gd_bam 0.1.0 → 0.1.1
Sign up to get free protection for your applications and to get access to all the features.
- data/bin/bam.orig +369 -0
- data/lib/bam/version.rb +1 -1
- data/lib/runtime.rb.orig +748 -0
- metadata +272 -284
data/lib/runtime.rb.orig
ADDED
@@ -0,0 +1,748 @@
|
|
1
|
+
require 'nodes/clover_gen'
|
2
|
+
require 'nodes/dependency'
|
3
|
+
require 'pathname'
|
4
|
+
require 'json'
|
5
|
+
require 'builder'
|
6
|
+
require 'erubis'
|
7
|
+
require 'salesforce'
|
8
|
+
require 'active_support/time'
|
9
|
+
require 'es'
|
10
|
+
require 'gooddata'
|
11
|
+
require 'logger'
|
12
|
+
require 'highline'
|
13
|
+
|
14
|
+
module GoodData
|
15
|
+
|
16
|
+
module CloverGenerator
|
17
|
+
|
18
|
+
GENERATOR_ROOT = Pathname(__FILE__).expand_path.dirname
|
19
|
+
BAM_ROOT = GENERATOR_ROOT.parent
|
20
|
+
|
21
|
+
BAM_DEFINED_GRAPHS_ROOT = GENERATOR_ROOT + "graphs"
|
22
|
+
REPO_ROOT = GENERATOR_ROOT + "repo"
|
23
|
+
TEMPLATE_ROOT = BAM_ROOT + "templates"
|
24
|
+
|
25
|
+
DEFINITION_ROOT = Pathname.new(Dir.pwd)
|
26
|
+
USER_DEFINED_GRAPHS_ROOT = DEFINITION_ROOT + "local_graphs"
|
27
|
+
|
28
|
+
CLOVER_PROJECT_ROOT = DEFINITION_ROOT + "clover-project"
|
29
|
+
CLOVER_DOWNLOADERS_ROOT = DEFINITION_ROOT + "downloader-project"
|
30
|
+
CONTRACT_PROJECT_ROOT = DEFINITION_ROOT + "contract-project"
|
31
|
+
|
32
|
+
PROJECT_GRAPHS_ROOT = "graphs"
|
33
|
+
PROJECT_METADATA_ROOT = "metadata"
|
34
|
+
PROJECT_DATA_ROOT = "data"
|
35
|
+
|
36
|
+
PARAM_FILE = DEFINITION_ROOT + 'params.json'
|
37
|
+
FLOWS_ROOT = DEFINITION_ROOT + "flows"
|
38
|
+
USER_DEFINED_PARAMS = File.exist?(PARAM_FILE) ? JSON.parse(File.read(PARAM_FILE), :symbolize_names => true) : {}
|
39
|
+
|
40
|
+
PARAMS = {:additional_params => {}}.merge(USER_DEFINED_PARAMS).merge(USER_DEFINED_PARAMS[:additional_params] || {})
|
41
|
+
|
42
|
+
# TODO: REMOVE WHEN WEBDAV cann use SST
|
43
|
+
PARAMS[:additional_params][:GD_LOGIN] = PARAMS[:gd_login]
|
44
|
+
PARAMS[:additional_params][:GD_PASSWORD] = PARAMS[:gd_pass]
|
45
|
+
PARAMS[:additional_params][:GDC_WEBDAV_HOST] = PARAMS[:web_dav_host] || "secure-di.gooddata.com"
|
46
|
+
|
47
|
+
<<<<<<< Updated upstream
|
48
|
+
=======
|
49
|
+
|
50
|
+
def self.render_template(template, params, options={})
|
51
|
+
to_file = options[:to_file]
|
52
|
+
root = options[:root] || TEMPLATE_ROOT
|
53
|
+
t = Pathname(template)
|
54
|
+
output = Erubis::Eruby.new(File.read(root + t)).result(params)
|
55
|
+
if to_file.nil?
|
56
|
+
output
|
57
|
+
else
|
58
|
+
File.open(to_file, 'w') {|f| f.write(output)}
|
59
|
+
end
|
60
|
+
end
|
61
|
+
|
62
|
+
def self.set_up_debug(project, flow, graph)
|
63
|
+
fail "Project \"#{project}\" was not found" unless File.exist?(project)
|
64
|
+
fail "Project \"#{project}\" is not a directory" unless File.directory?(project)
|
65
|
+
# :TODO make the proram really check there is a flow and a graph etc before writing to the file
|
66
|
+
project = Pathname(project)
|
67
|
+
File.open(project + 'params.prm', 'w') do |f|
|
68
|
+
f << "FLOW=#{flow}\n"
|
69
|
+
f << "NAME=#{graph}\n"
|
70
|
+
end
|
71
|
+
end
|
72
|
+
|
73
|
+
def self.setup_clover_project(base, options={})
|
74
|
+
|
75
|
+
name = options[:name] || PARAMS[:project_name]
|
76
|
+
|
77
|
+
[PROJECT_GRAPHS_ROOT, PROJECT_METADATA_ROOT, PROJECT_DATA_ROOT].each do |dir|
|
78
|
+
FileUtils::mkdir_p base + dir
|
79
|
+
end
|
80
|
+
File.open(base + 'params.prm', 'w') do |f|
|
81
|
+
f << "TRASH=0"
|
82
|
+
end
|
83
|
+
render_template("project.erb", PARAMS.merge(:project_name => name), :to_file => base + '.project')
|
84
|
+
render_template("workspace.prm.erb", PARAMS, :to_file => base + 'workspace.prm')
|
85
|
+
end
|
86
|
+
|
87
|
+
def self.setup_bash_structure(name)
|
88
|
+
fail "Directory \"#{name}\" already exists. Please remove it if you want to move forward." if File.exist?(name)
|
89
|
+
FileUtils::mkdir_p name
|
90
|
+
FileUtils::cd(name) do
|
91
|
+
render_template("params.json.erb", PARAMS, :to_file => 'params.json')
|
92
|
+
['flows', 'sinks', 'taps'].each do |dir|
|
93
|
+
FileUtils::mkdir_p dir
|
94
|
+
end
|
95
|
+
|
96
|
+
setup_flow('user')
|
97
|
+
# render_template("flow.rb.erb", PARAMS, :to_file => 'flows/flow_example.rb')
|
98
|
+
render_template("source.json.erb", PARAMS, :to_file => 'taps/source_example.json')
|
99
|
+
render_template("dataset.json.erb", PARAMS, :to_file => 'sinks/dataset_example.json')
|
100
|
+
end
|
101
|
+
|
102
|
+
end
|
103
|
+
|
104
|
+
def self.setup_flow(name)
|
105
|
+
render_template("flow.rb.erb", {:flow_name => name}, :to_file => "flows/#{name}.rb")
|
106
|
+
end
|
107
|
+
|
108
|
+
def self.setup_tap(name)
|
109
|
+
render_template("tap.json.erb", {:tap_name => name}, :to_file => "taps/#{name}.json")
|
110
|
+
end
|
111
|
+
|
112
|
+
def self.clobber_clover_project()
|
113
|
+
FileUtils::rm_rf(CLOVER_PROJECT_ROOT)
|
114
|
+
end
|
115
|
+
|
116
|
+
def self.clobber_downloader_project()
|
117
|
+
FileUtils::rm_rf(CLOVER_DOWNLOADERS_ROOT)
|
118
|
+
end
|
119
|
+
|
120
|
+
def self.build_project()
|
121
|
+
p = GoodData::CloverGenerator::DSL.project do |p|
|
122
|
+
project_name("Goodsales for ABCD")
|
123
|
+
|
124
|
+
use_dims ["GoodSales/opportunity", "GoodSales/user"]
|
125
|
+
use_usecase("GoodSales/quota_attainment")
|
126
|
+
end
|
127
|
+
|
128
|
+
repos = GoodData::CloverGenerator::Repo.load(REPO_ROOT).map {|config| N.new(config)}
|
129
|
+
p.run(repos)
|
130
|
+
p
|
131
|
+
end
|
132
|
+
|
133
|
+
def self.validate_datasets
|
134
|
+
project = build_project
|
135
|
+
datasets = project.get_datasets
|
136
|
+
GoodData.connect(PARAMS[:gd_login], PARAMS[:gd_pass])
|
137
|
+
GoodData.project = PARAMS[:project_pid]
|
138
|
+
gd_datasets = GoodData.get("/gdc/md/#{PARAMS[:project_pid]}/query/datasets")['query']['entries'].map {|m| GoodData.get(m['link'])}
|
139
|
+
report1 = GoodData::CloverGenerator::validate_gd_datasets_metadata(datasets, gd_datasets)
|
140
|
+
report2 = GoodData::CloverGenerator::validate_gd_datasets(datasets, gd_datasets)
|
141
|
+
report1.merge(report2)
|
142
|
+
end
|
143
|
+
|
144
|
+
def self.procs_list(options={})
|
145
|
+
project = build_project
|
146
|
+
datasets = project.get_datasets
|
147
|
+
GoodData.connect(PARAMS[:gd_login], PARAMS[:gd_pass])
|
148
|
+
procs = GoodData.get("/gdc/projects/#{PARAMS[:project_pid]}/dataload/processes")
|
149
|
+
procs["processes"]["items"].map {|p| [p["process"]["name"], p["process"]["links"]["self"]]}
|
150
|
+
end
|
151
|
+
|
152
|
+
def self.validate_taps
|
153
|
+
project = build_project
|
154
|
+
sources = project.get_sources
|
155
|
+
client = get_sf_client(PARAMS)
|
156
|
+
sf_sources = sources.find_all {|tap| tap[:source] == "salesforce"}
|
157
|
+
GoodData::CloverGenerator::validate_sf_metadata(client, sf_sources)
|
158
|
+
end
|
159
|
+
|
160
|
+
def self.sf_jack_in
|
161
|
+
client = get_sf_client(PARAMS)
|
162
|
+
client.pry
|
163
|
+
end
|
164
|
+
|
165
|
+
def self.generate_docs
|
166
|
+
project = build_project
|
167
|
+
sources = project.get_sources
|
168
|
+
datasets = project.get_datasets
|
169
|
+
|
170
|
+
taps = ""
|
171
|
+
sources.each do |source|
|
172
|
+
taps += "####{source[:object]}"
|
173
|
+
taps += "\n"
|
174
|
+
source[:fields].each do |f|
|
175
|
+
if f[:acts_as]
|
176
|
+
taps += " #{f[:name]} -> #{f[:acts_as].join(", ")}"
|
177
|
+
else
|
178
|
+
taps += " #{f[:name]}"
|
179
|
+
end
|
180
|
+
taps += "\n"
|
181
|
+
end
|
182
|
+
|
183
|
+
taps += "\n"
|
184
|
+
end
|
185
|
+
|
186
|
+
sinks = ""
|
187
|
+
datasets.each do |sink|
|
188
|
+
name = sink[:gd_name] || sink[:id]
|
189
|
+
sinks += "####{name}\n"
|
190
|
+
sink[:fields].each do |field|
|
191
|
+
name = field[:name] || "#{field[:schema]}:#{field[:ref]}"
|
192
|
+
type = field[:type]
|
193
|
+
sinks += " #{type.upcase} #{field[:meta]} => #{name}\n"
|
194
|
+
end
|
195
|
+
sinks += "\n"
|
196
|
+
|
197
|
+
end
|
198
|
+
|
199
|
+
|
200
|
+
render_template("README.md.erb", PARAMS.merge(:taps => taps, :sinks => sinks), :to_file => 'README.md', :root => DEFINITION_ROOT)
|
201
|
+
end
|
202
|
+
|
203
|
+
def self.model_sync(options)
|
204
|
+
dry_run = options[:dry]
|
205
|
+
project = build_project
|
206
|
+
datasets = project.get_datasets
|
207
|
+
model_update_dir = Pathname('model_update')
|
208
|
+
cl_home = ENV['CL_HOME'] || PARAMS['CL_HOME'] || fail("Home of cl tool cannot be found. Either set up CL_HOME in your env with 'export CL_HOME=path/to/cl or set it up in your params.json. Point to the directory of CL not to the bin dir.'")
|
209
|
+
cl_home = Pathname(cl_home) + 'bin/gdi.sh'
|
210
|
+
|
211
|
+
FileUtils::mkdir_p(model_update_dir)
|
212
|
+
File.open(model_update_dir + 'dummy', 'w')
|
213
|
+
FileUtils::cd(model_update_dir) do
|
214
|
+
datasets.each do |ds|
|
215
|
+
dataset_path = Pathname("cl_file_#{ds[:id]}")
|
216
|
+
File.open(dataset_path, "w") do |temp|
|
217
|
+
builder = Builder::XmlMarkup.new(:target=>temp, :indent=>2)
|
218
|
+
builder.schema do |builder|
|
219
|
+
builder.name(ds[:gd_name])
|
220
|
+
builder.title(ds[:gd_name])
|
221
|
+
builder.columns do |b|
|
222
|
+
ds[:fields].each do |f|
|
223
|
+
builder.column do |builder|
|
224
|
+
builder.title(f[:name])
|
225
|
+
builder.ldmType(f[:type].upcase)
|
226
|
+
builder.reference(f[:for]) if f.has_key?(:for)
|
227
|
+
builder.reference(f[:ref]) if f.has_key?(:ref)
|
228
|
+
builder.schemaReference(f[:schema]) if f.has_key?(:schema)
|
229
|
+
if f[:type] == "date"
|
230
|
+
builder.schemaReference("#{f[:dd]}")
|
231
|
+
builder.name("#{f[:name]}")
|
232
|
+
else
|
233
|
+
builder.name(f[:name] || f[:ref])
|
234
|
+
end
|
235
|
+
end
|
236
|
+
end
|
237
|
+
end
|
238
|
+
end
|
239
|
+
end
|
240
|
+
template_name = dry_run ? "update_dataset_dry.script.erb" : "update_dataset.script.erb"
|
241
|
+
render_template(template_name, PARAMS.merge({"config_file" => dataset_path.expand_path}), :to_file => 'update_dataset.script')
|
242
|
+
puts "Generate #{ds[:id]}"
|
243
|
+
|
244
|
+
system("#{cl_home} update_dataset.script --username #{PARAMS[:gd_login]} --password #{PARAMS[:gd_pass]}")
|
245
|
+
File.delete(dataset_path)
|
246
|
+
end
|
247
|
+
end
|
248
|
+
FileUtils::rm_rf(model_update_dir)
|
249
|
+
end
|
250
|
+
|
251
|
+
def self.generate_downloaders(options={})
|
252
|
+
setup_clover_project(CLOVER_DOWNLOADERS_ROOT, :name => "downloaders-#{PARAMS[:project_name]}")
|
253
|
+
project = build_project
|
254
|
+
sources = project.get_sources
|
255
|
+
sf_sources = sources.find_all {|tap| tap[:source] == "salesforce" && tap[:incremental] == true}
|
256
|
+
create_incremental_downloader_run_graph(CLOVER_DOWNLOADERS_ROOT + PROJECT_GRAPHS_ROOT + "main.grf", sf_sources)
|
257
|
+
s3_backup = PARAMS[:S3_SECRET_ACCESS_KEY] && PARAMS[:S3_ACCESS_KEY_ID] && PARAMS[:S3_BUCKETNAME]
|
258
|
+
|
259
|
+
GoodData::CloverGenerator::create_incremental_downloading_graph(CLOVER_DOWNLOADERS_ROOT + PROJECT_GRAPHS_ROOT + "incremental.grf", sf_sources, {
|
260
|
+
:password => PARAMS[:sf_password],
|
261
|
+
:token => PARAMS[:sf_token],
|
262
|
+
:login => PARAMS[:sf_login],
|
263
|
+
:sf_server => PARAMS[:sf_server],
|
264
|
+
:s3_backup => s3_backup
|
265
|
+
})
|
266
|
+
end
|
267
|
+
|
268
|
+
|
269
|
+
def self.run(dir, options={})
|
270
|
+
email = options[:email]
|
271
|
+
verbose = options[:v]
|
272
|
+
|
273
|
+
dir = Pathname(dir)
|
274
|
+
|
275
|
+
GoodData::CloverGenerator.deploy(dir, options.merge(:name => "Temporary deploy[#{dir}][#{PARAMS[:project_name]}]")) do |deploy_response|
|
276
|
+
puts HighLine::color("Executing", HighLine::BOLD) if verbose
|
277
|
+
if email.nil?
|
278
|
+
result = GoodData::CloverGenerator.execute_process(deploy_response["process"]["links"]["executions"], dir)
|
279
|
+
else
|
280
|
+
GoodData::CloverGenerator.create_email_channel(options) do |channel_response|
|
281
|
+
GoodData::CloverGenerator.subscribe_on_finish(:success, channel_response, deploy_response, options)
|
282
|
+
result = GoodData::CloverGenerator.execute_process(deploy_response["process"]["links"]["executions"], dir)
|
283
|
+
end
|
284
|
+
end
|
285
|
+
end
|
286
|
+
end
|
287
|
+
|
288
|
+
def self.execute_process(link, dir)
|
289
|
+
result = GoodData.post(link, {
|
290
|
+
:execution => {
|
291
|
+
:graph => "./#{dir}/graphs/main.grf",
|
292
|
+
:params => {}
|
293
|
+
}
|
294
|
+
})
|
295
|
+
begin
|
296
|
+
GoodData.poll(result, "executionTask")
|
297
|
+
rescue RestClient::RequestFailed => e
|
298
|
+
|
299
|
+
ensure
|
300
|
+
result = GoodData.get(result["executionTask"]["links"]["detail"])
|
301
|
+
if result["executionDetail"]["status"] == "ERROR"
|
302
|
+
fail "Runing process failed. You can look at a log here #{result["executionDetail"]["logFileName"]}"
|
303
|
+
end
|
304
|
+
end
|
305
|
+
result
|
306
|
+
end
|
307
|
+
|
308
|
+
def self.connect_to_gd(options={})
|
309
|
+
GoodData.connect(PARAMS[:gd_login], PARAMS[:gd_pass])
|
310
|
+
GoodData.project = PARAMS[:project_pid] if !PARAMS[:project_pid].nil? && !PARAMS[:project_pid].empty?
|
311
|
+
end
|
312
|
+
|
313
|
+
def self.subscribe_on_finish(event_type, channel, deploy, options={})
|
314
|
+
channel_uri = channel["channelConfiguration"]["meta"]["uri"]
|
315
|
+
process_id = deploy["process"]["links"]["self"].split('/').last
|
316
|
+
event_id = case event_type
|
317
|
+
when :success
|
318
|
+
"dataload.process.finish.ok"
|
319
|
+
when :failure
|
320
|
+
"dataload.process.finish.error"
|
321
|
+
else
|
322
|
+
fail "You specified unknown event \"#{event_type}\""
|
323
|
+
end
|
324
|
+
|
325
|
+
templates = {
|
326
|
+
:success => {
|
327
|
+
:message => "Just wanted to let you know that ETL for \"#{deploy["process"]["name"]}\" Succeeded",
|
328
|
+
:subject => "GoodData ETL SUCCESS: process with name \"#{deploy["process"]["name"]}\""
|
329
|
+
},
|
330
|
+
:failure => {
|
331
|
+
:message => "Just wanted to let you know that ETL for \"#{deploy["process"]["name"]}\" Failed",
|
332
|
+
:subject => "GoodData ETL FAILURE: process with name \"#{deploy["process"]["name"]}\""
|
333
|
+
}
|
334
|
+
}
|
335
|
+
|
336
|
+
templates_to_be_used = templates[event_type]
|
337
|
+
|
338
|
+
data = {
|
339
|
+
:subscription => {
|
340
|
+
:triggers => [
|
341
|
+
{
|
342
|
+
:projectEventTrigger => {
|
343
|
+
:types => [event_id]
|
344
|
+
}
|
345
|
+
}
|
346
|
+
],
|
347
|
+
:condition => {
|
348
|
+
:condition => {
|
349
|
+
:expression => "params.PROCESS_ID==\"#{process_id}\""
|
350
|
+
}
|
351
|
+
},
|
352
|
+
:subject => {
|
353
|
+
:template => {
|
354
|
+
:expression => templates_to_be_used[:subject]
|
355
|
+
}
|
356
|
+
},
|
357
|
+
:message => {
|
358
|
+
:template => {
|
359
|
+
:expression => templates_to_be_used[:message]
|
360
|
+
}
|
361
|
+
},
|
362
|
+
:channels => [channel_uri],
|
363
|
+
:meta => {
|
364
|
+
:title => "Notification for process #{process_id}"
|
365
|
+
}
|
366
|
+
}
|
367
|
+
}
|
368
|
+
profile_id = GoodData.connection.user["profile"].split("/").last
|
369
|
+
GoodData.post("/gdc/projects/#{PARAMS[:project_pid]}/users/#{profile_id}/subscriptions", data)
|
370
|
+
end
|
371
|
+
|
372
|
+
def self.clone_project(options={})
|
373
|
+
pid = case options[:blueprint]
|
374
|
+
when "goodsales"
|
375
|
+
"nt935rwzls50zfqwy6dh62tabu8h0ocy"
|
376
|
+
when nil
|
377
|
+
fail "Empty project not supported now"
|
378
|
+
end
|
379
|
+
|
380
|
+
project_name = PARAMS[:project_name]
|
381
|
+
fail "project name has to be filled in" if project_name.blank?
|
382
|
+
with_users = options[:with_users]
|
383
|
+
|
384
|
+
export = {
|
385
|
+
:exportProject => {
|
386
|
+
:exportUsers => with_users ? 1 : 0,
|
387
|
+
:exportData => 1
|
388
|
+
}
|
389
|
+
}
|
390
|
+
|
391
|
+
result = GoodData.post("/gdc/md/#{pid}/maintenance/export", export)
|
392
|
+
token = result["exportArtifact"]["token"]
|
393
|
+
status_url = result["exportArtifact"]["status"]["uri"]
|
394
|
+
|
395
|
+
state = GoodData.get(status_url)["taskState"]["status"]
|
396
|
+
while state == "RUNNING"
|
397
|
+
sleep 5
|
398
|
+
result = GoodData.get(status_url)
|
399
|
+
state = result["taskState"]["status"]
|
400
|
+
end
|
401
|
+
|
402
|
+
old_project = GoodData::Project[pid]
|
403
|
+
|
404
|
+
pr = {
|
405
|
+
:project => {
|
406
|
+
:content => {
|
407
|
+
:guidedNavigation => 1,
|
408
|
+
:driver => "Pg",
|
409
|
+
:authorizationToken => options[:token]
|
410
|
+
},
|
411
|
+
:meta => {
|
412
|
+
:title => project_name,
|
413
|
+
:summary => "Testing Project"
|
414
|
+
}
|
415
|
+
}
|
416
|
+
}
|
417
|
+
result = GoodData.post("/gdc/projects/", pr)
|
418
|
+
uri = result["uri"]
|
419
|
+
while(GoodData.get(uri)["project"]["content"]["state"] == "LOADING")
|
420
|
+
sleep(5)
|
421
|
+
end
|
422
|
+
|
423
|
+
new_project = GoodData::Project[uri]
|
424
|
+
|
425
|
+
import = {
|
426
|
+
:importProject => {
|
427
|
+
:token => token
|
428
|
+
}
|
429
|
+
}
|
430
|
+
|
431
|
+
result = GoodData.post("/gdc/md/#{new_project.obj_id}/maintenance/import", import)
|
432
|
+
status_url = result["uri"]
|
433
|
+
state = GoodData.get(status_url)["taskState"]["status"]
|
434
|
+
while state == "RUNNING"
|
435
|
+
sleep 5
|
436
|
+
result = GoodData.get(status_url)
|
437
|
+
state = result["taskState"]["status"]
|
438
|
+
end
|
439
|
+
GoodData.post "/gdc/projects/#{new_project.obj_id}/eventStore/stores", {:store => {:storeId => "es_0"}}
|
440
|
+
new_project.obj_id
|
441
|
+
end
|
442
|
+
|
443
|
+
def self.create_email_channel(options={}, &block)
|
444
|
+
email = options[:email]
|
445
|
+
|
446
|
+
data = {
|
447
|
+
:channelConfiguration => {
|
448
|
+
:configuration => {
|
449
|
+
:emailConfiguration => {
|
450
|
+
:to => email
|
451
|
+
}
|
452
|
+
},
|
453
|
+
:meta => {
|
454
|
+
:title => "temporary email channel"
|
455
|
+
}
|
456
|
+
}
|
457
|
+
}
|
458
|
+
profile_id = GoodData.connection.user["profile"].split("/").last
|
459
|
+
res = GoodData.post("/gdc/account/profile/#{profile_id}/channelConfigurations", data)
|
460
|
+
self_link = res["channelConfiguration"]["meta"]["uri"]
|
461
|
+
if block
|
462
|
+
begin
|
463
|
+
block.call(res)
|
464
|
+
ensure
|
465
|
+
GoodData.delete(self_link)
|
466
|
+
end
|
467
|
+
else
|
468
|
+
res
|
469
|
+
end
|
470
|
+
end
|
471
|
+
|
472
|
+
def self.deploy_graph(dir, options={})
|
473
|
+
deploy_name = options[:name] || "#{PARAMS[:project_name]}"
|
474
|
+
verbose = options[:verbose] || false
|
475
|
+
puts HighLine::color("Deploying #{dir}", HighLine::BOLD) if verbose
|
476
|
+
res = nil
|
477
|
+
|
478
|
+
Tempfile.open("deploy-graph-archive") do |temp|
|
479
|
+
Zip::ZipOutputStream.open(temp.path) do |zio|
|
480
|
+
Dir.glob("./#{dir}/**/*") do |item|
|
481
|
+
puts "including #{item}" if verbose
|
482
|
+
unless File.directory?(item)
|
483
|
+
zio.put_next_entry(item)
|
484
|
+
zio.print IO.read(item)
|
485
|
+
end
|
486
|
+
end
|
487
|
+
end
|
488
|
+
|
489
|
+
GoodData.connection.upload(temp.path)
|
490
|
+
process_id = options[:process]
|
491
|
+
|
492
|
+
binding.pry
|
493
|
+
|
494
|
+
data = {
|
495
|
+
:process => {
|
496
|
+
:name => deploy_name,
|
497
|
+
:path => "/uploads/#{File.basename(temp.path)}"
|
498
|
+
}
|
499
|
+
}
|
500
|
+
res = if process_id.nil?
|
501
|
+
GoodData.post("/gdc/projects/#{PARAMS[:project_pid]}/dataload/processes", data)
|
502
|
+
else
|
503
|
+
GoodData.put("/gdc/projects/#{PARAMS[:project_pid]}/dataload/processes/#{process_id}", data)
|
504
|
+
end
|
505
|
+
end
|
506
|
+
puts HighLine::color("Deploy DONE #{dir}", HighLine::BOLD) if verbose
|
507
|
+
res
|
508
|
+
end
|
509
|
+
|
510
|
+
def self.deploy(dir, options={}, &block)
|
511
|
+
verbose = options[:verbose] || false
|
512
|
+
if block
|
513
|
+
begin
|
514
|
+
res = deploy_graph(dir, options)
|
515
|
+
block.call(res)
|
516
|
+
ensure
|
517
|
+
self_link = res["process"]["links"]["self"]
|
518
|
+
GoodData.delete(self_link)
|
519
|
+
end
|
520
|
+
else
|
521
|
+
deploy_graph(dir, options)
|
522
|
+
end
|
523
|
+
end
|
524
|
+
|
525
|
+
def self.generate_graph_template(name, target)
|
526
|
+
template_name = "#{name}_template.grf.erb"
|
527
|
+
render_template(template_name, PARAMS, :to_file => USER_DEFINED_GRAPHS_ROOT + target)
|
528
|
+
end
|
529
|
+
|
530
|
+
|
531
|
+
def self.generate(options)
|
532
|
+
|
533
|
+
only_flow = options[:only]
|
534
|
+
setup_clover_project(CLOVER_PROJECT_ROOT, :name => "etl-#{PARAMS[:project_name]}")
|
535
|
+
p = build_project
|
536
|
+
sources = p.get_sources
|
537
|
+
datasets = p.get_datasets
|
538
|
+
s3_backup = PARAMS[:S3_SECRET_ACCESS_KEY] && PARAMS[:S3_ACCESS_KEY_ID] && PARAMS[:S3_BUCKETNAME]
|
539
|
+
|
540
|
+
flows = []
|
541
|
+
FileUtils::cd FLOWS_ROOT do
|
542
|
+
flows_sources = Dir.glob("*.rb")
|
543
|
+
flows = flows_sources.map do |f|
|
544
|
+
instance_eval(File.read(f))
|
545
|
+
end
|
546
|
+
end
|
547
|
+
|
548
|
+
flows = flows.find_all {|flow| flow && flow.name == only_flow} unless only_flow.nil? || only_flow.empty?
|
549
|
+
fail "The flow you specified was not found" if flows.empty? && !only_flow.nil?
|
550
|
+
fail "There are no flows to generate from" if flows.empty?
|
551
|
+
super_flow = []
|
552
|
+
FileUtils::cd CLOVER_PROJECT_ROOT do
|
553
|
+
|
554
|
+
flows.each do |f|
|
555
|
+
|
556
|
+
current_metadata = {}
|
557
|
+
steps_to_be_wrapped = []
|
558
|
+
flow_sources = []
|
559
|
+
if f.nil?
|
560
|
+
puts "Flow skipped"
|
561
|
+
next
|
562
|
+
end
|
563
|
+
name = f.name
|
564
|
+
step_no = 0
|
565
|
+
|
566
|
+
begin
|
567
|
+
f.steps.each_with_index do |s, i|
|
568
|
+
|
569
|
+
if s[:type] == :tap
|
570
|
+
source_name = s[:source_name] || f.name
|
571
|
+
source = sources.find do |source|
|
572
|
+
source[:id] == source_name
|
573
|
+
end
|
574
|
+
|
575
|
+
fail "Tap \"#{source_name}\" was not found" if source.nil?
|
576
|
+
|
577
|
+
flow_sources << source
|
578
|
+
|
579
|
+
dataset_name = source[:dataset] || source[:id]
|
580
|
+
|
581
|
+
current_metadata[source_name] = GoodData::CloverGenerator::create_metadata(source)
|
582
|
+
graph_name = "graphs/#{dataset_name}_#{source[:source]}_#{source[:type]}.grf"
|
583
|
+
steps_to_be_wrapped << {
|
584
|
+
:name => "#{source_name}_download",
|
585
|
+
:file => graph_name,
|
586
|
+
:flow => source_name
|
587
|
+
}
|
588
|
+
|
589
|
+
if source[:incremental] == true
|
590
|
+
current_metadata[source_name] = GoodData::CloverGenerator::DSL::Metadata.new(current_metadata[source_name]).change do |m|
|
591
|
+
m.remove("timestamp")
|
592
|
+
end.to_hash
|
593
|
+
|
594
|
+
GoodData::CloverGenerator::create_es_downloading_graph(graph_name, [source], {
|
595
|
+
:metadata => current_metadata[source_name],
|
596
|
+
:s3_backup => s3_backup
|
597
|
+
})
|
598
|
+
else
|
599
|
+
GoodData::CloverGenerator::create_sf_downloading_graph(graph_name, [source], {
|
600
|
+
:password => PARAMS[:sf_password],
|
601
|
+
:token => PARAMS[:sf_token],
|
602
|
+
:login => PARAMS[:sf_login],
|
603
|
+
:sf_server => PARAMS[:sf_server],
|
604
|
+
:metadata => current_metadata[source_name],
|
605
|
+
:s3_backup => s3_backup
|
606
|
+
})
|
607
|
+
end
|
608
|
+
|
609
|
+
step_no += 1
|
610
|
+
elsif s[:type] == :upload
|
611
|
+
source_name = s[:id] || f.name
|
612
|
+
dataset = datasets.find {|d| d[:id] == source_name}
|
613
|
+
|
614
|
+
fail "Sink \"#{source_name}\" was not found" if dataset.nil?
|
615
|
+
fail "Sink needs to have id defined" if dataset[:id].nil?
|
616
|
+
|
617
|
+
unless current_metadata.has_key?(source_name)
|
618
|
+
fail("Tap #{source_name} was not found in current metadata")
|
619
|
+
end
|
620
|
+
fail "Sink \"#{f.name}\" was not found" if dataset.nil?
|
621
|
+
metadata = current_metadata[source_name]
|
622
|
+
|
623
|
+
graph_name = "graphs/#{dataset[:id]}_#{dataset[:type]}.grf"
|
624
|
+
steps_to_be_wrapped << {
|
625
|
+
:name => "#{name}_upload",
|
626
|
+
:file => graph_name,
|
627
|
+
:flow => source_name
|
628
|
+
}
|
629
|
+
|
630
|
+
GoodData::CloverGenerator::create_uploading_graph(graph_name, {
|
631
|
+
:datasets => dataset,
|
632
|
+
:metadata => current_metadata[source_name]
|
633
|
+
})
|
634
|
+
step_no += 1
|
635
|
+
elsif s[:type] == :user_provided
|
636
|
+
# check that what is getting in a step is check
|
637
|
+
s[:metadata_block] && s[:metadata_block].each do |val|
|
638
|
+
name = val[:name]
|
639
|
+
next if(name.nil?)
|
640
|
+
sources_names = flow_sources.map {|flow_source| flow_source[:id]}
|
641
|
+
included_in_flow = sources_names.include?(name)
|
642
|
+
unless included_in_flow
|
643
|
+
# binding.pry
|
644
|
+
# fail "Metadata \"#{name}\" is not in the defined by any source"
|
645
|
+
end
|
646
|
+
end
|
647
|
+
graph_name = s[:graph]
|
648
|
+
graph_filename = "#{s[:graph]}.grf"
|
649
|
+
|
650
|
+
if File.exist?(USER_DEFINED_GRAPHS_ROOT + graph_filename)
|
651
|
+
FileUtils::cp(USER_DEFINED_GRAPHS_ROOT + graph_filename, CLOVER_PROJECT_ROOT + PROJECT_GRAPHS_ROOT)
|
652
|
+
elsif File.exist?(BAM_DEFINED_GRAPHS_ROOT + graph_filename)
|
653
|
+
FileUtils::cp(BAM_DEFINED_GRAPHS_ROOT + graph_filename, CLOVER_PROJECT_ROOT + PROJECT_GRAPHS_ROOT)
|
654
|
+
else
|
655
|
+
fail("The graph \"#{graph_filename}\" was not found in any location (local, global)")
|
656
|
+
end
|
657
|
+
graph_filename = "graphs/#{graph_filename}"
|
658
|
+
|
659
|
+
step_no += 1
|
660
|
+
if s[:metadata_block].nil?
|
661
|
+
s[:metadata_block] = [{:name=> flow_sources.first[:id] }]
|
662
|
+
end
|
663
|
+
|
664
|
+
s[:metadata_block] && s[:metadata_block].each_with_index do |metadata, i|
|
665
|
+
j = i+1
|
666
|
+
bl = metadata[:block]
|
667
|
+
name = metadata[:name] || f.name
|
668
|
+
as = metadata[:out_as]
|
669
|
+
FileUtils::mkdir_p("metadata/#{f.name}/#{graph_name}")
|
670
|
+
|
671
|
+
input_meta = current_metadata[name].clone
|
672
|
+
|
673
|
+
input_meta[:name] = "in_#{j}"
|
674
|
+
GoodData::CloverGenerator::save_metadata("metadata/#{f.name}/#{graph_name}/#{j}_in.xml", input_meta)
|
675
|
+
m = GoodData::CloverGenerator::DSL::Metadata.new(current_metadata[name])
|
676
|
+
new_m = (bl && bl.call(m)) || current_metadata[name]
|
677
|
+
|
678
|
+
current_metadata[name] = new_m
|
679
|
+
unless as.nil?
|
680
|
+
as_metadata = new_m.clone
|
681
|
+
as_metadata[:name] = as
|
682
|
+
current_metadata[as] = as_metadata
|
683
|
+
end
|
684
|
+
|
685
|
+
output_meta = current_metadata[name].clone
|
686
|
+
output_meta[:name] = "out_#{j}"
|
687
|
+
GoodData::CloverGenerator::save_metadata("metadata/#{f.name}/#{graph_name}/#{j}_out.xml", output_meta)
|
688
|
+
|
689
|
+
GoodData::CloverGenerator::create_moving_graph("graphs/#{f.name}_#{graph_name}_move_in_#{j}.grf", {
|
690
|
+
:source => "${DATA}/#{name}.csv",
|
691
|
+
:target => "${DATA}/#{j}_in.csv",
|
692
|
+
:operation => "MOVE"
|
693
|
+
})
|
694
|
+
|
695
|
+
steps_to_be_wrapped << {
|
696
|
+
:name => "graphs/#{f.name}_#{graph_name}_move_in_#{j}.grf",
|
697
|
+
:file => "graphs/#{f.name}_#{graph_name}_move_in_#{j}.grf",
|
698
|
+
:flow => "kopirujeme"
|
699
|
+
}
|
700
|
+
end
|
701
|
+
|
702
|
+
steps_to_be_wrapped << {
|
703
|
+
:name => graph_name,
|
704
|
+
:file => graph_filename,
|
705
|
+
:flow => f.name
|
706
|
+
}
|
707
|
+
|
708
|
+
output_id = if s[:metadata_block] && s[:metadata_block].any? {|metadata| metadata[:out_as]}
|
709
|
+
s[:metadata_block].find {|metadata| metadata[:out_as]}[:out_as]
|
710
|
+
end
|
711
|
+
output_id = s[:metadata_block].first[:name] if output_id.nil? && s[:metadata_block] && s[:metadata_block].length == 1
|
712
|
+
|
713
|
+
GoodData::CloverGenerator::create_moving_graph("graphs/#{f.name}_#{graph_name}_move_out.grf", {
|
714
|
+
:source => "${DATA}/out.csv",
|
715
|
+
:target => "${DATA}/#{output_id}.csv",
|
716
|
+
:operation => "MOVE"
|
717
|
+
})
|
718
|
+
|
719
|
+
steps_to_be_wrapped << {
|
720
|
+
:name => "graphs/#{f.name}_#{graph_name}_move_out.grf",
|
721
|
+
:file => "graphs/#{f.name}_#{graph_name}_move_out.grf",
|
722
|
+
:flow => "kopirujeme"
|
723
|
+
}
|
724
|
+
|
725
|
+
end
|
726
|
+
end
|
727
|
+
rescue GoodData::CloverGenerator::DSL::RemoveMetadataFieldError => e
|
728
|
+
puts "Removing field \"#{e.field}\" from metadata \"#{e.metadata.name}\" in Flow \"#{f.name}\" there was a roblem with step X."
|
729
|
+
end
|
730
|
+
|
731
|
+
GoodData::CloverGenerator::create_run_graph("graphs/#{f.name}_main.grf", {
|
732
|
+
:subgraphs => steps_to_be_wrapped
|
733
|
+
})
|
734
|
+
super_flow << {
|
735
|
+
:name => name,
|
736
|
+
:file => "graphs/#{f.name}_main.grf",
|
737
|
+
:flow => f.name
|
738
|
+
}
|
739
|
+
end
|
740
|
+
|
741
|
+
GoodData::CloverGenerator::create_run_graph("graphs/main.grf", {
|
742
|
+
:subgraphs => super_flow
|
743
|
+
})
|
744
|
+
end
|
745
|
+
end
|
746
|
+
>>>>>>> Stashed changes
|
747
|
+
end
|
748
|
+
end
|