Skip to content

Add time deposit and unit trust categories, omit time deposit rows, and add investment balances to the total balance. #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
ckyOL wants to merge 3 commits into
base: master
Choose a base branch
from
Open

Add time deposit and unit trust categories, omit time deposit rows, and add investment balances to the total balance. #3

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
25bbbdf
:sparkles: Add time deposit and unit trust categories, omit time depo…
ckyOL May 3, 2025
cd4f671
:sparkles: Add bond category and skip exchange rate row
ckyOL Jun 2, 2025
534192e
:sparkle: Add FCY time deposit category. Skip (DR=Debit \n) line. Add…
ckyOL Mar 1, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 49 additions & 4 deletions hsbcpdf/helpers/accountstatement.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,12 +23,20 @@ class EnumSumAccountTypes:
HKDCURRENT = 'HKD Current'
FCYSAVINGS = 'FCY Savings'
FCYCURRENT = 'FCY Current'
TIMEDEPOSIT = 'Time Deposit'
FCYTIMEDEPOSIT = 'FCY Time Deposit'
UNITTRUST = 'Unit Trusts'
BOND = 'Bonds/Certificates of Deposit'

class AccountTypes:
HKDSAVINGS = 'HKDSavings'
HKDCURRENT = 'HKDCurrent'
FCYSAVINGS = 'FCYSavings'
FCYCURRENT = 'FCYCurrent'
TIMEDEPOSIT = 'TimeDeposit'
FCYTIMEDEPOSIT = 'FCYTimeDeposit'
UNITTRUST = 'UnitTrust'
BOND = 'Bond'


class TableZone:
Expand Down Expand Up @@ -206,11 +214,21 @@ def clean_table(self):
logger.debug(shape)
logger.debug('table shape: {}'.format(self.table.shape))

last_dt = ""
dt = ""
ccy = ""
desc = ""
new_balance = 0.
for index, row in self.table.iterrows():
# skip time deposit lines
if (
row.isnull().all()
or str(row[0]).strip() in ("Time", "No.", "Deposi")
or str(row[0]).strip().isdigit()
or str(row[0]).strip() == "" and str(row[1]).strip() == "" and str(row[2]).strip().isdigit()
or str(row[0]).strip() == "" and str(row[1]).strip() == "" and str(row[2]).strip() == ""
):
continue
# first line with new currency is previous balance
if row[0] != ccy and row[0] != "":
if ccy != "":
Expand All @@ -231,13 +249,18 @@ def clean_table(self):
new_balance = previous_balance


if row[1] != "": dt = self.extract_date(row[1])
if row[1] != "":
dt = self.extract_date(row[1])
last_dt = dt
else:
dt = last_dt

desc = (desc + " " if desc != "" else "") + row[2]
credit = row[3]
debit = row[4]
amount = None

logger.debug("ccy[{}] date[{}] desc[{}] credit[{}] debit[{}]".format(ccy, dt, desc, credit, debit))
logger.debug("[{}] ccy[{}] date[{}] desc[{}] credit[{}] debit[{}]".format(index, ccy, dt, desc, credit, debit))
if credit is not None and credit != "":
amount = float(credit.replace(",", ""))
elif debit is not None and debit != "":
Expand Down Expand Up @@ -266,7 +289,11 @@ class TableZoneSum(TableZone):
EnumSumAccountTypes.HKDSAVINGS: AccountTypes.HKDSAVINGS,
EnumSumAccountTypes.HKDCURRENT: AccountTypes.HKDCURRENT,
EnumSumAccountTypes.FCYSAVINGS: AccountTypes.FCYSAVINGS,
EnumSumAccountTypes.FCYCURRENT: AccountTypes.FCYCURRENT
EnumSumAccountTypes.FCYCURRENT: AccountTypes.FCYCURRENT,
EnumSumAccountTypes.TIMEDEPOSIT: AccountTypes.TIMEDEPOSIT,
EnumSumAccountTypes.FCYTIMEDEPOSIT: AccountTypes.FCYTIMEDEPOSIT,
EnumSumAccountTypes.UNITTRUST: AccountTypes.UNITTRUST,
EnumSumAccountTypes.BOND: AccountTypes.BOND
}

def __init__(self, page_height, page_width, section, account, st_date):
Expand All @@ -293,9 +320,27 @@ def clean_table(self):
# skip first 2 lines that are header part and account narrative
for index, row in self.table[2:].iterrows():
logger.debug("process row ({}): <{}>".format(index, row))

# skip inverstment title
if row[0] == 'HSBC Premier\n- Investments':
continue

# skip exchange rate
if row[0] == 'Exchange Rate':
continue

# skip (DR=Debit \n) in other line
if row[4] == "(DR=Debit \n)" and row[6] == "(DR=Debit \n)":
continue

if row[0] is not None and row[0] != "":
if row[0] == 'Total':
self.summary['total_balance_hkd'] = self.extract_amount(row[6], row[7])
total_amount = self.extract_amount(row[6], row[7])
if self.summary['total_balance_hkd'] is None:
self.summary['total_balance_hkd'] = total_amount
else:
# add investments total
self.summary['total_balance_hkd'] += total_amount
continue
elif row[0] not in TableZoneSum.map_type.keys():
raise TemplateException("Summary contains an unknow Account type [{}]".format(row[0]))
Expand Down