def invalid_extraction(entity_type, entity_value, domain):
user = session_helper.get_user()
user_name = user.get_user_name()
org = session_helper.get_org()
success = db.mark_invalid_extracted_entity(user_name, entity_type,
entity_value, domain, org) == 0
return json.dumps(dict(success=success))
def add_irrelevant_trail_entity(domain, trail, entity):
org = helper.get_org()
if not db.does_irrelevant_entity_exist(org, domain, trail, entity):
success = db.add_irrelevant_trail_entity(org, domain, trail, entity.encode("utf-8")) == 0
if success:
kafka_producer.send_trail_term_message(org, domain, trail, entity, False)
return json.dumps(dict(success=success))
return json.dumps(dict(success=True))
def add_irrelevant_trail_entity(domain, trail, entity):
org = helper.get_org()
if not db.does_irrelevant_entity_exist(org, domain, trail, entity):
success = db.add_irrelevant_trail_entity(org, domain, trail, entity.encode("utf-8")) == 0
if success:
kafka_producer.send_trail_term_message(org, domain, trail, entity, False)
return json.dumps(dict(success=success))
return json.dumps(dict(success=True))
def get_trail_entity_links(domain, trail):
org = helper.get_org()
links = json.dumps(
dict(
visited=db.get_visited_trail_entity_links(org, domain, trail),
notVisited=db.get_not_visited_trail_entity_links(org, domain, trail),
)
)
return links
def getTimeWindow(users, trail=u'*'):
org = helper.get_org()
if trail == u'':
trail = u'*'
print 'getTimeWindow(', users, ',', trail, ')'
if len(users) > 0:
users = users.split(",")
else:
users = []
return json.dumps(datawake_mysql.getTimeWindow(org, users, trail))
def getTimeWindow(users, trail=u'*'):
org = helper.get_org()
if trail == u'':
trail = u'*'
print 'getTimeWindow(', users, ',', trail, ')'
if len(users) > 0:
users = users.split(",")
else:
users = []
return json.dumps(datawake_mysql.getTimeWindow(org, users, trail))
def get(domain,trail,stars,newdomain):
org = helper.get_org().upper()
if not db.domain_exists(newdomain):
db.add_new_domain(newdomain,'auto generated domain from trail: '+trail)
features = set([])
url_set = set([])
stars = int(stars)
# get all stared urls for the trail
for (url,rank) in db.getRankedUrls(org,trail,domain):
url_set.add(url)
if stars < 1:
urls = db.getBrowsePathUrls(org,trail)
for url in urls:
url_set.add(url)
# get the list of invalid entities for the domain
markedEntities = set([])
for (type,value) in db.get_marked_entities_for_domain(org, domain):
markedEntities.add(value)
# for each url get all extracted entities
entity_data_connector = factory.get_entity_data_connector()
all_entities = entity_data_connector.get_extracted_entities_from_urls(url_set)
for url,featureDict in all_entities.iteritems():
for type,values in featureDict.iteritems():
type = type.replace(',',' ')
filtered_values = []
for value in values:
if value not in markedEntities:
value = value.replace(',',' ')
features.add(type+"\0"+value)
# for each url get any manually extracted entities
for url in url_set:
for featureObj in db.get_feedback_entities(org, domain, url):
type = featureObj['type'].replace(',',' ')
value = featureObj['value'].replace(',',' ')
features.add(type+"\0"+value)
entity_data_connector.add_new_domain_items( map(lambda x: newdomain+'\0'+x,features))
def get(domain, trail):
org = helper.get_org()
trail_report = {}
# get all stared urls for the trail
for (url, rank) in db.getRankedUrls(org, trail, domain):
trail_report[url] = {
'rank': rank,
}
# get the list of invalid entities for the domain
markedEntities = set([])
for (type, value) in db.get_marked_entities_for_domain(org, domain):
markedEntities.add(value)
# for each url get all extracted entities
entity_data_connector = factory.get_entity_data_connector()
all_entities = entity_data_connector.get_extracted_entities_from_urls(
trail_report.keys())
for url, featureDict in all_entities.iteritems():
for type, values in featureDict.iteritems():
filtered_values = []
for value in values:
if value not in markedEntities:
filtered_values.append(value)
if len(filtered_values) > 0:
try:
if 'auto_features' not in trail_report[url]:
trail_report[url]['auto_features'] = {}
trail_report[url]['auto_features'][type] = filtered_values
except:
tangelo.log("report generation error. skipping url.")
continue
# for each url get any manually extracted entities
for url in trail_report.keys():
for featureObj in db.get_feedback_entities(org, domain, url):
if 'manual_features' not in trail_report[url]:
trail_report[url]['manual_features'] = {}
if featureObj['type'] not in trail_report[url]['manual_features']:
trail_report[url]['manual_features'][featureObj['type']] = []
trail_report[url]['manual_features'][featureObj['type']].append(
featureObj['value'])
# for each url get any highlighted text
for url in trail_report.keys():
selections = db.getSelections(domain, trail, url, org)
if len(selections) > 0:
trail_report[url]['selections'] = selections
result = {'trail': trail, 'urls': trail_report}
return json.dumps(result, sort_keys=True, indent=4, separators=(',', ':'))
def get_chart(users=u"", trail=u"*", domain=u""):
org = helper.get_org()
# tangelo.log('dataservice-get org=' + org + ' users=' + users + ' trail= ' + trail + ' domain=' + domain)
if trail == u"":
trail = u"*"
if len(users) > 0:
users = users.split(",")
else:
users = []
result = datawake_mysql.getHourlyBrowsePathCounts(org, users, trail, domain=domain)
return json.dumps(dict(data=result))
def get_chart(users=u'', trail=u'*', domain=u''):
org = helper.get_org()
#tangelo.log('dataservice-get org=' + org + ' users=' + users + ' trail= ' + trail + ' domain=' + domain)
if trail == u'':
trail = u'*'
if len(users) > 0:
users = users.split(",")
else:
users = []
result = datawake_mysql.getHourlyBrowsePathCounts(org, users, trail, domain=domain)
return json.dumps(dict(data=result))
def get(domain, trail, stars, newdomain):
org = helper.get_org().upper()
if not db.domain_exists(newdomain):
db.add_new_domain(newdomain,
'auto generated domain from trail: ' + trail)
features = set([])
url_set = set([])
stars = int(stars)
# get all stared urls for the trail
for (url, rank) in db.getRankedUrls(org, trail, domain):
url_set.add(url)
if stars < 1:
urls = db.getBrowsePathUrls(org, trail)
for url in urls:
url_set.add(url)
# get the list of invalid entities for the domain
markedEntities = set([])
for (type, value) in db.get_marked_entities_for_domain(org, domain):
markedEntities.add(value)
# for each url get all extracted entities
entity_data_connector = factory.get_entity_data_connector()
all_entities = entity_data_connector.get_extracted_entities_from_urls(
url_set)
for url, featureDict in all_entities.iteritems():
for type, values in featureDict.iteritems():
type = type.replace(',', ' ')
filtered_values = []
for value in values:
if value not in markedEntities:
value = value.replace(',', ' ')
features.add(type + "\0" + value)
# for each url get any manually extracted entities
for url in url_set:
for featureObj in db.get_feedback_entities(org, domain, url):
type = featureObj['type'].replace(',', ' ')
value = featureObj['value'].replace(',', ' ')
features.add(type + "\0" + value)
entity_data_connector.add_new_domain_items(
map(lambda x: newdomain + '\0' + x, features))
def getGraph(name, startdate=u'', enddate=u'', users=u'', trail=u'*', domain=u''):
org = helper.get_org()
if trail == u'':
trail = u'*'
userlist = map(lambda x: x.replace('\"', '').strip(), users.split(','))
userlist = filter(lambda x: len(x) > 0, userlist)
#tangelo.log('getGraph( ' + str(name) + ',' + str(startdate) + ',' + str(enddate) + ',' + str(userlist) + ',' + str(trail) + ',' + str(domain) + ')')
if name == 'browse path':
graph = graph_helper.getBrowsePathEdges(org, startdate, enddate, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent urls':
graph = graph_helper.getBrowsePathAndAdjacentWebsiteEdgesWithLimit(org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent urls min degree 2':
graph = graph_helper.getBrowsePathAndAdjacentWebsiteEdgesWithLimit(org, startdate, enddate, 2, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent phone #\'s':
graph = graph_helper.getBrowsePathAndAdjacentPhoneEdgesWithLimit(org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent email #\'s':
graph = graph_helper.getBrowsePathAndAdjacentEmailEdgesWithLimit(org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with text selections':
graph = graph_helper.getBrowsePathWithTextSelections(org, startdate, enddate, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path- with look ahead':
graph = graph_helper.getBrowsePathWithLookAhead(org, startdate, enddate, userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent info':
graph = graph_helper.getBrowsePathAndAdjacentInfoEdges(org, startdate, enddate,1,userlist, trail, domain)
return json.dumps(graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'OculusForensicRequest':
rows = graph_helper.getOculusForensicGraph(org,startdate,enddate,userlist,trail,domain)
return json.dumps(rows)
return json.dumps(dict(nodes=[], links=[]))
def get_trails(domain):
org = helper.get_org()
return get_trails_for_domain_and_org(org, domain)
def get_trail_based_entities(domain, trail):
entities = db.get_trail_based_entities(helper.get_org(), domain, trail)
irrelevantEntities = db.get_irrelevant_trail_based_entities(helper.get_org(), domain, trail)
return json.dumps(dict(entities=entities, irrelevantEntities=irrelevantEntities))
def get_trails(domain):
org = helper.get_org()
return get_trails_for_domain_and_org(org, domain)
def fetch_entities(domain, url):
org = session_helper.get_org()
entities = db.get_feedback_entities(org, domain, url)
return json.dumps(dict(entities=entities))
def marked_entities(domain):
user = session_helper.get_user()
user_name = user.get_user_name()
org = session_helper.get_org()
marked_entities_list = db.get_marked_entities(org, domain, user_name)
return json.dumps(dict(marked_entities=marked_entities_list))
def get_url_entities(domain, trail, url):
results = json.dumps(dict(entities=db.get_entities_on_url(helper.get_org(), domain, trail, url)))
tangelo.log_info(results)
return results
def good_extraction(raw_text, entity_type, entity_value, url, domain):
org = session_helper.get_org()
success = db.add_extractor_feedback(org, domain, raw_text, entity_type, entity_value, url) == 0
return json.dumps(dict(success=success))
def deleteUser(users, startdate, enddate):
org = helper.get_org()
tangelo.log('deleteUser(' + users + ',' + startdate + ',' + enddate + ')')
datawake_mysql.deleteUserData(org, users, startdate, enddate)
return json.dumps(dict(success=True))
def marked_entities(domain):
user = session_helper.get_user()
user_name = user.get_user_name()
org = session_helper.get_org()
marked_entities_list = db.get_marked_entities(org, domain, user_name)
return json.dumps(dict(marked_entities=marked_entities_list))
def getTrails():
org = helper.get_org()
results = datawake_mysql.getTrailsWithUserCounts(org)
results.insert(0, {})
return json.dumps(results)
def listUsers():
org = helper.get_org()
return json.dumps(datawake_mysql.getActiveUsers(org))
def delete_link_from_trail(domain, trail, url):
org = helper.get_org()
success = db.delete_link_from_trail(org, domain, trail, url) == 0
return json.dumps(dict(success=success))
def get_trail_entity_links(domain, trail):
org = helper.get_org()
links = json.dumps(dict(visited=db.get_visited_trail_entity_links(org, domain, trail), notVisited=db.get_not_visited_trail_entity_links(org, domain, trail)))
return links
def get_trail_based_entities(domain, trail):
entities = db.get_trail_based_entities(helper.get_org(), domain, trail)
irrelevantEntities = db.get_irrelevant_trail_based_entities(helper.get_org(), domain, trail)
return json.dumps(dict(entities=entities, irrelevantEntities=irrelevantEntities))
def get_url_entities(domain, trail, url):
results = json.dumps(dict(entities=db.get_entities_on_url(helper.get_org(), domain, trail, url)))
tangelo.log_info(results)
return results
def invalid_extraction(entity_type, entity_value, domain):
user = session_helper.get_user()
user_name = user.get_user_name()
org = session_helper.get_org()
success = db.mark_invalid_extracted_entity(user_name, entity_type, entity_value, domain, org) == 0
return json.dumps(dict(success=success))
def get_selections(domain, trail, url):
org = helper.get_org()
return json.dumps(
dict(selections=db.getSelections(domain, trail, url, org)))
def fetch_entities(domain, url):
org = session_helper.get_org()
entities = db.get_feedback_entities(org, domain, url)
return json.dumps(dict(entities=entities))
def good_extraction(raw_text, entity_type, entity_value, url, domain):
org = session_helper.get_org()
success = db.add_extractor_feedback(org, domain, raw_text, entity_type,
entity_value, url) == 0
return json.dumps(dict(success=success))
def listUsers():
org = helper.get_org()
return json.dumps(datawake_mysql.getActiveUsers(org))
def get_selections(domain, trail, url):
org = helper.get_org()
return json.dumps(dict(selections=db.getSelections(domain, trail, url, org)))
def getGraph(name,
startdate=u'',
enddate=u'',
users=u'',
trail=u'*',
domain=u''):
org = helper.get_org()
if trail == u'':
trail = u'*'
userlist = map(lambda x: x.replace('\"', '').strip(), users.split(','))
userlist = filter(lambda x: len(x) > 0, userlist)
#tangelo.log('getGraph( ' + str(name) + ',' + str(startdate) + ',' + str(enddate) + ',' + str(userlist) + ',' + str(trail) + ',' + str(domain) + ')')
if name == 'browse path':
graph = graph_helper.getBrowsePathEdges(org, startdate, enddate,
userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent urls':
graph = graph_helper.getBrowsePathAndAdjacentWebsiteEdgesWithLimit(
org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent urls min degree 2':
graph = graph_helper.getBrowsePathAndAdjacentWebsiteEdgesWithLimit(
org, startdate, enddate, 2, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent phone #\'s':
graph = graph_helper.getBrowsePathAndAdjacentPhoneEdgesWithLimit(
org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent email #\'s':
graph = graph_helper.getBrowsePathAndAdjacentEmailEdgesWithLimit(
org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with text selections':
graph = graph_helper.getBrowsePathWithTextSelections(
org, startdate, enddate, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path- with look ahead':
graph = graph_helper.getBrowsePathWithLookAhead(
org, startdate, enddate, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'browse path - with adjacent info':
graph = graph_helper.getBrowsePathAndAdjacentInfoEdges(
org, startdate, enddate, 1, userlist, trail, domain)
return json.dumps(
graph_helper.processEdges(graph['edges'], graph['nodes']))
if name == 'OculusForensicRequest':
rows = graph_helper.getOculusForensicGraph(org, startdate, enddate,
userlist, trail, domain)
return json.dumps(rows)
return json.dumps(dict(nodes=[], links=[]))
def getTrails():
org = helper.get_org()
results = datawake_mysql.getTrailsWithUserCounts(org)
results.insert(0, {})
return json.dumps(results)
def delete_link_from_trail(domain, trail, url):
org = helper.get_org()
success = db.delete_link_from_trail(org, domain, trail, url) == 0
return json.dumps(dict(success=success))
