#2: Use pdftotext -layout
instead of pdftotext -raw
to distinguish debit from credit
This commit is contained in:
parent
69d6442966
commit
2d4a10da05
@ -6,6 +6,7 @@ import matplotlib.ticker as ticker
|
|||||||
from datetime import datetime
|
from datetime import datetime
|
||||||
from utils import getDatetimeFromFileName, getMonthIndexSinceEpoch, getMonthNameFromMonthIndex, readPdfBankStatement
|
from utils import getDatetimeFromFileName, getMonthIndexSinceEpoch, getMonthNameFromMonthIndex, readPdfBankStatement
|
||||||
import operator
|
import operator
|
||||||
|
from pprint import pprint
|
||||||
|
|
||||||
PATH = f'/home/benjamin/Desktop/bens_folder/bazaar/documents/bnp/bank_statements/'
|
PATH = f'/home/benjamin/Desktop/bens_folder/bazaar/documents/bnp/bank_statements/'
|
||||||
|
|
||||||
@ -34,11 +35,12 @@ for folder in os.listdir():
|
|||||||
filePath = f'{folder}/{file}'
|
filePath = f'{folder}/{file}'
|
||||||
print(filePath)
|
print(filePath)
|
||||||
print(readPdfBankStatement(filePath))
|
print(readPdfBankStatement(filePath))
|
||||||
exit(0)
|
#exit(0)
|
||||||
transactions = readPdfBankStatement(filePath)[3]
|
transactions = readPdfBankStatement(filePath)[3]
|
||||||
|
pprint(transactions)
|
||||||
appendTransactions(transactions, folder)
|
appendTransactions(transactions, folder)
|
||||||
|
|
||||||
#exit(1)
|
exit(1)
|
||||||
os.chdir(f'{MAIN_BANK_ACCOUNT}/')
|
os.chdir(f'{MAIN_BANK_ACCOUNT}/')
|
||||||
|
|
||||||
for folder in sorted(os.listdir()):
|
for folder in sorted(os.listdir()):
|
||||||
@ -53,7 +55,6 @@ for folder in sorted(os.listdir()):
|
|||||||
#break
|
#break
|
||||||
#break
|
#break
|
||||||
|
|
||||||
from pprint import pprint
|
|
||||||
allTransactions.sort(key = operator.itemgetter('date'))
|
allTransactions.sort(key = operator.itemgetter('date'))
|
||||||
print(len(allTransactions))
|
print(len(allTransactions))
|
||||||
pprint(allTransactions)
|
pprint(allTransactions)
|
||||||
|
41
utils.py
41
utils.py
@ -2,17 +2,18 @@ import subprocess
|
|||||||
from datetime import datetime
|
from datetime import datetime
|
||||||
import re
|
import re
|
||||||
|
|
||||||
FIRST_LINE_OF_PAYMENT_REGEX = re.compile('(\\d{2}\\.\\d{2}) (\\d{2}\\.\\d{2}) ([\\d ]+,\\d{2})')
|
# Source: [the Stack Overflow answer 766377](https://stackoverflow.com/a/766377)
|
||||||
|
FIRST_LINE_OF_PAYMENT_REGEX = re.compile('\\ +(\\d{2}\\.\\d{2})\\ +([A-Z\\d /.()*]+?)\\ +(\\d{2}\\.\\d{2})\\ +([\\d ]+,\\d{2})')
|
||||||
END_PAGE_AFTER_THE_FIRST_ONE_REGEX = re.compile('P\\. \\d+/\\d+')
|
END_PAGE_AFTER_THE_FIRST_ONE_REGEX = re.compile('P\\. \\d+/\\d+')
|
||||||
SOLDE_CREDITEUR_AU_REGEX = re.compile('SOLDE CREDITEUR AU (\\d{2}\\.\\d{2}\\.\\d{4}) ([\\d ]+,\\d{2})')
|
SOLDE_CREDITEUR_AU_REGEX = re.compile('\\ +SOLDE CREDITEUR AU (\\d{2}\\.\\d{2}\\.\\d{4})\\ +([\\d ]+,\\d{2})')
|
||||||
TOTAL_DES_OPERATIONS_REGEX = re.compile('TOTAL\\ DES\\ OPERATIONS\\ ([\\d ]+,\\d{2})\\ ([\\d ]+,\\d{2})')
|
TOTAL_DES_OPERATIONS_REGEX = re.compile('\\ +TOTAL\\ DES\\ OPERATIONS\\ +([\\d ]+,\\d{2})\\ +([\\d ]+,\\d{2})')
|
||||||
TOTAL_DES_OPERATIONS_CREDIT_ONLY_REGEX = re.compile('TOTAL\\ DES\\ OPERATIONS\\ ([\\d ]+,\\d{2})')
|
TOTAL_DES_OPERATIONS_CREDIT_ONLY_REGEX = re.compile('\\ +TOTAL\\ DES\\ OPERATIONS\\ +([\\d ]+,\\d{2})')
|
||||||
|
|
||||||
def execute(command):
|
def execute(command):
|
||||||
return subprocess.check_output(command).decode('utf-8')
|
return subprocess.check_output(command).decode('utf-8')
|
||||||
|
|
||||||
def getTextFromPdf(pdfPath):
|
def getTextFromPdf(pdfPath):
|
||||||
return execute(['pdftotext', '-raw', pdfPath, '-'])
|
return execute(['pdftotext', '-layout', pdfPath, '-'])
|
||||||
|
|
||||||
def getDatetimeFromFileName(aDatetimeStr):
|
def getDatetimeFromFileName(aDatetimeStr):
|
||||||
#aDatetime = datetime.strptime(aDatetimeStr, '%Y%m%d.pdf')
|
#aDatetime = datetime.strptime(aDatetimeStr, '%Y%m%d.pdf')
|
||||||
@ -56,6 +57,7 @@ def readPdfBankStatement(filePath):
|
|||||||
if soldeCrediteurAuRegexMatch is not None:
|
if soldeCrediteurAuRegexMatch is not None:
|
||||||
initialDate = datetime.strptime(soldeCrediteurAuRegexMatch.group(1), '%d.%m.%Y')
|
initialDate = datetime.strptime(soldeCrediteurAuRegexMatch.group(1), '%d.%m.%Y')
|
||||||
initialAmount = toFloat(soldeCrediteurAuRegexMatch.group(2))
|
initialAmount = toFloat(soldeCrediteurAuRegexMatch.group(2))
|
||||||
|
print(f'{initialAmount=}')
|
||||||
#currentAmount = initialAmount
|
#currentAmount = initialAmount
|
||||||
started = True
|
started = True
|
||||||
continue
|
continue
|
||||||
@ -66,16 +68,20 @@ def readPdfBankStatement(filePath):
|
|||||||
started = False
|
started = False
|
||||||
continue
|
continue
|
||||||
# We aren't interested in the content after this line
|
# We aren't interested in the content after this line
|
||||||
elif line.startswith('TOTAL DES OPERATIONS'):
|
else:
|
||||||
totalDesOperationsRegexMatch = TOTAL_DES_OPERATIONS_REGEX.match(line)
|
totalDesOperationsRegexMatch = TOTAL_DES_OPERATIONS_REGEX.match(line)
|
||||||
# Note that transfer between accounts will be noted in both debits and credits, as trying to cancel would make benefits show as negative debit which does not make sense.
|
totalDesOperationsCreditOnlyRegexMatch = TOTAL_DES_OPERATIONS_CREDIT_ONLY_REGEX.match(line)
|
||||||
# Cannot just consider January as benefits only as `20240122.pdf` also contains an additional transfer between my accounts.
|
if totalDesOperationsRegexMatch is not None or totalDesOperationsCreditOnlyRegexMatch is not None:
|
||||||
if totalDesOperationsRegexMatch is not None:
|
# Note that transfer between accounts will be noted in both debits and credits, as trying to cancel would make benefits show as negative debit which does not make sense.
|
||||||
totalMonthlyDebit, totalMonthlyCredit = [toFloat(group) for group in totalDesOperationsRegexMatch.groups()]
|
# Cannot just consider January as benefits only as `20240122.pdf` also contains an additional transfer between my accounts.
|
||||||
else:
|
if totalDesOperationsRegexMatch is not None:
|
||||||
totalMonthlyCredit = toFloat(TOTAL_DES_OPERATIONS_CREDIT_ONLY_REGEX.match(line).group(1))
|
totalMonthlyDebit, totalMonthlyCredit = [toFloat(group) for group in totalDesOperationsRegexMatch.groups()]
|
||||||
totalMonthlyDebit = 0
|
else:
|
||||||
break
|
totalMonthlyCredit = toFloat(totalDesOperationsCreditOnlyRegexMatch.group(1))
|
||||||
|
totalMonthlyDebit = 0
|
||||||
|
print(f'{totalMonthlyDebit=}')
|
||||||
|
print(f'{totalMonthlyCredit=}')
|
||||||
|
break
|
||||||
firstLineOfPaymentRegexMatch = FIRST_LINE_OF_PAYMENT_REGEX.match(line)
|
firstLineOfPaymentRegexMatch = FIRST_LINE_OF_PAYMENT_REGEX.match(line)
|
||||||
if firstLineOfPaymentRegexMatch is not None:
|
if firstLineOfPaymentRegexMatch is not None:
|
||||||
print(line)
|
print(line)
|
||||||
@ -88,11 +94,12 @@ def readPdfBankStatement(filePath):
|
|||||||
'comment': '\n'.join(comment)
|
'comment': '\n'.join(comment)
|
||||||
}]
|
}]
|
||||||
date = None
|
date = None
|
||||||
date, valeur, amount = firstLineOfPaymentRegexMatch.groups()
|
date, firstCommentLine, valeur, amount = firstLineOfPaymentRegexMatch.groups()
|
||||||
amount = toFloat(amount)
|
amount = toFloat(amount)
|
||||||
#currentAmount -= amount
|
#currentAmount -= amount
|
||||||
comment = []
|
comment = [firstCommentLine]
|
||||||
else:
|
elif line != '':
|
||||||
|
print(f'comment: {line}')
|
||||||
comment += [line]
|
comment += [line]
|
||||||
if date is not None:
|
if date is not None:
|
||||||
transactions += [{
|
transactions += [{
|
||||||
|
Loading…
Reference in New Issue
Block a user