-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathsearch_server.py
427 lines (370 loc) · 15.7 KB
/
search_server.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
from __future__ import unicode_literals
from flask import Flask
from flask import request
from flask_api import status
from flask_cors import CORS, cross_origin
import os
import json
import requests
import StringIO
import codecs
from math import log
from engine import Engine
from elasticsearch_mapping.generate import generate_from_project_config
from elasticsearch_mapping.generate import generate_from_etk_config
from elasticsearch_indexing.index_knowledge_graph import index_knowledge_graph_fields
from urllib import unquote
from urlparse import urlparse
from copy import deepcopy
app = Flask(__name__)
CORS(app, supports_credentials=True)
engine = None
project_engines = {}
current_project = "static"
_location__ = os.path.realpath(
os.path.join(os.getcwd(), os.path.dirname(__file__)))
def load_project_json_file(file_name):
file = json.load(codecs.open(os.path.join(_location__, file_name),
'r', 'utf-8'))
return file
def get_engine(project=None):
global engine
global project_engines
if not project:
if engine:
return engine
else:
import sys
c = load_json_file(sys.argv[11])
engine = Engine(c)
set_engine(engine)
return engine
else:
if project in project_engines:
return project_engines.get(project)
else:
raise ValueError("project {} has not been configured yet\n".format(project))
def get_default_es_endpoint(project=None):
global project_engines
if project and project not in project_engines:
project = None
execute_component = get_engine(project).config["coarse"]["execute"]["components"][0]
if "endpoints" in execute_component:
default_es_endpoint = execute_component["endpoints"]
if "host" in execute_component and "port" in execute_component:
default_es_endpoint = ["{}:{}".format(execute_component["host"], execute_component["port"])]
return default_es_endpoint
@app.route("/")
def hello():
return "DIG Sandpaper\n"
def post_url(url, data):
url = unquote(url)
parsed_url = urlparse(url)
# for some reason requests is ignoring usernames and passwords in urls
if parsed_url.username:
response = requests.post(url, auth=(parsed_url.username, parsed_url.password), data=data)
else:
response = requests.post(url, data=data)
return response
def put_url(url, data):
url = unquote(url)
parsed_url = urlparse(url)
# for some reason requests is ignoring usernames and passwords in urls
if parsed_url.username:
response = requests.put(url, auth=(parsed_url.username, parsed_url.password), data=data)
else:
response = requests.put(url, data=data)
return response
def get_url(url):
url = unquote(url)
parsed_url = urlparse(url)
# for some reason requests is ignoring usernames and passwords in urls
if parsed_url.username:
response = requests.get(url, auth=(parsed_url.username, parsed_url.password))
else:
response = requests.get(url)
return response
def get_project_config(url, project):
if url and project:
response = get_url('{}/projects/{}'.format(url, project))
elif url and not project:
response = get_url(url)
elif not url and project:
response = get_url('{}/projects/{}'.format(default_project_url, project))
else:
return "Please provide either a url and/or a project as url params to retrieve fields to generate an elasticserach mapping\n", status.HTTP_400_BAD_REQUEST
project_config = response.json()
response.raise_for_status()
return project_config
def call_generate_mapping(url, project, project_config=None, shards=5):
if not project_config:
project_config = get_project_config(url, project)
return generate_from_project_config(project_config, shards=shards)
def call_generate_mapping_from_etk_config(etk_config, shards=5):
return generate_from_etk_config(etk_config, shards=shards)
@app.route("/mapping/generate/etk", methods=['POST'])
def generate_mapping_from_etk():
shards = request.args.get('shards', 5)
etk_config = json.loads(request.data)
m = call_generate_mapping_from_etk_config(etk_config, shards=shards)
return json.dumps(m)
@app.route("/mapping/generate", methods=['GET'])
def generate_mapping():
url = request.args.get('url', None)
project = request.args.get('project', None)
shards = request.args.get('shards', 5)
m = call_generate_mapping(url, project, shards=shards)
return json.dumps(m)
@app.route("/mapping", methods=['PUT', 'POST'])
def add_mapping():
shards = request.args.get('shards', 5)
url = request.args.get('url', None)
project = request.args.get('project', None)
etk = request.args.get('etk', False)
if etk:
etk_config = json.loads(request.data)
m = call_generate_mapping_from_etk_config(etk_config, shards=shards)
index = request.args.get('index', None)
if not index:
return "Please provide an index to create for the mapping \n",\
status.HTTP_400_BAD_REQUEST
else:
if request.data:
project_config = json.loads(request.data)
else:
project_config = get_project_config(url, project)
m = call_generate_mapping(url, project, project_config, shards=shards)
index = request.args.get('index', project_config["index"]["full"])
if 'endpoint' in request.args:
endpoint = request.args.get('endpoint')
else:
endpoint = get_default_es_endpoint(project)
if not isinstance(endpoint, basestring):
endpoint = endpoint[0]
response = put_url('{}/{}'.format(endpoint, index),
data=json.dumps(m))
response.raise_for_status()
return "index {} added for project {}\n".format(index, project)
def jl_file_iterator(file):
line = file.readline()
while line:
document = json.loads(line)
yield document
line = file.readline()
def _index_fields(request):
if (request.headers['Content-Type'] == 'application/x-gzip'):
gz_data_as_file = StringIO.StringIO(request.data)
uncompressed = gzip.GzipFile(fileobj=gz_data_as_file, mode='rb')
jls = uncompressed.read()
elif (request.headers['Content-Type'] == 'application/json' or
request.headers['Content-Type'] == 'application/x-jsonlines'):
jls = request.data
else:
return "Only supported content types are application/x-gzip, application/json and application/x-jsonlines", status.HTTP_400_BAD_REQUEST
reader = codecs.getreader('utf-8')
jls_as_file = reader(StringIO.StringIO(jls))
jls = [json.dumps(jl) for jl in [index_knowledge_graph_fields(jl) for jl in jl_file_iterator(jls_as_file)] if jl is not None]
return jls
@app.route("/indexing/fields", methods=['POST'])
def index_fields():
jls = _index_fields(request)
indexed_jls = "\n".join(jls)
if (request.headers['Content-Type'] == 'application/x-gzip'):
indexed_jls_as_file = StringIO.StringIO()
compressed = gzip.GzipFile(
filename=FILENAME, mode='wb', fileobj=indexed_jls_as_file)
compressed.write(indexed_jls)
compressed.close()
return indexed_jls_as_file.getvalue()
else:
return indexed_jls
def chunker(seq, size):
return (seq[pos:pos + size] for pos in xrange(0, len(seq), size))
@app.route("/indexing", methods=['POST'])
def index():
project = request.args.get('project', None)
endpoint = request.args.get('endpoint', get_default_es_endpoint(project))
if not isinstance(endpoint, basestring):
endpoint = endpoint[0]
index = request.args.get('index', None)
t = request.args.get('type', "ads")
jls = _index_fields(request)
log_requests = get_engine(project).config.get("indexing",{}).get("log_requests", None)
if log_requests:
with open(os.path.join(log_requests, "indexing.{}.jl".format(index)), "a") as myfile:
for jl in jls:
myfile.write(jl + '\n')
url = "{}/{}/{}/_bulk".format(endpoint, index, t)
counter = 0
after_filtering = len(jls)
failed = 0
succeeded = 0
# this is inefficent
for chunk in chunker(jls, 100):
bulk_request = ""
bulk_request_format = '{"index":{}}\n'
for c in chunk:
doc_id = json.loads(c).get("doc_id", None)
if doc_id:
doc_request = '{"index":{"_id":"' + doc_id + '"}}\n' + c + '\n'
else:
doc_request = bulk_request_format + c + '\n'
bulk_request = bulk_request + doc_request
counter += len(chunk)
r = post_url(url, bulk_request)
bulk_response = r.json()
for doc_response in bulk_response.get("items", []):
index_response = doc_response.get("index", {})
failed = index_response.get("shards", {}).get("failed", 0)
succeeded = index_response.get("shards", {}).get("succeeded", 0)
log_responses = get_engine(project).config.get("indexing",{}).get("log_responses", None)
if log_responses:
with open(os.path.join(log_responses, "indexing.{}.responses.jl".format(index)), "a") as myfile:
myfile.write(json.dumps(bulk_response))
return "Posted {} documents. {} succeeded. {} 0 failed.\n".format(counter, succeeded, failed)
@app.route("/search", methods=['POST'])
def search():
project = request.args.get("project", None)
query = json.loads(request.data)
(qs, rs) = get_engine(project).execute_coarse(query)
answers = get_engine(project).execute_fine(qs, rs)
return json.dumps(answers)
def coarse_results_to_dict(r):
if isinstance(r, list):
return [rr.to_dict() for rr in r]
else:
return r.to_dict()
@app.route("/search/coarse", methods=['POST'])
def coarse():
project = request.args.get("project", None)
query = json.loads(request.data)
log_requests = get_engine(project).config.get("coarse",{}).get("log_requests", None)
if log_requests:
with open(os.path.join(log_requests, "coarse.{}.jl".format(current_project)), "a") as myfile:
myfile.write(json.dumps(query) + '\n')
(qs, rs) = get_engine(project).execute_coarse(query)
qs_with_rs = [{"query": q, "result": coarse_results_to_dict(r)} for q, r in zip(qs, rs)]
return json.dumps(qs_with_rs)
@app.route("/search/coarse/generate", methods=['POST'])
def coarse_generate():
project = request.args.get("project", None)
query = json.loads(request.data)
qs = get_engine(project).generate_coarse(query)
return json.dumps(qs)
@app.route("/search/fine", methods=['POST'])
def fine():
return "Hello World!\n"
def set_engine(e, project=None):
global engine
engine = e
if project:
global project_engines
if project in project_engines:
old_e = project_engines[project]
project_engines[project] = e
old_e.teardown()
else:
project_engines[project] = e
def multiply_values(w, multiplier):
for k, v in w.iteritems():
if isinstance(v, (int, float, long, complex)):
w[k] = v * multiplier
elif isinstance(v, dict):
multiply_values(v, multiplier)
def update_endpoint(config, endpoint):
if endpoint:
execute_component = config["coarse"]["execute"]["components"][0]
execute_component.pop("host", None)
execute_component.pop("port", None)
if isinstance(endpoint, basestring):
endpoints = list()
endpoints.append(unquote(endpoint))
else:
endpoints = endpoint
execute_component["endpoints"] = endpoints
return config
def apply_config_from_project(url, project, endpoint, index=None,
default_config=None, sample=False,
search_importance_enabled=False,
project_config=None):
if not project_config:
project_config = get_project_config(url, project)
global current_project
current_project = project
if not index:
if not sample:
index = project_config["index"]["full"]
else:
index = project_config["index"]["sample"]
if not default_config:
default_config = load_project_json_file("default_config.json")
c = default_config
update_endpoint(c, endpoint)
generate_components = c["coarse"]["generate"]["components"]
preprocess_components = c["coarse"]["preprocess"]["components"]
predicate_type_mapping = {}
type_field_mapping = {}
type_group_field_mapping = {}
field_weight_mapping = {}
methods = ["extract_from_landmark", "other_method"]
segments = ["title", "content_strict", "other_segment"]
for gc in generate_components:
if gc["name"] == "TypeIndexMapping":
gc["type_index_mappings"]["Ad"] = index
if gc["name"] == "TypeFieldGroupByMapping":
type_group_field_mapping = gc["type_field_mappings"]
if gc["name"] == "TypeFieldMapping":
type_field_mapping = gc["type_field_mappings"]
if gc["name"] == "FieldWeightMapping":
field_weight_mapping = gc["field_weight_mappings"]
for pc in preprocess_components:
if pc["name"] == "PredicateDictConstraintTypeMapper":
predicate_type_mapping = pc["predicate_range_mappings"]
for field_name, spec in project_config["fields"].iteritems():
predicate_type_mapping[field_name] = field_name.lower()
type_group_field_mapping[field_name.lower()] = "indexed.{}.high_confidence_keys".format(field_name)
fields = list()
if spec.get("type", "string") == "string" and\
("email" not in field_name.lower() and
"website" not in field_name.lower() and
"tld" not in field_name.lower() and
"date" not in field_name.lower() and
"image" not in field_name.lower()):
fields.extend(type_field_mapping["owl:Thing"])
if "search_importance" in spec and search_importance_enabled:
search_importance = spec["search_importance"]
if search_importance > 1:
multiplier = log(search_importance, 4)
else:
multiplier = 0.25
weights = deepcopy(field_weight_mapping["indexed"]["*"])
field_weight_mapping["indexed"][field_name] = weights
multiply_values(weights, multiplier)
for method in methods:
for segment in segments:
fields.append("indexed.{}.{}.{}.value".format(field_name, method, segment))
if spec.get("type", "string") == "email" or "email" in field_name.lower():
fields.append("indexed.{}.{}.{}.key".format(field_name, method, segment))
type_field_mapping[field_name.lower()] = fields
set_engine(Engine(c), project)
@app.route("/config", methods=['POST'])
def config():
url = request.args.get('url', None)
project = request.args.get('project', None)
endpoint = request.args.get('endpoint', get_default_es_endpoint())
index = request.args.get('index', None)
sample = request.args.get('sample', False)
search_importance_enabled = request.args.get('searchimportanceenabled',
False)
if request.data and len(request.data) > 0:
project_config = json.loads(request.data)
else:
project_config = None
apply_config_from_project(url, project, endpoint, index,
None, sample,
search_importance_enabled,
project_config)
return "Applied config for project {}\n".format(project)
def load_json_file(file_name):
rules = json.load(codecs.open(file_name, 'r', 'utf-8'))
return rules