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
|
import psycopg2
import os.path
import re
import csv
import datetime
import pdb
import common
from common import query_db
import sys
fields = ['Asset Name', 'Issuer', 'Contributed Balance', 'Maturity Date', \
'Asset Subtype', 'Asset Type', 'Gross Coupon', 'Spread', 'Frequency', \
'Next Paydate', 'Second Lien', 'LoanX ID', 'CUSIP', 'Market Price', \
'Market Price Source', 'Price Date', 'Fixed or Float', \
'Defaulted Flag', 'Security Sub-Category', \
'Structured Finance Security', 'Life Floor', 'Reinvest Collat', 'Native Currency']
def convertToNone(s):
return None if s=='' else s
def sanitize_float(intex_float):
intex_float = intex_float.replace(",", "")
if "(" in intex_float:
intex_float = - float(intex_float[1:-1])
else:
intex_float = float(intex_float)
return intex_float
def upload_data(dealnames, workdate, conn):
for dealname in dealnames:
#dealname, updatedate = line.rstrip().split()
# updatedate = datetime.datetime.strptime(updatedate, '%m/%d/%Y')
#dealname = dealname.upper() + ",AD.txt
with open( os.path.join(common.root, "data", "Collaterals_" + workdate, dealname.upper() + "_AD.txt"),
"r", encoding='windows-1252') as fh:
dr = csv.DictReader(fh, dialect = 'excel-tab')
missingfields = set(fields) - set(dr.fieldnames)
if "LoanX ID" in missingfields:
print("LoanX ID column is missing. Probably an error in exporting from intex")
pdb.set_trace()
data = []
if 'Reinvest Collat' in missingfields:
print('no reinvestment asset')
pdb.set_trace()
for line in dr:
for f in missingfields:
line[f] = None
if 'Gross Margin' not in line:
line['Gross Margin'] = None
if line['Fixed or Float']:
line['Fixed or Float'] = line['Fixed or Float'].upper()
if line['LoanX ID']:
line['LoanX ID'] = line['LoanX ID'].upper()
if len(line['LoanX ID']) > 8:
print("dubious id found: {0}".format(line['LoanX ID']))
line['LoanX ID'] = line['LoanX ID'][:8]
# make sure the string is utf8 safe
line['Issuer'] = line['Issuer']
if 'Reinvest Collat' not in missingfields and \
line['Reinvest Collat'].upper() == 'Y' or line['Issuer'] == '':
# assume it's a reinvestment asset
line['Reinvest Collat'] = True
line['Issuer'] = line['ID Number']
if 'Spread' in missingfields or not line['Spread']:
line['Spread'] = line['Gross Margin']
for field in ['Spread', 'Gross Coupon', 'Market Price', 'Contributed Balance']:
if line[field]:
line[field] = sanitize_float(line[field])
if line['Market Price'] is not None and line['Market Price'] == 0:
line['Market Price'] = None
#we store the Libor FLoor in the database, so Life Floor is really Libor Floor
if line['Life Floor'] == "No limit":
line['Life Floor'] = 0
elif line['Life Floor']:
try:
line['Life Floor'] = float(line['Life Floor']) - float(line['Spread'])
except ValueError:
line['Life Floor'] = float('Nan')
if line['CUSIP']== 'XAQ3930AAB43':
line['CUSIP']='BL078321'
if line['CUSIP']:
line['CUSIP'] = line['CUSIP'].upper()
r = [convertToNone(line[field]) for field in fields]
if r[fields.index('CUSIP')] and len(r[fields.index('CUSIP')])>9:
pdb.set_trace()
data.append(r)
sqlstr = "select distinct(updatedate) from et_collateral where dealname= %s"
old_update_dates = [date[0] for date in query_db(sqlstr, params=(dealname,), one=False)]
sqlstr = 'select max("Latest Update") from clo_universe where dealname= %s'
updatedate = query_db(sqlstr, params=(dealname,))[0]
reinsert = False
if updatedate in old_update_dates:
sqlstr = "SELECT count(*) FROM et_collateral where dealname = %s and updatedate= %s"
currlen = query_db(sqlstr, params = (dealname, updatedate))[0]
if currlen != len(data):
print("{0} has {1} rows in the database and current collateral file has {2}".format(dealname, currlen, len(data)))
c = conn.cursor()
sqlstr = "DELETE FROM et_collateral where dealname = %s and updatedate = %s"
c.execute(sqlstr, (dealname, updatedate))
conn.commit()
c.close()
reinsert = True
if reinsert or not old_update_dates or updatedate not in old_update_dates:
tag = 0
c = conn.cursor()
for row in data:
# these next three ifs are to take care of reinvestment asset lines
if not row[0]:
row[0] = 'Reinv'
if row[4] and 'Reinvest' in row[4]:
row[4] = row[4].replace("Reinvest ", "")
if row[4] and len(row[4])>10:
row[4] = row[4][:9]
sql_fields = ["dealname", "updatedate", "name", "IssuerName", "CurrentBalance",
"Maturity", "AssetSubtype", "AssetType", "GrossCoupon",
"Spread", "Frequency", "NextPaydate", "SecondLien", "LoanXID",
"Cusip", "IntexPrice", "IntexPriceSource", "IntexPriceDate",
"FixedOrFloat", "DefaultedFlag", "CovLite", "isCDO",
"Liborfloor", "ReinvFlag", "Currency"]
sqlstr = "INSERT INTO ET_COLLATERAL({0}) VALUES({1})".format(",".join(sql_fields),
",".join(["%s"] * len(sql_fields)))
try:
c.execute(sqlstr, (dealname, updatedate) + tuple(row))
except psycopg2.DataError as detail:
print(detail)
pdb.set_trace()
except psycopg2.IntegrityError as detail:
print(detail)
# crazy hack intex unique id is not really unique
conn.rollback()
# make sure the loan name is unique by tagging it
row[0] = row[0] + "_tag_" + str(tag)
tag = tag + 1
c.execute(sqlstr, (dealname, updatedate) + tuple(row))
except TypeError:
pdb.set_trace()
conn.commit()
c.close()
if __name__ == "__main__":
if len(sys.argv) > 1:
workdate = sys.argv[1]
else:
workdate = str(datetime.date.today())
if len(sys.argv) > 2:
dealnames = sys.argv[2:]
else:
dealnames = [re.sub("_AD.txt", "", d).lower() for d in
os.listdir(os.path.join(common.root, "data", "Collaterals_" + workdate))]
upload_data(dealnames, workdate, common.conn)
common.conn.close()
print("done")
|
