lowang-rubberband 0.0.3
Sign up to get free protection for your applications and to get access to all the features.
- data/LICENSE +202 -0
- data/README.rdoc +36 -0
- data/Rakefile +59 -0
- data/TODO +29 -0
- data/VERSION +1 -0
- data/lib/elasticsearch.rb +14 -0
- data/lib/elasticsearch/client.rb +20 -0
- data/lib/elasticsearch/client/abstract_client.rb +34 -0
- data/lib/elasticsearch/client/admin_cluster.rb +56 -0
- data/lib/elasticsearch/client/admin_index.rb +109 -0
- data/lib/elasticsearch/client/auto_discovering_client.rb +21 -0
- data/lib/elasticsearch/client/default_scope.rb +29 -0
- data/lib/elasticsearch/client/hits.rb +83 -0
- data/lib/elasticsearch/client/index.rb +102 -0
- data/lib/elasticsearch/client/retrying_client.rb +81 -0
- data/lib/elasticsearch/encoding.rb +7 -0
- data/lib/elasticsearch/encoding/base.rb +17 -0
- data/lib/elasticsearch/encoding/json.rb +19 -0
- data/lib/elasticsearch/transport.rb +14 -0
- data/lib/elasticsearch/transport/base.rb +37 -0
- data/lib/elasticsearch/transport/base_protocol.rb +235 -0
- data/lib/elasticsearch/transport/http.rb +50 -0
- data/lib/elasticsearch/transport/thrift.rb +96 -0
- data/lib/elasticsearch/transport/thrift/elasticsearch.thrift +81 -0
- data/lib/elasticsearch/transport/thrift/elasticsearch_constants.rb +12 -0
- data/lib/elasticsearch/transport/thrift/elasticsearch_types.rb +124 -0
- data/lib/elasticsearch/transport/thrift/rest.rb +83 -0
- data/lowang-rubberband.gemspec +81 -0
- data/rubberband.gemspec +80 -0
- data/test/elasticsearch_test.rb +7 -0
- data/test/test_helper.rb +10 -0
- metadata +141 -0
@@ -0,0 +1,19 @@
|
|
1
|
+
require 'yajl'
|
2
|
+
|
3
|
+
module ElasticSearch
|
4
|
+
module Encoding
|
5
|
+
class JSON < Base
|
6
|
+
def encode(object)
|
7
|
+
Yajl::Encoder.encode(object)
|
8
|
+
end
|
9
|
+
|
10
|
+
def decode(string)
|
11
|
+
Yajl::Parser.parse(string)
|
12
|
+
end
|
13
|
+
|
14
|
+
def is_encoded?(object)
|
15
|
+
object.is_a?(String)
|
16
|
+
end
|
17
|
+
end
|
18
|
+
end
|
19
|
+
end
|
@@ -0,0 +1,14 @@
|
|
1
|
+
require "transport/base_protocol"
|
2
|
+
require "transport/base"
|
3
|
+
|
4
|
+
module ElasticSearch
|
5
|
+
class ConnectionFailed < RetryableError; end
|
6
|
+
class HostResolutionError < RetryableError; end
|
7
|
+
class TimeoutError < RetryableError; end
|
8
|
+
class RequestError < FatalError; end
|
9
|
+
|
10
|
+
module Transport
|
11
|
+
autoload :HTTP, 'transport/http'
|
12
|
+
autoload :Thrift, 'transport/thrift'
|
13
|
+
end
|
14
|
+
end
|
@@ -0,0 +1,37 @@
|
|
1
|
+
module ElasticSearch
|
2
|
+
class RetryableError < StandardError; end
|
3
|
+
class FatalError < StandardError; end
|
4
|
+
|
5
|
+
module Transport
|
6
|
+
|
7
|
+
DEFAULTS = {
|
8
|
+
:encoder => ElasticSearch::Encoding::JSON
|
9
|
+
}.freeze
|
10
|
+
|
11
|
+
class Base
|
12
|
+
include BaseProtocol
|
13
|
+
|
14
|
+
attr_accessor :server, :options
|
15
|
+
|
16
|
+
def initialize(server, options={})
|
17
|
+
@server = server
|
18
|
+
@options = DEFAULTS.merge(options)
|
19
|
+
end
|
20
|
+
|
21
|
+
def connect!
|
22
|
+
raise NotImplementedError
|
23
|
+
end
|
24
|
+
|
25
|
+
def close
|
26
|
+
end
|
27
|
+
|
28
|
+
def encoder
|
29
|
+
@encoder ||= @options[:encoder].new
|
30
|
+
end
|
31
|
+
|
32
|
+
def request(method, operation, params={}, body=nil, headers={})
|
33
|
+
raise NotImplementedError
|
34
|
+
end
|
35
|
+
end
|
36
|
+
end
|
37
|
+
end
|
@@ -0,0 +1,235 @@
|
|
1
|
+
module ElasticSearch
|
2
|
+
module Transport
|
3
|
+
module IndexProtocol
|
4
|
+
def index(index, type, id, document, options={})
|
5
|
+
body = encoder.is_encoded?(document) ? document : encoder.encode(document)
|
6
|
+
if id.nil?
|
7
|
+
response = request(:post, {:index => index, :type => type}, {}, body)
|
8
|
+
else
|
9
|
+
response = request(:put, {:index => index, :type => type, :id => id}, {}, body)
|
10
|
+
end
|
11
|
+
handle_error(response) unless response.status == 200
|
12
|
+
encoder.decode(response.body)
|
13
|
+
end
|
14
|
+
|
15
|
+
def get(index, type, id, options={})
|
16
|
+
response = request(:get, {:index => index, :type => type, :id => id})
|
17
|
+
return nil if response.status == 404
|
18
|
+
|
19
|
+
handle_error(response) unless response.status == 200
|
20
|
+
hit = encoder.decode(response.body)
|
21
|
+
unescape_id!(hit) #TODO extract these two calls from here and search
|
22
|
+
set_encoding!(hit)
|
23
|
+
hit # { "_id", "_index", "_type", "_source" }
|
24
|
+
end
|
25
|
+
|
26
|
+
def delete(index, type, id, options={})
|
27
|
+
response = request(:delete,{:index => index, :type => type, :id => id})
|
28
|
+
handle_error(response) unless response.status == 200 # ElasticSearch always returns 200 on delete, even if the object doesn't exist
|
29
|
+
encoder.decode(response.body)
|
30
|
+
end
|
31
|
+
|
32
|
+
def search(index, type, query, options={})
|
33
|
+
if query.is_a?(Hash)
|
34
|
+
# patron cannot submit get requests with content, so if query is a hash, post it instead (assume a query hash is using the query dsl)
|
35
|
+
response = request(:post, {:index => index, :type => type, :op => "_search"}, options, encoder.encode(query))
|
36
|
+
else
|
37
|
+
response = request(:get, {:index => index, :type => type, :op => "_search"}, options.merge(:q => query))
|
38
|
+
end
|
39
|
+
handle_error(response) unless response.status == 200
|
40
|
+
results = encoder.decode(response.body)
|
41
|
+
# unescape ids
|
42
|
+
results["hits"]["hits"].each do |hit|
|
43
|
+
unescape_id!(hit)
|
44
|
+
set_encoding!(hit)
|
45
|
+
end
|
46
|
+
results # {"hits"=>{"hits"=>[{"_id", "_type", "_source", "_index", "_score"}], "total"}, "_shards"=>{"failed", "total", "successful"}}
|
47
|
+
end
|
48
|
+
|
49
|
+
def scroll(scroll_id)
|
50
|
+
response = request(:get, {:op => "_search/scroll"}, {:scroll_id => scroll_id })
|
51
|
+
handle_error(response) unless response.status == 200
|
52
|
+
results = encoder.decode(response.body)
|
53
|
+
# unescape ids
|
54
|
+
results["hits"]["hits"].each do |hit|
|
55
|
+
unescape_id!(hit)
|
56
|
+
set_encoding!(hit)
|
57
|
+
end
|
58
|
+
results # {"hits"=>{"hits"=>[{"_id", "_type", "_source", "_index", "_score"}], "total"}, "_shards"=>{"failed", "total", "successful"}, "_scrollId"}
|
59
|
+
end
|
60
|
+
|
61
|
+
def count(index, type, query, options={})
|
62
|
+
if query.is_a?(Hash)
|
63
|
+
# patron cannot submit get requests with content, so if query is a hash, post it instead (assume a query hash is using the query dsl)
|
64
|
+
response = request(:post, {:index => index, :type => type, :op => "_count"}, options, encoder.encode(query))
|
65
|
+
else
|
66
|
+
response = request(:get, {:index => index, :type => type, :op => "_count"}, options.merge(:q => query))
|
67
|
+
end
|
68
|
+
handle_error(response) unless response.status == 200
|
69
|
+
encoder.decode(response.body) # {"count", "_shards"=>{"failed", "total", "successful"}}
|
70
|
+
end
|
71
|
+
end
|
72
|
+
|
73
|
+
module IndexAdminProtocol
|
74
|
+
def index_status(index_list, options={})
|
75
|
+
standard_request(:get, {:index => index_list, :op => "_status"})
|
76
|
+
end
|
77
|
+
|
78
|
+
def create_index(index, create_options={}, options={})
|
79
|
+
standard_request(:put, {:index => index}, {}, encoder.encode(create_options))
|
80
|
+
end
|
81
|
+
|
82
|
+
def delete_index(index, options={})
|
83
|
+
standard_request(:delete, {:index => index})
|
84
|
+
end
|
85
|
+
|
86
|
+
def alias_index(operations, options={})
|
87
|
+
standard_request(:post, {:op => "_aliases"}, {}, encoder.encode(operations))
|
88
|
+
end
|
89
|
+
|
90
|
+
def update_mapping(index, type, mapping, options)
|
91
|
+
standard_request(:put, {:index => index, :type => type, :op => "_mapping"}, options, encoder.encode(mapping))
|
92
|
+
end
|
93
|
+
|
94
|
+
def flush(index_list, options={})
|
95
|
+
standard_request(:post, {:index => index_list, :op => "_flush"}, options, "")
|
96
|
+
end
|
97
|
+
|
98
|
+
def refresh(index_list, options={})
|
99
|
+
standard_request(:post, {:index => index_list, :op => "_refresh"}, {}, "")
|
100
|
+
end
|
101
|
+
|
102
|
+
def snapshot(index_list, options={})
|
103
|
+
standard_request(:post, {:index => index_list, :type => "_gateway", :op => "snapshot"}, {}, "")
|
104
|
+
end
|
105
|
+
|
106
|
+
def optimize(index_list, options={})
|
107
|
+
standard_request(:post, {:index => index_list, :op => "_optimize"}, options, {}, "")
|
108
|
+
end
|
109
|
+
end
|
110
|
+
|
111
|
+
module ClusterAdminProtocol
|
112
|
+
def cluster_health(index_list, options={})
|
113
|
+
standard_request(:get, {:index => "_cluster", :type => "health", :id => index_list}, options)
|
114
|
+
end
|
115
|
+
|
116
|
+
def cluster_state(options={})
|
117
|
+
standard_request(:get, {:index => "_cluster", :op => "state"})
|
118
|
+
end
|
119
|
+
|
120
|
+
def nodes_info(node_list, options={})
|
121
|
+
standard_request(:get, {:index => "_cluster", :type => "nodes", :id => node_list})
|
122
|
+
end
|
123
|
+
|
124
|
+
def nodes_stats(node_list, options={})
|
125
|
+
standard_request(:get, {:index => "_cluster", :type => "nodes", :id => node_list, :op => "stats"})
|
126
|
+
end
|
127
|
+
|
128
|
+
def shutdown_nodes(node_list, options={})
|
129
|
+
standard_request(:post, {:index => "_cluster", :type => "nodes", :id => node_list, :op => "_shutdown"}, options, "")
|
130
|
+
end
|
131
|
+
|
132
|
+
def restart_nodes(node_list, options={})
|
133
|
+
standard_request(:post, {:index => "_cluster", :type => "nodes", :id => node_list, :op => "_restart"}, options, "")
|
134
|
+
end
|
135
|
+
end
|
136
|
+
|
137
|
+
module ProtocolHelpers
|
138
|
+
private
|
139
|
+
|
140
|
+
def standard_request(*args)
|
141
|
+
response = request(*args)
|
142
|
+
handle_error(response) unless response.status == 200
|
143
|
+
encoder.decode(response.body)
|
144
|
+
end
|
145
|
+
|
146
|
+
def handle_error(response)
|
147
|
+
raise RequestError, "(#{response.status}) #{response.body}"
|
148
|
+
end
|
149
|
+
|
150
|
+
# :index - one or many index names
|
151
|
+
# :type - one or many types
|
152
|
+
# :id - one id
|
153
|
+
# :op - optional operation
|
154
|
+
def generate_uri(options)
|
155
|
+
path = ""
|
156
|
+
path << "/#{Array(options[:index]).collect { |i| escape(i.downcase) }.join(",")}" if options[:index] && !options[:index].empty?
|
157
|
+
path << "/#{Array(options[:type]).collect { |t| escape(t) }.join(",")}" if options[:type] && !options[:type].empty?
|
158
|
+
path << "/#{Array(options[:id]).collect { |id| escape(id) }.join(",")}" if options[:id] && !options[:id].empty?
|
159
|
+
path << "/#{options[:op]}" if options[:op]
|
160
|
+
path
|
161
|
+
end
|
162
|
+
|
163
|
+
#doesn't handle arrays or hashes or what have you
|
164
|
+
def generate_query_string(params)
|
165
|
+
params.collect { |k,v| "#{escape(k.to_s)}=#{escape(v.to_s)}" }.join("&")
|
166
|
+
end
|
167
|
+
|
168
|
+
def unescape_id!(hit)
|
169
|
+
hit["_id"] = unescape(hit["_id"])
|
170
|
+
nil
|
171
|
+
end
|
172
|
+
|
173
|
+
def set_encoding!(hit)
|
174
|
+
encode_utf8(hit["_source"]) if hit["_source"].is_a?(String)
|
175
|
+
nil
|
176
|
+
end
|
177
|
+
|
178
|
+
# faster than CGI.escape
|
179
|
+
# stolen from RSolr, who stole it from Rack
|
180
|
+
def escape(string)
|
181
|
+
string.to_s.gsub(/([^ a-zA-Z0-9_.-]+)/n) {
|
182
|
+
#'%'+$1.unpack('H2'*$1.size).join('%').upcase
|
183
|
+
'%'+$1.unpack('H2'*bytesize($1)).join('%').upcase
|
184
|
+
}.tr(' ', '+')
|
185
|
+
end
|
186
|
+
|
187
|
+
def unescape(string)
|
188
|
+
CGI.unescape(string)
|
189
|
+
end
|
190
|
+
|
191
|
+
if ''.respond_to?(:force_encoding) && ''.respond_to?(:encoding)
|
192
|
+
# encodes the string as utf-8 in Ruby 1.9
|
193
|
+
def encode_utf8(string)
|
194
|
+
# ElasticSearch only ever returns json in UTF-8 (per the JSON spec) so we can use force_encoding here (#TODO what about ids? can we assume those are always ascii?)
|
195
|
+
string.force_encoding(::Encoding::UTF_8)
|
196
|
+
end
|
197
|
+
else
|
198
|
+
# returns the unaltered string in Ruby 1.8
|
199
|
+
def encode_utf8(string)
|
200
|
+
string
|
201
|
+
end
|
202
|
+
end
|
203
|
+
|
204
|
+
# Return the bytesize of String; uses String#size under Ruby 1.8 and
|
205
|
+
# String#bytesize under 1.9.
|
206
|
+
if ''.respond_to?(:bytesize)
|
207
|
+
def bytesize(string)
|
208
|
+
string.bytesize
|
209
|
+
end
|
210
|
+
else
|
211
|
+
def bytesize(string)
|
212
|
+
string.size
|
213
|
+
end
|
214
|
+
end
|
215
|
+
end
|
216
|
+
|
217
|
+
module BaseProtocol
|
218
|
+
include IndexProtocol
|
219
|
+
include IndexAdminProtocol
|
220
|
+
include ClusterAdminProtocol
|
221
|
+
include ProtocolHelpers
|
222
|
+
|
223
|
+
def all_nodes
|
224
|
+
http_addresses = nodes_info([])["nodes"].collect { |id, node| node["http_address"] }
|
225
|
+
http_addresses.collect! do |a|
|
226
|
+
if a =~ /inet\[.*\/([\d.:]+)\]/
|
227
|
+
$1
|
228
|
+
end
|
229
|
+
end.compact!
|
230
|
+
http_addresses
|
231
|
+
end
|
232
|
+
|
233
|
+
end
|
234
|
+
end
|
235
|
+
end
|
@@ -0,0 +1,50 @@
|
|
1
|
+
require 'patron'
|
2
|
+
require 'cgi'
|
3
|
+
|
4
|
+
module ElasticSearch
|
5
|
+
module Transport
|
6
|
+
class HTTP < Base
|
7
|
+
|
8
|
+
DEFAULTS = {
|
9
|
+
:timeout => 5
|
10
|
+
}.freeze
|
11
|
+
|
12
|
+
def initialize(server, options={})
|
13
|
+
super
|
14
|
+
@options = DEFAULTS.merge(@options)
|
15
|
+
end
|
16
|
+
|
17
|
+
def connect!
|
18
|
+
@session = Patron::Session.new
|
19
|
+
@session.base_url = @server
|
20
|
+
@session.timeout = @options[:timeout]
|
21
|
+
@session.headers['User-Agent'] = 'ElasticSearch.rb v0.1'
|
22
|
+
end
|
23
|
+
|
24
|
+
private
|
25
|
+
|
26
|
+
def request(method, operation, params={}, body=nil, headers={})
|
27
|
+
begin
|
28
|
+
uri = generate_uri(operation)
|
29
|
+
query = generate_query_string(params)
|
30
|
+
path = [uri, query].join("?")
|
31
|
+
#puts "request: #{@server} #{path} #{body}"
|
32
|
+
response = @session.request(method, path, headers, :data => body)
|
33
|
+
handle_error(response) if response.status >= 500
|
34
|
+
response
|
35
|
+
rescue Exception => e
|
36
|
+
case e
|
37
|
+
when Patron::ConnectionFailed
|
38
|
+
raise ConnectionFailed
|
39
|
+
when Patron::HostResolutionError
|
40
|
+
raise HostResolutionError
|
41
|
+
when Patron::TimeoutError
|
42
|
+
raise TimeoutError
|
43
|
+
else
|
44
|
+
raise e
|
45
|
+
end
|
46
|
+
end
|
47
|
+
end
|
48
|
+
end
|
49
|
+
end
|
50
|
+
end
|
@@ -0,0 +1,96 @@
|
|
1
|
+
require 'thrift'
|
2
|
+
$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), "thrift"))
|
3
|
+
require 'rest'
|
4
|
+
|
5
|
+
module ElasticSearch
|
6
|
+
module Transport
|
7
|
+
class Thrift < Base
|
8
|
+
|
9
|
+
DEFAULTS = {
|
10
|
+
:timeout => 5,
|
11
|
+
:thrift_protocol => ::Thrift::BinaryProtocol,
|
12
|
+
:thrift_protocol_extra_params => [],
|
13
|
+
:thrift_transport => ::Thrift::Socket,
|
14
|
+
:thrift_transport_wrapper => ::Thrift::BufferedTransport,
|
15
|
+
:client_class => ElasticSearch::Thrift::Rest::Client
|
16
|
+
}.freeze
|
17
|
+
|
18
|
+
def initialize(server, options={})
|
19
|
+
super
|
20
|
+
@options = DEFAULTS.merge(@options)
|
21
|
+
end
|
22
|
+
|
23
|
+
def connect!
|
24
|
+
host, port = parse_server(@server)
|
25
|
+
|
26
|
+
@transport = @options[:thrift_transport].new(host, port.to_i, @options[:timeout])
|
27
|
+
@transport = @transport_wrapper.new(@transport) if @transport_wrapper
|
28
|
+
@transport.open
|
29
|
+
|
30
|
+
@client = @options[:client_class].new(@options[:thrift_protocol].new(@transport, *@options[:thrift_protocol_extra_params]))
|
31
|
+
rescue ::Thrift::TransportException, Errno::ECONNREFUSED
|
32
|
+
close
|
33
|
+
raise ConnectionFailed
|
34
|
+
end
|
35
|
+
|
36
|
+
def close
|
37
|
+
@transport.close rescue nil
|
38
|
+
end
|
39
|
+
|
40
|
+
private
|
41
|
+
|
42
|
+
def parse_server(server)
|
43
|
+
host, port = server.to_s.split(":")
|
44
|
+
raise ArgumentError, 'Servers must be in the form "host:port"' unless host and port
|
45
|
+
[host, port]
|
46
|
+
end
|
47
|
+
|
48
|
+
def stringify!(hash)
|
49
|
+
hash.keys.each do |k|
|
50
|
+
hash[k.to_s] = hash.delete(k).to_s
|
51
|
+
end
|
52
|
+
hash
|
53
|
+
end
|
54
|
+
|
55
|
+
def request(method, operation, params={}, body=nil, headers={})
|
56
|
+
begin
|
57
|
+
uri = generate_uri(operation)
|
58
|
+
#puts "request: #{@server} #{method} #{uri} #{params.inspect} #{body}"
|
59
|
+
request = ElasticSearch::Thrift::RestRequest.new
|
60
|
+
case method
|
61
|
+
when :get
|
62
|
+
request.method = ElasticSearch::Thrift::Method::GET
|
63
|
+
when :put
|
64
|
+
request.method = ElasticSearch::Thrift::Method::GET
|
65
|
+
when :post
|
66
|
+
request.method = ElasticSearch::Thrift::Method::POST
|
67
|
+
when :delete
|
68
|
+
request.method = ElasticSearch::Thrift::Method::DELETE
|
69
|
+
end
|
70
|
+
|
71
|
+
request.uri = uri
|
72
|
+
request.params = stringify!(params) #TODO this will change to parameters= in versions > 0.11.0
|
73
|
+
request.body = body
|
74
|
+
request.headers = stringify!(headers)
|
75
|
+
response = @client.execute(request)
|
76
|
+
handle_error(response) if response.status >= 500
|
77
|
+
response
|
78
|
+
rescue Exception => e
|
79
|
+
case e
|
80
|
+
when ::Thrift::TransportException
|
81
|
+
case e.type
|
82
|
+
when ::Thrift::TransportException::TIMED_OUT
|
83
|
+
raise TimeoutError
|
84
|
+
else
|
85
|
+
raise ConnectionFailed
|
86
|
+
end
|
87
|
+
#TODO Thrift::ApplicationException, Thrift::ProtocolException, IOError.. retryable or fatal?
|
88
|
+
else
|
89
|
+
raise e
|
90
|
+
end
|
91
|
+
end
|
92
|
+
end
|
93
|
+
|
94
|
+
end
|
95
|
+
end
|
96
|
+
end
|