to be_nil
|
35
|
+
expect(p.url.to_s).to eq('http://www.google.com')
|
36
|
+
expect(p.body).to eq('<html></html>')
|
37
|
+
@storage.remove(p)
|
38
|
+
p = @storage.get(p)
|
39
|
+
expect(p).to be_nil
|
40
|
+
end
|
41
|
+
|
42
|
+
it 'should store all the relevant data from the page' do
|
43
|
+
url = "http://www.duckduckgo.com"
|
44
|
+
referer = "http://www.actually.nowhere.com"
|
45
|
+
redirectto = "#{url}/your_super_awesome_results?page=42"
|
46
|
+
now = Time.now.to_i
|
47
|
+
p = page_factory(
|
48
|
+
url,
|
49
|
+
{
|
50
|
+
referer: referer,
|
51
|
+
redirect_to: redirectto,
|
52
|
+
fetched_at: now
|
53
|
+
})
|
54
|
+
uuid = @storage.add p
|
55
|
+
expect(uuid).to eq('3cd657f53c74f22c1a21b420ce3863fd')
|
56
|
+
p = @storage.get p
|
57
|
+
|
58
|
+
expect(p.url.to_s).to eq(url)
|
59
|
+
expect(p.referer.to_s).to eq(referer)
|
60
|
+
expect(p.redirect_to.to_s).to eq(redirectto)
|
61
|
+
expect(p.fetched_at).to eq(now)
|
62
|
+
expect(p.body).to eq('<html></html>')
|
63
|
+
|
64
|
+
# for the sake of the other tests...
|
65
|
+
expect(@storage.remove(p)).to be_truthy
|
66
|
+
end
|
67
|
+
|
68
|
+
it 'should update a page' do
|
69
|
+
p = page_factory 'http://www.google.com', code: 301
|
70
|
+
@storage.add p
|
71
|
+
p = @storage.get p
|
72
|
+
expect(p.code).to eq(301)
|
73
|
+
end
|
74
|
+
|
75
|
+
it 'should iterate over stored pages' do
|
76
|
+
p = page_factory('http://www.google.com')
|
77
|
+
@storage.add(p)
|
78
|
+
@storage.each do |k, page|
|
79
|
+
expect(k).to eq('ed646a3334ca891fd3467db131372140')
|
80
|
+
expect(page.url.to_s).to eq('http://www.google.com')
|
81
|
+
end
|
82
|
+
end
|
83
|
+
|
84
|
+
it 'should delete a page' do
|
85
|
+
p = page_factory 'http://www.google.com', code: 301
|
86
|
+
@storage.remove p
|
87
|
+
expect(@storage.get(p)).to be_nil
|
88
|
+
end
|
89
|
+
|
90
|
+
it 'should store a page removing a query string from the uuid generation' do
|
91
|
+
p = page_factory 'http://www.asd.com/?asd=lol'
|
92
|
+
p_no_query = page_factory 'http://www.asd.com/?asdas=dasda&adsda=1'
|
93
|
+
@storage.include_query_string_in_uuid = false
|
94
|
+
@storage.add p
|
95
|
+
expect(@storage.exists?(p_no_query)).to be_truthy
|
96
|
+
@storage.remove p
|
97
|
+
end
|
98
|
+
|
99
|
+
it 'should store a page removing a query string from the uuid generation no ending slash' do
|
100
|
+
p = page_factory 'http://www.asd.com?asd=lol'
|
101
|
+
p_no_query = page_factory 'http://www.asd.com'
|
102
|
+
@storage.include_query_string_in_uuid = false
|
103
|
+
@storage.add p
|
104
|
+
expect(@storage.exists?(p_no_query)).to be_truthy
|
105
|
+
@storage.remove p
|
106
|
+
end
|
107
|
+
|
108
|
+
it 'should store a page with user data associated' do
|
109
|
+
p = page_factory 'http://www.user.com'
|
110
|
+
p.user_data.name = 'Test User Data'
|
111
|
+
@storage.add p
|
112
|
+
expect(@storage.exists?(p)).to be_truthy
|
113
|
+
p = @storage.get(p)
|
114
|
+
expect(p.user_data.name).to eq('Test User Data')
|
115
|
+
@storage.remove p
|
116
|
+
end
|
117
|
+
|
118
|
+
it 'should honor the except parameters' do
|
119
|
+
pag = page_factory 'http://www.user-doo.com'
|
120
|
+
expect(pag.code).to eq(200)
|
121
|
+
expect(pag.body).to eq('<html></html>')
|
122
|
+
|
123
|
+
@storage_without_code_and_body.add(pag)
|
124
|
+
pag = @storage_without_code_and_body.get(pag)
|
125
|
+
|
126
|
+
expect(pag.body).to be_nil
|
127
|
+
expect(pag.code).to eq(0)
|
128
|
+
@storage_without_code_and_body.remove(pag)
|
129
|
+
end
|
130
|
+
|
131
|
+
it 'should return false if a doc not exists' do
|
132
|
+
@storage.include_query_string_in_uuid = false
|
133
|
+
p_other = page_factory 'http://www.asdrrrr.com'
|
134
|
+
expect(@storage.exists?(p_other)).to be_falsey
|
135
|
+
@storage.add p_other
|
136
|
+
expect(@storage.exists?(p_other)).to be_truthy
|
137
|
+
p_other = page_factory 'http://www.asdrrrr.com?trk=asd-lol'
|
138
|
+
expect(@storage.exists?(p_other)).to be_truthy
|
139
|
+
@storage.include_query_string_in_uuid = true
|
140
|
+
expect(@storage.exists?(p_other)).to be_falsey
|
141
|
+
@storage.include_query_string_in_uuid = false
|
142
|
+
@storage.remove p_other
|
143
|
+
end
|
144
|
+
|
145
|
+
it 'should set page.fetched_at based on the id creation' do
|
146
|
+
p = page_factory 'http://www.user-doojo.com'
|
147
|
+
@storage.add p
|
148
|
+
expect(p.fetched_at).not_to be_nil
|
149
|
+
p = @storage.get p
|
150
|
+
expect(p.fetched_at).not_to be_nil
|
151
|
+
@storage.remove p
|
152
|
+
end
|
153
|
+
|
154
|
+
it 'should NOT set page.fetched_at if already present' do
|
155
|
+
p = page_factory 'http://www.user-doojooo.com'
|
156
|
+
p.fetched_at = 10
|
157
|
+
@storage.add p
|
158
|
+
p = @storage.get p
|
159
|
+
expect(p.fetched_at).to be 10
|
160
|
+
@storage.remove p
|
161
|
+
end
|
162
|
+
|
163
|
+
it 'should store two pages and the count will be two' do
|
164
|
+
pages = ['http://www.google.com', 'http://www.duckduckgo.com'].map do |url|
|
165
|
+
page_factory(url).tap do |page|
|
166
|
+
@storage.add(page)
|
167
|
+
end
|
168
|
+
end
|
169
|
+
expect(@storage.count).to be 2
|
170
|
+
pages.each do |page|
|
171
|
+
@storage.remove(page)
|
172
|
+
end
|
173
|
+
expect(@storage.count).to be 0
|
174
|
+
end
|
175
|
+
end
|
data/spec/spec_helper.rb
ADDED
@@ -0,0 +1,51 @@
|
|
1
|
+
# Require this file using `require "spec_helper"`
|
2
|
+
# to ensure that it is only loaded once.
|
3
|
+
#
|
4
|
+
# See http://rubydoc.info/gems/rspec-core/RSpec/Core/Configuration
|
5
|
+
require 'digest/md5'
|
6
|
+
require 'coveralls'
|
7
|
+
require 'vcr'
|
8
|
+
require 'webmock/rspec'
|
9
|
+
|
10
|
+
Coveralls.wear!
|
11
|
+
|
12
|
+
require 'polipus'
|
13
|
+
|
14
|
+
VCR.configure do |c|
|
15
|
+
c.cassette_library_dir = "#{File.dirname(__FILE__)}/cassettes"
|
16
|
+
c.hook_into :webmock
|
17
|
+
c.allow_http_connections_when_no_cassette = true
|
18
|
+
c.ignore_localhost = true
|
19
|
+
end
|
20
|
+
|
21
|
+
RSpec.configure do |config|
|
22
|
+
config.run_all_when_everything_filtered = true
|
23
|
+
config.filter_run :focus
|
24
|
+
|
25
|
+
# Run specs in random order to surface order dependencies. If you find an
|
26
|
+
# order dependency and want to debug it, you can fix the order by providing
|
27
|
+
# the seed, which is printed after each run.
|
28
|
+
# --seed 1234
|
29
|
+
config.order = 'random'
|
30
|
+
config.mock_with :flexmock
|
31
|
+
config.around(:each) do |example|
|
32
|
+
t = Time.now
|
33
|
+
print example.metadata[:full_description]
|
34
|
+
VCR.use_cassette(
|
35
|
+
Digest::MD5.hexdigest(example.metadata[:full_description]),
|
36
|
+
record: :all
|
37
|
+
) do
|
38
|
+
example.run
|
39
|
+
end
|
40
|
+
puts " [#{Time.now - t}s]"
|
41
|
+
end
|
42
|
+
config.before(:each) { Polipus::SignalHandler.disable }
|
43
|
+
end
|
44
|
+
|
45
|
+
def page_factory(url, params = {})
|
46
|
+
params[:code] ||= 200 unless params.has_key?(:code)
|
47
|
+
params[:body] = '<html></html>' unless params.has_key?(:body)
|
48
|
+
params[:fetched_at] = Time.now.to_i
|
49
|
+
sleep(1)
|
50
|
+
Polipus::Page.new(url, params)
|
51
|
+
end
|
metadata
ADDED
@@ -0,0 +1,215 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: polipus-elasticsearch
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Stefano Fontanelli
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2015-07-17 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: elasticsearch
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ~>
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: 1.0.4
|
22
|
+
type: :runtime
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ~>
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: 1.0.4
|
30
|
+
- !ruby/object:Gem::Dependency
|
31
|
+
name: elasticsearch-model
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
33
|
+
none: false
|
34
|
+
requirements:
|
35
|
+
- - ~>
|
36
|
+
- !ruby/object:Gem::Version
|
37
|
+
version: 0.1.4
|
38
|
+
type: :runtime
|
39
|
+
prerelease: false
|
40
|
+
version_requirements: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ~>
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: 0.1.4
|
46
|
+
- !ruby/object:Gem::Dependency
|
47
|
+
name: polipus
|
48
|
+
requirement: !ruby/object:Gem::Requirement
|
49
|
+
none: false
|
50
|
+
requirements:
|
51
|
+
- - ~>
|
52
|
+
- !ruby/object:Gem::Version
|
53
|
+
version: '0.3'
|
54
|
+
- - ! '>='
|
55
|
+
- !ruby/object:Gem::Version
|
56
|
+
version: 0.3.0
|
57
|
+
type: :runtime
|
58
|
+
prerelease: false
|
59
|
+
version_requirements: !ruby/object:Gem::Requirement
|
60
|
+
none: false
|
61
|
+
requirements:
|
62
|
+
- - ~>
|
63
|
+
- !ruby/object:Gem::Version
|
64
|
+
version: '0.3'
|
65
|
+
- - ! '>='
|
66
|
+
- !ruby/object:Gem::Version
|
67
|
+
version: 0.3.0
|
68
|
+
- !ruby/object:Gem::Dependency
|
69
|
+
name: rake
|
70
|
+
requirement: !ruby/object:Gem::Requirement
|
71
|
+
none: false
|
72
|
+
requirements:
|
73
|
+
- - ~>
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: '10.3'
|
76
|
+
type: :development
|
77
|
+
prerelease: false
|
78
|
+
version_requirements: !ruby/object:Gem::Requirement
|
79
|
+
none: false
|
80
|
+
requirements:
|
81
|
+
- - ~>
|
82
|
+
- !ruby/object:Gem::Version
|
83
|
+
version: '10.3'
|
84
|
+
- !ruby/object:Gem::Dependency
|
85
|
+
name: rspec
|
86
|
+
requirement: !ruby/object:Gem::Requirement
|
87
|
+
none: false
|
88
|
+
requirements:
|
89
|
+
- - ~>
|
90
|
+
- !ruby/object:Gem::Version
|
91
|
+
version: 3.1.0
|
92
|
+
type: :development
|
93
|
+
prerelease: false
|
94
|
+
version_requirements: !ruby/object:Gem::Requirement
|
95
|
+
none: false
|
96
|
+
requirements:
|
97
|
+
- - ~>
|
98
|
+
- !ruby/object:Gem::Version
|
99
|
+
version: 3.1.0
|
100
|
+
- !ruby/object:Gem::Dependency
|
101
|
+
name: flexmock
|
102
|
+
requirement: !ruby/object:Gem::Requirement
|
103
|
+
none: false
|
104
|
+
requirements:
|
105
|
+
- - ~>
|
106
|
+
- !ruby/object:Gem::Version
|
107
|
+
version: '1.3'
|
108
|
+
type: :development
|
109
|
+
prerelease: false
|
110
|
+
version_requirements: !ruby/object:Gem::Requirement
|
111
|
+
none: false
|
112
|
+
requirements:
|
113
|
+
- - ~>
|
114
|
+
- !ruby/object:Gem::Version
|
115
|
+
version: '1.3'
|
116
|
+
- !ruby/object:Gem::Dependency
|
117
|
+
name: vcr
|
118
|
+
requirement: !ruby/object:Gem::Requirement
|
119
|
+
none: false
|
120
|
+
requirements:
|
121
|
+
- - ~>
|
122
|
+
- !ruby/object:Gem::Version
|
123
|
+
version: 2.9.0
|
124
|
+
type: :development
|
125
|
+
prerelease: false
|
126
|
+
version_requirements: !ruby/object:Gem::Requirement
|
127
|
+
none: false
|
128
|
+
requirements:
|
129
|
+
- - ~>
|
130
|
+
- !ruby/object:Gem::Version
|
131
|
+
version: 2.9.0
|
132
|
+
- !ruby/object:Gem::Dependency
|
133
|
+
name: webmock
|
134
|
+
requirement: !ruby/object:Gem::Requirement
|
135
|
+
none: false
|
136
|
+
requirements:
|
137
|
+
- - ~>
|
138
|
+
- !ruby/object:Gem::Version
|
139
|
+
version: 1.20.0
|
140
|
+
type: :development
|
141
|
+
prerelease: false
|
142
|
+
version_requirements: !ruby/object:Gem::Requirement
|
143
|
+
none: false
|
144
|
+
requirements:
|
145
|
+
- - ~>
|
146
|
+
- !ruby/object:Gem::Version
|
147
|
+
version: 1.20.0
|
148
|
+
- !ruby/object:Gem::Dependency
|
149
|
+
name: coveralls
|
150
|
+
requirement: !ruby/object:Gem::Requirement
|
151
|
+
none: false
|
152
|
+
requirements:
|
153
|
+
- - ! '>='
|
154
|
+
- !ruby/object:Gem::Version
|
155
|
+
version: '0'
|
156
|
+
type: :development
|
157
|
+
prerelease: false
|
158
|
+
version_requirements: !ruby/object:Gem::Requirement
|
159
|
+
none: false
|
160
|
+
requirements:
|
161
|
+
- - ! '>='
|
162
|
+
- !ruby/object:Gem::Version
|
163
|
+
version: '0'
|
164
|
+
description: Add support for ElasticSearch in Polipus crawler
|
165
|
+
email:
|
166
|
+
- s.fontanelli@gmail.com
|
167
|
+
executables: []
|
168
|
+
extensions: []
|
169
|
+
extra_rdoc_files: []
|
170
|
+
files:
|
171
|
+
- .gitignore
|
172
|
+
- Gemfile
|
173
|
+
- LICENSE.txt
|
174
|
+
- README.md
|
175
|
+
- Rakefile
|
176
|
+
- lib/polipus-elasticsearch.rb
|
177
|
+
- lib/polipus-elasticsearch/index/page.rb
|
178
|
+
- lib/polipus-elasticsearch/storage/elasticsearch_store.rb
|
179
|
+
- polipus-elasticsearch.gemspec
|
180
|
+
- spec/polipus-elasticsearch/storage/elasticsearch_store_spec.rb
|
181
|
+
- spec/spec_helper.rb
|
182
|
+
homepage: https://github.com/stefanofontanelli/polipus-elasticsearch
|
183
|
+
licenses:
|
184
|
+
- MIT
|
185
|
+
post_install_message:
|
186
|
+
rdoc_options: []
|
187
|
+
require_paths:
|
188
|
+
- lib
|
189
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
190
|
+
none: false
|
191
|
+
requirements:
|
192
|
+
- - ! '>='
|
193
|
+
- !ruby/object:Gem::Version
|
194
|
+
version: '0'
|
195
|
+
segments:
|
196
|
+
- 0
|
197
|
+
hash: -3053519168912849089
|
198
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
199
|
+
none: false
|
200
|
+
requirements:
|
201
|
+
- - ! '>='
|
202
|
+
- !ruby/object:Gem::Version
|
203
|
+
version: '0'
|
204
|
+
segments:
|
205
|
+
- 0
|
206
|
+
hash: -3053519168912849089
|
207
|
+
requirements: []
|
208
|
+
rubyforge_project:
|
209
|
+
rubygems_version: 1.8.23.2
|
210
|
+
signing_key:
|
211
|
+
specification_version: 3
|
212
|
+
summary: Add support for ElasticSearch in Polipus crawler
|
213
|
+
test_files:
|
214
|
+
- spec/polipus-elasticsearch/storage/elasticsearch_store_spec.rb
|
215
|
+
- spec/spec_helper.rb
|