diff options
Diffstat (limited to 'python/parse_gs_exchange.py')
| -rw-r--r-- | python/parse_gs_exchange.py | 20 |
1 files changed, 12 insertions, 8 deletions
diff --git a/python/parse_gs_exchange.py b/python/parse_gs_exchange.py index e80d8b1e..0d6de7c0 100644 --- a/python/parse_gs_exchange.py +++ b/python/parse_gs_exchange.py @@ -1,4 +1,3 @@ -from db import dbconn from exchange import get_msgs from pytz import timezone from parse_emails import write_todb @@ -7,9 +6,11 @@ import logging import pandas as pd import re + class ParseError(Exception): pass + def parse_email(email, fwd_index): m = re.search("(IG|HY)(\d{2}) 5y (?:.*)SWAPTION (?:UPDATE|CLOSES|CLOSE) - Ref\D+(.+)$", email.subject) @@ -18,7 +19,8 @@ def parse_email(email, fwd_index): series = int(series) if indextype == 'HY': refprice, refspread = map(float, - re.match("([\S]+)\s+\(([^)]+)\)", ref).groups()) + re.match(r"([\S]+)\s+\(([^)]+)\)", ref). + groups()) else: refspread = float(ref) else: @@ -51,25 +53,25 @@ def parse_email(email, fwd_index): r.append(vals) continue else: - if indextype=='HY': + if indextype == 'HY': cols = ['Strike', 'Sprd', 'Pay', 'DeltaPay', 'Rec', 'Vol', 'VolChg', 'VolBpd', 'Tail'] else: cols = ['Strike', 'Pay', 'DeltaPay', 'Rec', 'Vol', 'VolChg', 'VolBpd', 'Tail'] - df = pd.DataFrame.from_records(r, columns = cols) + df = pd.DataFrame.from_records(r, columns=cols) df[['PayBid', 'PayOffer']] = df.Pay.str.split('/', expand=True) df[['RecBid', 'RecOffer']] = df.Rec.str.split('/', expand=True) df.drop(['Pay', 'Rec'], axis=1, inplace=True) for col in df: - df[col] = pd.to_numeric(df[col], errors = 'coerce') + df[col] = pd.to_numeric(df[col], errors='coerce') df.set_index('Strike', inplace=True) d = {'quotedate': quotedate, 'expiry': date, 'index': indextype, 'series': series, - 'ref': refspread if indextype =="IG" else refprice} + 'ref': refspread if indextype == "IG" else refprice} if indextype == "IG": d['fwdspread'] = float(fwspread) else: @@ -80,7 +82,8 @@ def parse_email(email, fwd_index): flag = False r = [] continue - return (quotedate, indextype, series, pd.concat(masterdf, names=['expiry'])) + return quotedate, indextype, series, pd.concat(masterdf, names=['expiry']) + def clean_df(all_df): all_df = pd.concat(all_df, names=['quotedate', 'index', 'series'], sort=True) @@ -100,7 +103,8 @@ def clean_df(all_df): del all_df['Sprd'] return all_df -if __name__=="__main__": + +if __name__ == "__main__": fwd_index = [] swaption_quotes = {} for email in get_msgs(count=20): |