dstk 0.50.1
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/dstk.rb +158 -0
- metadata +106 -0
data/lib/dstk.rb
ADDED
@@ -0,0 +1,158 @@
|
|
1
|
+
# Client interface to the Data Science Toolkit API
|
2
|
+
# See http://www.datasciencetoolkit.org/developerdocs for details
|
3
|
+
# By Pete Warden, pete@petewarden.com
|
4
|
+
|
5
|
+
require 'rubygems' if RUBY_VERSION < '1.9'
|
6
|
+
|
7
|
+
require 'json'
|
8
|
+
require 'httparty'
|
9
|
+
require 'httmultiparty'
|
10
|
+
|
11
|
+
module DSTK
|
12
|
+
class DSTK
|
13
|
+
def initialize(options = {})
|
14
|
+
default_options = {
|
15
|
+
:api_base => 'http://www.datasciencetoolkit.org',
|
16
|
+
:check_version => true,
|
17
|
+
}
|
18
|
+
|
19
|
+
if ENV['DSTK_API_BASE']
|
20
|
+
default_options[:api_base] = ENV['DSTK_API_BASE']
|
21
|
+
end
|
22
|
+
|
23
|
+
default_options.each do |key, value|
|
24
|
+
if !options.has_key?(key)
|
25
|
+
options[key] = value
|
26
|
+
end
|
27
|
+
end
|
28
|
+
|
29
|
+
@api_base = options[:api_base]
|
30
|
+
|
31
|
+
if options[:check_version]:
|
32
|
+
self.check_version()
|
33
|
+
end
|
34
|
+
end
|
35
|
+
|
36
|
+
# A short-hand method to URL encode a string. See http://web.elctech.com/?p=58
|
37
|
+
def u(str)
|
38
|
+
str.gsub(/[^a-zA-Z0-9_\.\-]/n) {|s| sprintf('%%%02x', s[0]) }
|
39
|
+
end
|
40
|
+
|
41
|
+
def json_api_call(endpoint, arguments = {}, data_payload = nil, data_payload_type = 'json')
|
42
|
+
|
43
|
+
api_url = @api_base + endpoint
|
44
|
+
arguments_list = arguments.map do |name, value|
|
45
|
+
name + '=' + u(value)
|
46
|
+
end
|
47
|
+
if arguments_list.length > 0
|
48
|
+
arguments_string = '?' + arguments_list.join('&')
|
49
|
+
api_url += arguments_string
|
50
|
+
end
|
51
|
+
response = nil
|
52
|
+
if !data_payload
|
53
|
+
response = HTTParty.get(api_url)
|
54
|
+
else
|
55
|
+
if data_payload_type == 'json'
|
56
|
+
data_payload_value = data_payload.to_json
|
57
|
+
else
|
58
|
+
data_payload_value = data_payload
|
59
|
+
end
|
60
|
+
response = HTTParty.post(api_url, { :body => data_payload_value })
|
61
|
+
end
|
62
|
+
|
63
|
+
if !response.body or response.code != 200
|
64
|
+
raise "DSTK::json_api_call('#{endpoint}', #{arguments.to_json}, '#{data_payload}', '#{data_payload_type}') call to '#{api_url}' failed with code #{response.code} : '#{response.message}'"
|
65
|
+
end
|
66
|
+
|
67
|
+
json_string = response.body
|
68
|
+
result = nil
|
69
|
+
begin
|
70
|
+
result = JSON.parse(json_string)
|
71
|
+
rescue JSON::ParseError => e
|
72
|
+
raise "DSTK::json_api_call('#{endpoint}', #{arguments.to_json}, '#{data_payload}', '#{data_payload_type}') call to '#{api_url}' failed to parse response '#{json_string}' as JSON - #{e.message}"
|
73
|
+
end
|
74
|
+
if !result.is_a?(Array) and result['error']
|
75
|
+
raise result['error']
|
76
|
+
end
|
77
|
+
result
|
78
|
+
end
|
79
|
+
|
80
|
+
def check_version
|
81
|
+
required_version = 50
|
82
|
+
response = json_api_call('/info')
|
83
|
+
actual_version = response['version']
|
84
|
+
if actual_version < required_version:
|
85
|
+
raise "DSTK: Version #{actual_version.to_s} found but #{required_version.to_s} is required"
|
86
|
+
end
|
87
|
+
end
|
88
|
+
|
89
|
+
def ip2coordinates(ips)
|
90
|
+
if !ips.is_a?(Array) then ips = [ips] end
|
91
|
+
response = json_api_call('/ip2coordinates', {}, ips)
|
92
|
+
response
|
93
|
+
end
|
94
|
+
|
95
|
+
def street2coordinates(addresses)
|
96
|
+
if !addresses.is_a?(Array) then addresses = [addresses] end
|
97
|
+
response = json_api_call('/street2coordinates', {}, addresses)
|
98
|
+
response
|
99
|
+
end
|
100
|
+
|
101
|
+
def coordinates2politics(coordinates)
|
102
|
+
response = json_api_call('/coordinates2politics', {}, coordinates)
|
103
|
+
response
|
104
|
+
end
|
105
|
+
|
106
|
+
def text2places(text)
|
107
|
+
response = json_api_call('/text2places', {}, text, 'string')
|
108
|
+
response
|
109
|
+
end
|
110
|
+
|
111
|
+
def file2text(inputfile)
|
112
|
+
response = json_api_call('/text2places', {}, {:inputfile => inputfile}, 'file')
|
113
|
+
response
|
114
|
+
end
|
115
|
+
|
116
|
+
def text2sentences(text)
|
117
|
+
response = json_api_call('/text2sentences', {}, text, 'string')
|
118
|
+
response
|
119
|
+
end
|
120
|
+
|
121
|
+
def html2text(html)
|
122
|
+
response = json_api_call('/html2text', {}, html, 'string')
|
123
|
+
response
|
124
|
+
end
|
125
|
+
|
126
|
+
def html2story(html)
|
127
|
+
response = json_api_call('/html2story', {}, html, 'string')
|
128
|
+
response
|
129
|
+
end
|
130
|
+
|
131
|
+
def text2people(text)
|
132
|
+
response = json_api_call('/text2people', {}, text, 'string')
|
133
|
+
response
|
134
|
+
end
|
135
|
+
|
136
|
+
def text2times(text)
|
137
|
+
response = json_api_call('/text2times', {}, text, 'string')
|
138
|
+
response
|
139
|
+
end
|
140
|
+
|
141
|
+
def text2sentiment(text)
|
142
|
+
response = json_api_call('/text2sentiment', {}, text, 'string')
|
143
|
+
response
|
144
|
+
end
|
145
|
+
|
146
|
+
def coordinates2statistics(coordinates, statistics = nil)
|
147
|
+
if statistics
|
148
|
+
if !statistics.is_a?(Array) then statistics = [statistics] end
|
149
|
+
arguments = { 'statistics' => statistics.join(',') }
|
150
|
+
else
|
151
|
+
arguments = {}
|
152
|
+
end
|
153
|
+
response = json_api_call('/coordinates2statistics', arguments, coordinates)
|
154
|
+
response
|
155
|
+
end
|
156
|
+
|
157
|
+
end
|
158
|
+
end
|
metadata
ADDED
@@ -0,0 +1,106 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: dstk
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
hash: 213
|
5
|
+
prerelease:
|
6
|
+
segments:
|
7
|
+
- 0
|
8
|
+
- 50
|
9
|
+
- 1
|
10
|
+
version: 0.50.1
|
11
|
+
platform: ruby
|
12
|
+
authors:
|
13
|
+
- Pete Warden
|
14
|
+
autorequire:
|
15
|
+
bindir: bin
|
16
|
+
cert_chain: []
|
17
|
+
|
18
|
+
date: 2013-05-14 00:00:00 Z
|
19
|
+
dependencies:
|
20
|
+
- !ruby/object:Gem::Dependency
|
21
|
+
name: json
|
22
|
+
prerelease: false
|
23
|
+
requirement: &id001 !ruby/object:Gem::Requirement
|
24
|
+
none: false
|
25
|
+
requirements:
|
26
|
+
- - ">="
|
27
|
+
- !ruby/object:Gem::Version
|
28
|
+
hash: 3
|
29
|
+
segments:
|
30
|
+
- 0
|
31
|
+
version: "0"
|
32
|
+
type: :runtime
|
33
|
+
version_requirements: *id001
|
34
|
+
- !ruby/object:Gem::Dependency
|
35
|
+
name: httparty
|
36
|
+
prerelease: false
|
37
|
+
requirement: &id002 !ruby/object:Gem::Requirement
|
38
|
+
none: false
|
39
|
+
requirements:
|
40
|
+
- - ">="
|
41
|
+
- !ruby/object:Gem::Version
|
42
|
+
hash: 3
|
43
|
+
segments:
|
44
|
+
- 0
|
45
|
+
version: "0"
|
46
|
+
type: :runtime
|
47
|
+
version_requirements: *id002
|
48
|
+
- !ruby/object:Gem::Dependency
|
49
|
+
name: httmultiparty
|
50
|
+
prerelease: false
|
51
|
+
requirement: &id003 !ruby/object:Gem::Requirement
|
52
|
+
none: false
|
53
|
+
requirements:
|
54
|
+
- - ">="
|
55
|
+
- !ruby/object:Gem::Version
|
56
|
+
hash: 3
|
57
|
+
segments:
|
58
|
+
- 0
|
59
|
+
version: "0"
|
60
|
+
type: :runtime
|
61
|
+
version_requirements: *id003
|
62
|
+
description: An interface to the datasciencetoolkit.org open API for geocoding addresses, extracting entities and sentiment from unstructured text, and other common semantic and geo data tasks.
|
63
|
+
email: pete@petewarden.com
|
64
|
+
executables: []
|
65
|
+
|
66
|
+
extensions: []
|
67
|
+
|
68
|
+
extra_rdoc_files: []
|
69
|
+
|
70
|
+
files:
|
71
|
+
- lib/dstk.rb
|
72
|
+
homepage: http://rubygems.org/gems/dstk
|
73
|
+
licenses: []
|
74
|
+
|
75
|
+
post_install_message:
|
76
|
+
rdoc_options: []
|
77
|
+
|
78
|
+
require_paths:
|
79
|
+
- lib
|
80
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
81
|
+
none: false
|
82
|
+
requirements:
|
83
|
+
- - ">="
|
84
|
+
- !ruby/object:Gem::Version
|
85
|
+
hash: 3
|
86
|
+
segments:
|
87
|
+
- 0
|
88
|
+
version: "0"
|
89
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
90
|
+
none: false
|
91
|
+
requirements:
|
92
|
+
- - ">="
|
93
|
+
- !ruby/object:Gem::Version
|
94
|
+
hash: 3
|
95
|
+
segments:
|
96
|
+
- 0
|
97
|
+
version: "0"
|
98
|
+
requirements: []
|
99
|
+
|
100
|
+
rubyforge_project:
|
101
|
+
rubygems_version: 1.8.24
|
102
|
+
signing_key:
|
103
|
+
specification_version: 3
|
104
|
+
summary: Data Science Toolkit client
|
105
|
+
test_files: []
|
106
|
+
|