opener-webservice 2.0.0 → 2.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/README.md +135 -17
- data/lib/opener/webservice/configuration.rb +90 -0
- data/lib/opener/webservice/error_handler.rb +29 -0
- data/lib/opener/webservice/input_extractor.rb +43 -0
- data/lib/opener/webservice/input_sanitizer.rb +65 -0
- data/lib/opener/webservice/option_parser.rb +175 -0
- data/lib/opener/webservice/server.rb +297 -0
- data/lib/opener/webservice/uploader.rb +50 -0
- data/lib/opener/webservice/version.rb +4 -6
- data/lib/opener/webservice.rb +18 -417
- data/opener-webservice.gemspec +18 -10
- metadata +109 -5
- data/lib/opener/webservice/opt_parser.rb +0 -103
@@ -0,0 +1,297 @@
|
|
1
|
+
module Opener
|
2
|
+
module Webservice
|
3
|
+
##
|
4
|
+
# The meat of the webservices: the actual Sinatra application. Components
|
5
|
+
# should extend this class and configure it (e.g. to specify what component
|
6
|
+
# class to use).
|
7
|
+
#
|
8
|
+
class Server < Sinatra::Base
|
9
|
+
##
|
10
|
+
# List of fields that can contain input to process.
|
11
|
+
#
|
12
|
+
# @return [Array]
|
13
|
+
#
|
14
|
+
INPUT_FIELDS = %w{input input_url}
|
15
|
+
|
16
|
+
##
|
17
|
+
# Sets the accepted component parameters. Parameter names are always
|
18
|
+
# stored as symbols.
|
19
|
+
#
|
20
|
+
# @param [Array] params
|
21
|
+
#
|
22
|
+
def self.accepted_params=(params)
|
23
|
+
@accepted_params = params.map(&:to_sym)
|
24
|
+
end
|
25
|
+
|
26
|
+
##
|
27
|
+
# Returns the accepted component parameters.
|
28
|
+
#
|
29
|
+
# @return [Array]
|
30
|
+
#
|
31
|
+
def self.accepted_params
|
32
|
+
return @accepted_params ||= []
|
33
|
+
end
|
34
|
+
|
35
|
+
##
|
36
|
+
# Sets the text processor to use.
|
37
|
+
#
|
38
|
+
# @param [Class] processor
|
39
|
+
#
|
40
|
+
def self.text_processor=(processor)
|
41
|
+
@text_processor = processor
|
42
|
+
end
|
43
|
+
|
44
|
+
##
|
45
|
+
# Returns the text processor to use.
|
46
|
+
#
|
47
|
+
# @return [Class]
|
48
|
+
#
|
49
|
+
def self.text_processor
|
50
|
+
return @text_processor
|
51
|
+
end
|
52
|
+
|
53
|
+
configure :production do
|
54
|
+
set :raise_errors, false
|
55
|
+
set :dump_errors, false
|
56
|
+
end
|
57
|
+
|
58
|
+
error do
|
59
|
+
Rollbar.report_exception(env['sinatra.error'])
|
60
|
+
|
61
|
+
halt(
|
62
|
+
500,
|
63
|
+
'An error occurred. A team of garden gnomes has been dispatched to ' \
|
64
|
+
'look into the problem.',
|
65
|
+
)
|
66
|
+
end
|
67
|
+
|
68
|
+
# Require authentication for non static files if authentication is
|
69
|
+
# enabled.
|
70
|
+
before %r{^((?!.css|.jpg|.png|.js|.ico).)+$} do
|
71
|
+
authenticate! if Configuration.authentication?
|
72
|
+
end
|
73
|
+
|
74
|
+
##
|
75
|
+
# Shows a form that allows users to submit data directly from their
|
76
|
+
# browser.
|
77
|
+
#
|
78
|
+
get '/' do
|
79
|
+
erb :index
|
80
|
+
end
|
81
|
+
|
82
|
+
##
|
83
|
+
# Processes the input using a component.
|
84
|
+
#
|
85
|
+
# Data can be submitted in two ways:
|
86
|
+
#
|
87
|
+
# 1. As regular POST fields
|
88
|
+
# 2. A single JSON object as the POST body
|
89
|
+
#
|
90
|
+
# When submitting data, you can use the following fields (either as POST
|
91
|
+
# fields or as the fields of a JSON object):
|
92
|
+
#
|
93
|
+
# | Field | Description |
|
94
|
+
# |:---------------|:--------------------------------------------|
|
95
|
+
# | input | The raw input text/KAF to process |
|
96
|
+
# | input_url | A URL to a document to download and process |
|
97
|
+
# | callbacks | An array of callback URLs |
|
98
|
+
# | error_callback | A URL to submit errors to |
|
99
|
+
# | request_id | A unique ID to associate with the document |
|
100
|
+
# | metadata | A custom metadata object to store in S3 |
|
101
|
+
#
|
102
|
+
# In case of a JSON object the input body would look something like the
|
103
|
+
# following:
|
104
|
+
#
|
105
|
+
# {"input": "Hello world, this is....", request_id: "123abc"}
|
106
|
+
#
|
107
|
+
post '/' do
|
108
|
+
if json_input?
|
109
|
+
options = params_from_json
|
110
|
+
else
|
111
|
+
options = params
|
112
|
+
end
|
113
|
+
|
114
|
+
options = InputSanitizer.new.prepare_parameters(options)
|
115
|
+
has_input = false
|
116
|
+
|
117
|
+
INPUT_FIELDS.each do |field|
|
118
|
+
if options[field] and !options[field].empty?
|
119
|
+
has_input = true
|
120
|
+
|
121
|
+
break
|
122
|
+
end
|
123
|
+
end
|
124
|
+
|
125
|
+
unless has_input
|
126
|
+
halt(400, 'No input specified in the "input" or "input_url" field')
|
127
|
+
end
|
128
|
+
|
129
|
+
if options['callbacks'] and !options['callbacks'].empty?
|
130
|
+
process_async(options)
|
131
|
+
else
|
132
|
+
process_sync(options)
|
133
|
+
end
|
134
|
+
end
|
135
|
+
|
136
|
+
##
|
137
|
+
# Processes a request synchronously, results are sent as the response upon
|
138
|
+
# completion.
|
139
|
+
#
|
140
|
+
# @param [Hash] options
|
141
|
+
# @return [String]
|
142
|
+
#
|
143
|
+
def process_sync(options)
|
144
|
+
output, ctype = analyze(options)
|
145
|
+
|
146
|
+
content_type(ctype)
|
147
|
+
|
148
|
+
return output
|
149
|
+
end
|
150
|
+
|
151
|
+
##
|
152
|
+
# Processes a request asynchronously, results are submitted to the next
|
153
|
+
# callback URL.
|
154
|
+
#
|
155
|
+
# @param [Hash] options
|
156
|
+
# @return [Hash]
|
157
|
+
#
|
158
|
+
def process_async(options)
|
159
|
+
request_id = options['request_id'] || SecureRandom.hex
|
160
|
+
final_url = options['callbacks'].last
|
161
|
+
|
162
|
+
async { analyze_async(options, request_id) }
|
163
|
+
|
164
|
+
content_type :json
|
165
|
+
|
166
|
+
return JSON.dump(
|
167
|
+
:request_id => request_id,
|
168
|
+
:output_url => "#{final_url}/#{request_id}"
|
169
|
+
)
|
170
|
+
end
|
171
|
+
|
172
|
+
##
|
173
|
+
# Analyzes the input and returns an Array containing the output and
|
174
|
+
# content type.
|
175
|
+
#
|
176
|
+
# @param [Hash] options
|
177
|
+
# @return [Array]
|
178
|
+
#
|
179
|
+
def analyze(options)
|
180
|
+
comp_options = InputSanitizer.new.whitelist_options(
|
181
|
+
options,
|
182
|
+
self.class.accepted_params
|
183
|
+
)
|
184
|
+
|
185
|
+
input = InputExtractor.new.extract(options)
|
186
|
+
processor = self.class.text_processor.new(comp_options)
|
187
|
+
output = processor.run(input)
|
188
|
+
|
189
|
+
if processor.respond_to?(:output_type)
|
190
|
+
type = processor.output_type
|
191
|
+
else
|
192
|
+
type = :xml
|
193
|
+
end
|
194
|
+
|
195
|
+
return output, type
|
196
|
+
end
|
197
|
+
|
198
|
+
##
|
199
|
+
# Analyzes the input asynchronously.
|
200
|
+
#
|
201
|
+
# @param [Hash] options
|
202
|
+
# @param [String] request_id
|
203
|
+
#
|
204
|
+
def analyze_async(options, request_id)
|
205
|
+
output, _ = analyze(options)
|
206
|
+
|
207
|
+
submit_output(output, request_id, options)
|
208
|
+
|
209
|
+
# Submit the error to the error callback, re-raise so Rollbar can also
|
210
|
+
# report it.
|
211
|
+
rescue Exception => error
|
212
|
+
ErrorHandler.new.submit(error, request_id) if options['error_callback']
|
213
|
+
|
214
|
+
raise error
|
215
|
+
end
|
216
|
+
|
217
|
+
##
|
218
|
+
# Submits the output to the next callback URL.
|
219
|
+
#
|
220
|
+
# @param [String] output
|
221
|
+
# @param [String] request_id
|
222
|
+
# @param [Hash] options
|
223
|
+
#
|
224
|
+
def submit_output(output, request_id, options)
|
225
|
+
callbacks = options['callbacks'].dup
|
226
|
+
next_url = callbacks.shift
|
227
|
+
|
228
|
+
# Re-use the old payload so that any extra data (e.g. metadata) is kept
|
229
|
+
# in place.
|
230
|
+
new_payload = options.merge(
|
231
|
+
'callbacks' => callbacks,
|
232
|
+
'request_id' => request_id
|
233
|
+
)
|
234
|
+
|
235
|
+
# Make sure we don't re-send this to the next component.
|
236
|
+
new_payload.delete('input')
|
237
|
+
|
238
|
+
if Configuration.output_bucket
|
239
|
+
uploader = Uploader.new
|
240
|
+
object = uploader.upload(request_id, output, options['metadata'])
|
241
|
+
|
242
|
+
new_payload['input_url'] = object.url_for(:read, :expires => 3600)
|
243
|
+
else
|
244
|
+
new_payload['input'] = output
|
245
|
+
end
|
246
|
+
|
247
|
+
CallbackHandler.new.post(next_url, new_payload)
|
248
|
+
end
|
249
|
+
|
250
|
+
##
|
251
|
+
# Returns a Hash containing the parameters from a JSON payload. The keys
|
252
|
+
# of this Hash are returned as _strings_ to prevent Symbol DOS attacks.
|
253
|
+
#
|
254
|
+
# @return [Hash]
|
255
|
+
#
|
256
|
+
def params_from_json
|
257
|
+
return JSON.load(request.body.read)
|
258
|
+
end
|
259
|
+
|
260
|
+
##
|
261
|
+
# Returns `true` if the input data is in JSON, false otherwise
|
262
|
+
#
|
263
|
+
# @return [TrueClass|FalseClass]
|
264
|
+
#
|
265
|
+
def json_input?
|
266
|
+
return request.content_type == 'application/json'
|
267
|
+
end
|
268
|
+
|
269
|
+
##
|
270
|
+
# Authenticates the current request.
|
271
|
+
#
|
272
|
+
def authenticate!
|
273
|
+
token = Configuration.authentication_token
|
274
|
+
secret = Configuration.authentication_secret
|
275
|
+
creds = {token => params[token], secret => params[secret]}
|
276
|
+
|
277
|
+
response = HTTPClient.get(Configuration.authentication_endpoint, creds)
|
278
|
+
|
279
|
+
unless response.ok?
|
280
|
+
halt(403, "Authentication failed: #{response.body}")
|
281
|
+
end
|
282
|
+
end
|
283
|
+
|
284
|
+
##
|
285
|
+
# Runs the block in a separate thread. When running a test environment the
|
286
|
+
# block is instead yielded normally.
|
287
|
+
#
|
288
|
+
def async
|
289
|
+
if self.class.environment == :test
|
290
|
+
yield
|
291
|
+
else
|
292
|
+
Thread.new { yield }
|
293
|
+
end
|
294
|
+
end
|
295
|
+
end # Server
|
296
|
+
end # Webservice
|
297
|
+
end # Opener
|
@@ -0,0 +1,50 @@
|
|
1
|
+
module Opener
|
2
|
+
module Webservice
|
3
|
+
##
|
4
|
+
# Class for uploading KAF documents to Amazon S3.
|
5
|
+
#
|
6
|
+
class Uploader
|
7
|
+
##
|
8
|
+
# Uploads the given KAF document.
|
9
|
+
#
|
10
|
+
# @param [String] identifier
|
11
|
+
# @param [String] document
|
12
|
+
# @param [Hash] metadata description
|
13
|
+
#
|
14
|
+
# @return [AWS::S3::S3Object]
|
15
|
+
#
|
16
|
+
def upload(identifier, document, metadata = {})
|
17
|
+
object = create(
|
18
|
+
"#{identifier}.xml",
|
19
|
+
document,
|
20
|
+
:metadata => metadata,
|
21
|
+
:content_type => 'application/xml'
|
22
|
+
)
|
23
|
+
|
24
|
+
return object
|
25
|
+
end
|
26
|
+
|
27
|
+
##
|
28
|
+
# @param [Array] args
|
29
|
+
# @return [AWS::S3::S3Object]
|
30
|
+
#
|
31
|
+
def create(*args)
|
32
|
+
return bucket.objects.create(*args)
|
33
|
+
end
|
34
|
+
|
35
|
+
##
|
36
|
+
# @return [AWS::S3.new]
|
37
|
+
#
|
38
|
+
def s3
|
39
|
+
return @s3 ||= AWS::S3.new
|
40
|
+
end
|
41
|
+
|
42
|
+
##
|
43
|
+
# @return [AWS::S3::Bucket]
|
44
|
+
#
|
45
|
+
def bucket
|
46
|
+
return @bucket ||= s3.buckets[Configuration.output_bucket]
|
47
|
+
end
|
48
|
+
end # Uploader
|
49
|
+
end # Daemons
|
50
|
+
end # Opener
|