def writeRanks(ids):
with newman_connector() as read_cnx1, newman_connector() as read_cnx, newman_connector() as write_cnx:
with execute_query(read_cnx1.conn(), stmt) as qry:
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
facts = Fact(write_cnx.conn(), autocommit=False)
print "assigning ranks"
for mail in qry.cursor():
#print mail[0] #, "email_addr", "rank", ids.get(mail,'0'), txid
facts.addFact(mail[0], "email_addr", "rank", ids.get(mail[0],'0'), txid)
print "commit"
write_cnx.commit()
def writeRanks(ids):
with newman_connector() as read_cnx1, newman_connector(
) as read_cnx, newman_connector() as write_cnx:
with execute_query(read_cnx1.conn(), stmt) as qry:
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
facts = Fact(write_cnx.conn(), autocommit=False)
print "assigning ranks"
for mail in qry.cursor():
#print mail[0] #, "email_addr", "rank", ids.get(mail,'0'), txid
facts.addFact(mail[0], "email_addr", "rank",
ids.get(mail[0], '0'), txid)
print "commit"
write_cnx.commit()
for subgraph in clustering.subgraphs():
community_name = jsonGet(['name'], head(subgraph.vs['node']), 'n/a')
for node in subgraph.vs['node']:
node['community'] = community_name
#output format
#NODE\tCOMMUNITY
# for node in nodes:
# print "{}\t{}".format(node['name'], node['community'])
count = counter(1)
with newman_connector() as read_cnx, newman_connector() as write_cnx:
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
facts = Fact(write_cnx.conn(), autocommit=False)
print "assigning communities"
for node in nodes:
email_addr, community_id = node['name'], node['community']
facts.addFact(email_addr, "email_addr", "community", community_id,
txid)
facts.addFact(email_addr, "email_addr", "group_id", next(count),
txid)
print "commit"
write_cnx.commit()
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
print "assign community ids"
stmt = (
args = parser.parse_args()
headers = [
"id", "threadid", "dir", "category", "datetime", "from", "tos", "ccs",
"bccs", "subject", "body", "tosize", "ccsize", "attachsize", "attach",
"bodysize", "location"
]
#skip header row for counting
c = counter(-1)
with newman_connector() as cnx:
tx = Tx(cnx.conn()).next()
print "tx: %s" % tx
fact = Fact(cnx.conn(), autocommit=False)
for line in slurpA(args.input_tsv):
try:
count = c.next()
if count % 1000 == 0:
print "ingested count - %s " % count
row = line.split('\t')
row = (c.strip() for c in row)
num, dir, category, utc_date, importance, fromemail, ip, toemail, ccemail, bccemail, attach, messageid, inreplyto, references, subject, body = row
fromemail = lower(fromemail)
toemail = lower(toemail)
ccemail = lower(ccemail)
bccemail = lower(bccemail)
for subgraph in clustering.subgraphs():
community_name = jsonGet(['name'], head(subgraph.vs['node']), 'n/a')
for node in subgraph.vs['node']:
node['community'] = community_name
#output format
#NODE\tCOMMUNITY
# for node in nodes:
# print "{}\t{}".format(node['name'], node['community'])
count = counter(1)
with newman_connector() as read_cnx, newman_connector() as write_cnx:
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
facts = Fact(write_cnx.conn(), autocommit=False)
print "assigning communities"
for node in nodes:
email_addr, community_id = node['name'], node['community']
facts.addFact(email_addr, "email_addr", "community", community_id, txid)
facts.addFact(email_addr, "email_addr", "group_id", next(count), txid)
print "commit"
write_cnx.commit()
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
print "assign community ids"
stmt = (
" insert into facts (subject, schema_name, predicate, obj, tx) "
" select f.subject, f.schema_name, 'community_id', f2.obj, %s "
" where f.schema_name = 'email'"
" and f2.schema_name = f.schema_name"
" and f.predicate = 'from'"
" and f2.predicate in ('to', 'cc', 'bcc')")
if __name__ == "__main__":
parser = argparse.ArgumentParser(
description='Enrich Emails Communications')
args = parser.parse_args()
with newman_connector() as read_cnx, newman_connector() as write_cnx:
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
print "enrich sent to recipient communications"
facts = Fact(write_cnx.conn(), autocommit=False)
with execute_nonquery(write_cnx.conn(), stmt_sent_to, txid) as qry:
pass
write_cnx.commit()
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
print "enrich total sent to "
with execute_query(write_cnx.conn(), stmt_total_recipients,
txid) as qry:
pass
write_cnx.commit()
txid = Tx(read_cnx.conn()).next()
print "tx: %s" % txid
if __name__ == "__main__":
parser = argparse.ArgumentParser(description='Ingest Walker Email')
parser.add_argument("input_tsv", help="input of tsv file")
args= parser.parse_args()
headers = ["id","threadid", "dir","category","datetime","from","tos","ccs","bccs","subject","body","tosize","ccsize","attachsize","attach","bodysize","location"]
#skip header row for counting
c = counter(-1)
with newman_connector() as cnx:
tx = Tx(cnx.conn()).next()
print "tx: %s" % tx
fact = Fact(cnx.conn(), autocommit=False)
for line in slurpA(args.input_tsv):
try:
count = c.next()
if count % 1000 == 0:
print "ingested count - %s " % count
row = line.split('\t')
row = (c.strip() for c in row)
num,dir,category,utc_date,importance,fromemail,ip,toemail,ccemail,bccemail,attach,messageid,inreplyto,references,subject,body = row
fromemail = lower(fromemail)
toemail = lower(toemail)
ccemail = lower(ccemail)
bccemail = lower(bccemail)