fluent-plugin-datahub 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/Gemfile +3 -0
- data/README.md +38 -0
- data/Rakefile +14 -0
- data/VERSION +1 -0
- data/build.sh +11 -0
- data/fluent-plugin-datahub.gemspec +22 -0
- data/lib/fluent/plugin/datahub/datahub-client.rb +27 -0
- data/lib/fluent/plugin/datahub/datahub-http-client-test.rb +343 -0
- data/lib/fluent/plugin/datahub/datahub-http-client.rb +229 -0
- data/lib/fluent/plugin/datahub/datahub-project.rb +59 -0
- data/lib/fluent/plugin/datahub/datahub-put-record-result.rb +23 -0
- data/lib/fluent/plugin/datahub/datahub-record-entity.rb +136 -0
- data/lib/fluent/plugin/datahub/datahub-record-schema.rb +73 -0
- data/lib/fluent/plugin/datahub/datahub-shard.rb +13 -0
- data/lib/fluent/plugin/datahub/datahub-topic.rb +73 -0
- data/lib/fluent/plugin/out_datahub.rb +402 -0
- data/sample/csv_sample.conf +22 -0
- data/sample/csv_sample.csv +14 -0
- data/sample/log_sample.conf +17 -0
- data/sample/log_sample.log +1 -0
- metadata +97 -0
@@ -0,0 +1,229 @@
|
|
1
|
+
require "net/http"
|
2
|
+
require "openssl"
|
3
|
+
require "base64"
|
4
|
+
require "date"
|
5
|
+
require "time"
|
6
|
+
require "json"
|
7
|
+
require_relative "datahub-record-schema"
|
8
|
+
require_relative "datahub-record-entity"
|
9
|
+
|
10
|
+
########################
|
11
|
+
## @author pholy.ht ##
|
12
|
+
## @time 2016-03-24 ##
|
13
|
+
########################
|
14
|
+
class DatahubHttpClient
|
15
|
+
attr_accessor :access_id
|
16
|
+
attr_accessor :access_key
|
17
|
+
attr_accessor :endpoint
|
18
|
+
|
19
|
+
def initialize(endpoint, access_id, access_key)
|
20
|
+
@endpoint = endpoint
|
21
|
+
@access_id = access_id
|
22
|
+
@access_key = access_key
|
23
|
+
end
|
24
|
+
|
25
|
+
# Signature = base64(HmacSha1(AccessKeySecret, VERB + "\n"
|
26
|
+
# + CONTENT-TYPE + "\n"
|
27
|
+
# + DATE + "\n"
|
28
|
+
# + CanonicalizedDatahubHeaders + "\n"
|
29
|
+
# + CanonicalizedResource))
|
30
|
+
|
31
|
+
def get_signature(params)
|
32
|
+
verb = params["verb"]
|
33
|
+
content_type = params["content-type"]
|
34
|
+
resource = params["resource"]
|
35
|
+
gmt_time = params["date"]
|
36
|
+
data = verb + "\n" + content_type + "\n" + gmt_time + "\n" + "x-datahub-client-version:1\n" + resource
|
37
|
+
# data = "GET\napplication/json\nFri, 06 May 2016 06:43:31 GMT\nx-datahub-client-version:1\n/projects/test_project/topics/datahub_fluentd_out_1"
|
38
|
+
return "DATAHUB " + @access_id + ":" + Base64.encode64("#{OpenSSL::HMAC.digest('sha1', @access_key, data)}").chomp
|
39
|
+
end
|
40
|
+
|
41
|
+
def send_request(method, path, params={}, headers={})
|
42
|
+
uri = URI(@endpoint + path)
|
43
|
+
http_req = nil
|
44
|
+
if method == "GET"
|
45
|
+
http_req = Net::HTTP::Get.new(uri.path)
|
46
|
+
elsif method == "POST"
|
47
|
+
http_req = Net::HTTP::Post.new(uri.path)
|
48
|
+
# start_time = Time.now.to_f
|
49
|
+
# puts "start_time :" + start_time.to_s
|
50
|
+
http_req.body = JSON.generate(params)
|
51
|
+
# end_time = Time.now.to_f
|
52
|
+
# puts "end_time :" + end_time.to_s
|
53
|
+
# puts ((end_time - start_time)*1000).to_s
|
54
|
+
elsif method == "PUT"
|
55
|
+
http_req = Net::HTTP::Put.new(uri.path)
|
56
|
+
http_req.body = JSON.generate(params)
|
57
|
+
elsif method == "DELETE"
|
58
|
+
http_req = Net::HTTP::Delete.new(uri.path)
|
59
|
+
else
|
60
|
+
raise "Unsupported method: " + method
|
61
|
+
end
|
62
|
+
|
63
|
+
sig_params = {}
|
64
|
+
sig_params["verb"] = method
|
65
|
+
sig_params["content-type"] = "application/json"
|
66
|
+
sig_params["resource"] = path
|
67
|
+
sig_params["date"] = Time.now.utc.strftime("%a, %d %b %Y %T") + " GMT"
|
68
|
+
|
69
|
+
http_req["Authorization"] = get_signature(sig_params)
|
70
|
+
http_req["x-datahub-client-version"] = 1
|
71
|
+
http_req["Date"] = sig_params["date"]
|
72
|
+
http_req["Content-Type"] = sig_params["content-type"]
|
73
|
+
http_req["ACCEPT-ENCODING"] = nil
|
74
|
+
|
75
|
+
# puts http_req["Date"]
|
76
|
+
# puts http_req["Authorization"]
|
77
|
+
# puts http_req.to_hash
|
78
|
+
|
79
|
+
headers.each do |k, v|
|
80
|
+
http_req[k] = v
|
81
|
+
end
|
82
|
+
|
83
|
+
# puts path
|
84
|
+
|
85
|
+
return Net::HTTP.start(uri.host, uri.port) {
|
86
|
+
|client|
|
87
|
+
http_resp = client.request(http_req)
|
88
|
+
if http_resp.code != '200' and http_resp.code != '201'
|
89
|
+
if http_resp.body.empty?
|
90
|
+
raise "Send request failed, unknown response "
|
91
|
+
end
|
92
|
+
raise http_resp.body
|
93
|
+
end
|
94
|
+
result = http_resp.body
|
95
|
+
if result.empty?
|
96
|
+
return nil
|
97
|
+
end
|
98
|
+
return JSON.parse(result)
|
99
|
+
}
|
100
|
+
end
|
101
|
+
|
102
|
+
def list_projects()
|
103
|
+
path = "/projects"
|
104
|
+
return send_request("GET", path)
|
105
|
+
end
|
106
|
+
|
107
|
+
def get_project(project_name)
|
108
|
+
path = "/projects/" + project_name
|
109
|
+
return send_request("GET", path)
|
110
|
+
end
|
111
|
+
|
112
|
+
def create_project(project_name, comment)
|
113
|
+
path = "/projects/" + project_name
|
114
|
+
params = {}
|
115
|
+
params["Comment"] = comment
|
116
|
+
return send_request("POST", path, params)
|
117
|
+
end
|
118
|
+
|
119
|
+
# datahub目前未实现update操作
|
120
|
+
# def update_project(project_name, comment)
|
121
|
+
# path = "/projects/" + project_name
|
122
|
+
# params = {}
|
123
|
+
# params["Comment"] = comment
|
124
|
+
# return send_request("PUT", path, params)
|
125
|
+
# end
|
126
|
+
|
127
|
+
def delete_project(project_name)
|
128
|
+
path = "/projects/" + project_name
|
129
|
+
return send_request("DELETE", path)
|
130
|
+
end
|
131
|
+
|
132
|
+
def list_topics(project_name)
|
133
|
+
path = "/projects/" + project_name + "/topics"
|
134
|
+
return send_request("GET", path)
|
135
|
+
end
|
136
|
+
|
137
|
+
def get_topic(project_name, topic_name)
|
138
|
+
path = "/projects/" + project_name + "/topics/" + topic_name
|
139
|
+
return send_request("GET", path)
|
140
|
+
end
|
141
|
+
|
142
|
+
def create_topic(project_name, topic_name, shard_count, lifecycle, record_type, record_schema, comment)
|
143
|
+
path = "/projects/" + project_name + "/topics/" + topic_name
|
144
|
+
params = {}
|
145
|
+
params["ShardCount"] = shard_count
|
146
|
+
params["Lifecycle"] = lifecycle
|
147
|
+
params["RecordType"] = record_type
|
148
|
+
params["RecordSchema"] = JSON.generate(record_schema)
|
149
|
+
params["Comment"] = comment
|
150
|
+
return send_request("POST", path, params)
|
151
|
+
end
|
152
|
+
|
153
|
+
def update_topic(project_name, topic_name, lifecycle, comment)
|
154
|
+
path = "/projects/" + project_name + "/topics/" + topic_name
|
155
|
+
params = {}
|
156
|
+
params["Lifecycle"] = lifecycle
|
157
|
+
params["Comment"] = comment
|
158
|
+
return send_request("PUT", path, params)
|
159
|
+
end
|
160
|
+
|
161
|
+
def delete_topic(project_name, topic_name)
|
162
|
+
path = "/projects/" + project_name + "/topics/" + topic_name
|
163
|
+
return send_request("DELETE", path)
|
164
|
+
end
|
165
|
+
|
166
|
+
def list_shards(project_name, topic_name)
|
167
|
+
path = "/projects/" + project_name + "/topics/" + topic_name + "/shards"
|
168
|
+
return send_request("GET", path)
|
169
|
+
end
|
170
|
+
|
171
|
+
##########################################################
|
172
|
+
## shard的merge、split未实现
|
173
|
+
## datahub这一期并没有实现,所以api不开放
|
174
|
+
##########################################################
|
175
|
+
# def merge_shard(project_name, topic_name, shard_id, adjacent_shard_id)
|
176
|
+
# path = "/projects/" + project_name + "/topics/" + topic_name + "/shards"
|
177
|
+
# params = {}
|
178
|
+
# params["Action"] = "merge"
|
179
|
+
# params["ShardId"] = shard_id
|
180
|
+
# params["AdjacentShardId"] = adjacent_shard_id
|
181
|
+
# return send_request("POST", path, params)
|
182
|
+
# end
|
183
|
+
|
184
|
+
# def split_shard(project_name, topic_name, shard_id, split_key)
|
185
|
+
# path = "/projects/" + project_name + "/topics/" + topic_name + "/shards"
|
186
|
+
# params = {}
|
187
|
+
# params["Action"] = "split"
|
188
|
+
# params["ShardId"] = shard_id
|
189
|
+
# params["SplitKey"] = split_key
|
190
|
+
# return send_request("POST", path, params)
|
191
|
+
# end
|
192
|
+
|
193
|
+
def write_data_to_topic(project_name, topic_name, record_entities)
|
194
|
+
path = "/projects/" + project_name + "/topics/" + topic_name + "/shards"
|
195
|
+
params = {}
|
196
|
+
params["Action"] = "pub"
|
197
|
+
params["Records"] = record_entities
|
198
|
+
return send_request("POST", path, params)
|
199
|
+
end
|
200
|
+
|
201
|
+
def get_shard_cursor(project_name, topic_name, shard_id, offset=DateTime.now.strftime('%Q'), type="LATEST")
|
202
|
+
path = "/projects/" + project_name + "/topics/" + topic_name + "/shards/" + shard_id
|
203
|
+
params = {}
|
204
|
+
params["Action"] = "cursor"
|
205
|
+
params["SystemTime"] = offset
|
206
|
+
params["Type"] = type
|
207
|
+
return send_request("POST", path, params)
|
208
|
+
end
|
209
|
+
|
210
|
+
def read_data_from_shard_with_cursor(project_name, topic_name, shard_id, cursor, count)
|
211
|
+
path = "/projects/" + project_name + "/topics/" + topic_name + "/shards/" + shard_id
|
212
|
+
params = {}
|
213
|
+
params["Action"] = "sub"
|
214
|
+
params["Cursor"] = cursor
|
215
|
+
params["Limit"] = count
|
216
|
+
return send_request("POST", path, params)
|
217
|
+
end
|
218
|
+
|
219
|
+
def read_data_from_shard(project_name, topic_name, shard_id, count, offset=DateTime.now.strftime('%Q'), type="LATEST")
|
220
|
+
path = "/projects/" + project_name + "/topics/" + topic_name + "/shards/" + shard_id
|
221
|
+
cursor = get_shard_cursor(project_name, topic_name, shard_id, offset, type).to_hash["Cursor"]
|
222
|
+
params = {}
|
223
|
+
params["Action"] = "sub"
|
224
|
+
params["Cursor"] = cursor
|
225
|
+
params["Limit"] = count
|
226
|
+
return send_request("POST", path, params)
|
227
|
+
end
|
228
|
+
|
229
|
+
end
|
@@ -0,0 +1,59 @@
|
|
1
|
+
require_relative "datahub-http-client"
|
2
|
+
require_relative "datahub-topic"
|
3
|
+
|
4
|
+
class DatahubProject
|
5
|
+
def initialize(datahub_http_client, project_name)
|
6
|
+
@client = datahub_http_client
|
7
|
+
@project_name = project_name
|
8
|
+
end
|
9
|
+
|
10
|
+
def list_topics()
|
11
|
+
topics_map = @client.list_topics
|
12
|
+
topics_array = topics_map["TopicNames"]
|
13
|
+
topics = []
|
14
|
+
for i in 0...topics_array.size
|
15
|
+
topic_name = topics_array[i]
|
16
|
+
topic = DatahubTopic.new(datahub_http_client, @project_name, topic_name)
|
17
|
+
topics.push(topic)
|
18
|
+
end
|
19
|
+
|
20
|
+
return topics
|
21
|
+
end
|
22
|
+
|
23
|
+
def create_topic(topic_name, shard_count, lifecycle, record_type, record_schema, comment)
|
24
|
+
@client.create_topic(@project_name, topic_name, shard_count, lifecycle, record_type, record_schema, comment)
|
25
|
+
return self.get_topic(topic_name)
|
26
|
+
end
|
27
|
+
|
28
|
+
def update_topic(topic_name, lifecycle, desc)
|
29
|
+
@client.update_topic(@project_name, topic_name, lifecycle, desc)
|
30
|
+
end
|
31
|
+
|
32
|
+
def delete_topic(topic_name)
|
33
|
+
@client.delete_topic(@project_name, topic_name)
|
34
|
+
end
|
35
|
+
|
36
|
+
def get_topic(topic_name)
|
37
|
+
result_map = @client.get_topic(@project_name, topic_name)
|
38
|
+
record_schema_string = result_map["RecordSchema"]
|
39
|
+
record_schema_map = JSON.parse(record_schema_string)
|
40
|
+
fields = record_schema_map["fields"]
|
41
|
+
record_schema = RecordSchema.new()
|
42
|
+
for i in 0...fields.size
|
43
|
+
field = fields[i]
|
44
|
+
record_field = RecordField.new(field["name"], field["type"])
|
45
|
+
record_schema.add_field(record_field)
|
46
|
+
end
|
47
|
+
topic = DatahubTopic.new(@client, @project_name, topic_name)
|
48
|
+
topic.shard_count = result_map["ShardCount"]
|
49
|
+
topic.lifecycle = result_map["Lifecycle"]
|
50
|
+
topic.record_type = result_map["RecordType"]
|
51
|
+
topic.record_schema = record_schema
|
52
|
+
topic.comment = result_map["Comment"]
|
53
|
+
topic.create_time = result_map["CreateTime"]
|
54
|
+
topic.last_modify_time = result_map["LastModifyTime"]
|
55
|
+
|
56
|
+
return topic
|
57
|
+
end
|
58
|
+
|
59
|
+
end
|
@@ -0,0 +1,23 @@
|
|
1
|
+
|
2
|
+
class PutRecordResult
|
3
|
+
attr_accessor :failed_record_count
|
4
|
+
attr_accessor :failed_record_index
|
5
|
+
attr_accessor :failed_record_error
|
6
|
+
attr_accessor :failed_record_list
|
7
|
+
|
8
|
+
def initialize
|
9
|
+
@failed_record_count = 0
|
10
|
+
@failed_record_index = []
|
11
|
+
@failed_record_error = []
|
12
|
+
@failed_record_list = []
|
13
|
+
end
|
14
|
+
|
15
|
+
def to_json(*a)
|
16
|
+
{
|
17
|
+
"failed_record_count" => @failed_record_count,
|
18
|
+
"failed_record_index" => @failed_record_index,
|
19
|
+
"failed_record_error" => @failed_record_error,
|
20
|
+
"failed_record_list" => @failed_record_list
|
21
|
+
}.to_json(*a)
|
22
|
+
end
|
23
|
+
end
|
@@ -0,0 +1,136 @@
|
|
1
|
+
|
2
|
+
class RecordEntity
|
3
|
+
$STRING_MAX_LENGTH = 1 * 1024 * 1024
|
4
|
+
|
5
|
+
def initialize(schema)
|
6
|
+
@columns_map = {}
|
7
|
+
|
8
|
+
@shard_id = nil
|
9
|
+
@attributes = {}
|
10
|
+
|
11
|
+
@schema = schema
|
12
|
+
end
|
13
|
+
|
14
|
+
def set(name, value)
|
15
|
+
@columns_map[name] = value
|
16
|
+
end
|
17
|
+
|
18
|
+
def set_shard_id(shard_id)
|
19
|
+
if shard_id.is_a?String
|
20
|
+
@shard_id = shard_id
|
21
|
+
else
|
22
|
+
raise "shard_id must be String type!"
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
def add_attribute(name, value)
|
27
|
+
@attributes[name] = value
|
28
|
+
end
|
29
|
+
|
30
|
+
def get_columns_map
|
31
|
+
return @columns_map
|
32
|
+
end
|
33
|
+
|
34
|
+
def get_schema
|
35
|
+
return @schema
|
36
|
+
end
|
37
|
+
|
38
|
+
|
39
|
+
def setBigInt(name, value)
|
40
|
+
if value == nil
|
41
|
+
set(name, value)
|
42
|
+
elsif value.is_a?Integer
|
43
|
+
set(name, value)
|
44
|
+
elsif value.is_a?String and value.to_i.to_s == value
|
45
|
+
set(name, value.to_i)
|
46
|
+
else
|
47
|
+
raise "value show be Integer, name:" + name.to_s + " value:" + value.to_s
|
48
|
+
end
|
49
|
+
end
|
50
|
+
|
51
|
+
def setDouble(name, value)
|
52
|
+
if value == nil
|
53
|
+
set(name, value)
|
54
|
+
elsif value.is_a?Float
|
55
|
+
set(name, value)
|
56
|
+
elsif value.is_a?String
|
57
|
+
begin
|
58
|
+
set(name, Float(value))
|
59
|
+
rescue
|
60
|
+
raise "value show be Float, name:" + name.to_s + " value:" + value.to_s
|
61
|
+
end
|
62
|
+
else
|
63
|
+
raise "value show be Float, name:" + name.to_s + " value:" + value.to_s
|
64
|
+
end
|
65
|
+
end
|
66
|
+
|
67
|
+
def setBoolean(name, value)
|
68
|
+
if value == nil
|
69
|
+
set(name, value)
|
70
|
+
elsif value.is_a?String
|
71
|
+
if value == "true"
|
72
|
+
set(name, true)
|
73
|
+
elsif value == "false"
|
74
|
+
set(name, false)
|
75
|
+
else
|
76
|
+
raise "value must be true or false, name:" + name.to_s + " value:" + value.to_s
|
77
|
+
end
|
78
|
+
elsif value != false and value != true
|
79
|
+
raise "value must be bool or string[true,false], name:" + name.to_s + " value:" + value.to_s
|
80
|
+
end
|
81
|
+
set(name, value)
|
82
|
+
end
|
83
|
+
|
84
|
+
def setTimeStamp(name, value)
|
85
|
+
if value == nil
|
86
|
+
set(name, value)
|
87
|
+
elsif value.is_a?Integer
|
88
|
+
set(name, value)
|
89
|
+
elsif value.is_a?String and value.to_i.to_s == value
|
90
|
+
set(name, value.to_i)
|
91
|
+
else
|
92
|
+
raise "value should be Integer, name:" + name.to_s + " value:" + value.to_s
|
93
|
+
end
|
94
|
+
end
|
95
|
+
|
96
|
+
def setString(name, value)
|
97
|
+
if value == nil
|
98
|
+
set(name, value)
|
99
|
+
elsif value.is_a?String and value.length < $STRING_MAX_LENGTH
|
100
|
+
if @schema.get_encoding == nil
|
101
|
+
set(name, value)
|
102
|
+
else
|
103
|
+
# puts "encoding:" + @schema.get_encoding.to_s
|
104
|
+
value.force_encoding(@schema.get_encoding)
|
105
|
+
set(name, value)
|
106
|
+
# puts value
|
107
|
+
end
|
108
|
+
else
|
109
|
+
raise "value show be String and len < " + $STRING_MAX_LENGTH.to_s + ", name:" + name.to_s + " value:" + value.to_s
|
110
|
+
end
|
111
|
+
end
|
112
|
+
|
113
|
+
def get_stored_column_values
|
114
|
+
data = []
|
115
|
+
fields = @schema.get_fields
|
116
|
+
for i in 0...fields.size
|
117
|
+
field = fields[i]
|
118
|
+
name = field.get_name
|
119
|
+
if @columns_map[name] == nil
|
120
|
+
data.push(@columns_map[name])
|
121
|
+
elsif
|
122
|
+
data.push(@columns_map[name].to_s)
|
123
|
+
end
|
124
|
+
end
|
125
|
+
return data
|
126
|
+
end
|
127
|
+
|
128
|
+
def to_json(*a)
|
129
|
+
{
|
130
|
+
"Data" => get_stored_column_values,
|
131
|
+
"ShardId" => @shard_id,
|
132
|
+
"Attributes" => @attributes
|
133
|
+
}.to_json(*a)
|
134
|
+
end
|
135
|
+
|
136
|
+
end
|