Created
April 6, 2016 11:18
-
-
Save danieroux/236fd31aff25ebfa5b5963d0f44db786 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/evn python | |
# | |
# Rough but serviceable Standard Bank South Africa PDF statement to CSV extraction. | |
# Exports to Xero's import format | |
# | |
# Needs: | |
# - poppler | |
# - lxml for python to be installed (pip install lxml) | |
from lxml import etree as ET | |
from subprocess import call | |
import re | |
import os | |
import csv | |
class ParseStandardBankPDF: | |
def __init__(self, year, pdf_location): | |
self.pdf_location = pdf_location | |
filename, ext = os.path.splitext(pdf_location) | |
self.filename = filename | |
self.csv_location = filename + ".csv" | |
self.xml_location = filename + ".xml" | |
self.year = year | |
def makeXmlTree(self, pdf_location): | |
cmd = "pdftohtml -xml -nodrm '{}'".format(pdf_location) | |
os.system(cmd) | |
filename, ext = os.path.splitext(pdf_location) | |
return ET.parse(self.xml_location) | |
def parseTransactions(self): | |
tree = self.makeXmlTree(self.pdf_location) | |
transactions = [] | |
first_page_done = False | |
for page in tree.getroot().iter('page'): | |
found = page.xpath('.//text[text()="BALANCE BROUGHT FORWARD"]') | |
if (found): | |
transactions += self.doPageWithEntries(found[0].itersiblings(), first_page_done) | |
first_page_done = True | |
return transactions | |
def doPageWithEntries(self, cursor, first_page_done): | |
cursor.next() | |
transactions_on_page = [] | |
if (not first_page_done): | |
cursor.next() | |
while True: | |
try: | |
type_and_number = cursor.next() | |
description = cursor.next() | |
amount = cursor.next() | |
if (amount.text == '##'): | |
amount = cursor.next() | |
date_ish = cursor.next() | |
balance = cursor.next() | |
transactions_on_page.append([type_and_number.text, description.text, amount.text, date_ish.text, balance.text]) | |
except StopIteration: | |
break | |
return transactions_on_page | |
def makeAmountFromMatch(self, matchobject): | |
rand = matchobject.group(1).replace(".", "") | |
cents = matchobject.group(2) | |
minus = matchobject.group(3) | |
return "{}{}.{}".format(minus, rand, cents) | |
def makeAmount(self, string_amount): | |
return re.sub(r'([\d.]*),(\d*)(-?)', self.makeAmountFromMatch, string_amount) | |
def makeDate(self, string_date): | |
month, day = string_date.split(" ") | |
return "{}-{}-{}".format(self.year, month, day) | |
def makeLine(self, list): | |
type_and_number, description, amount, date_ish, balance = list | |
return { 'Reference': type_and_number, | |
'Description': description, | |
'Amount': self.makeAmount(amount), | |
'Date': self.makeDate(date_ish) } | |
def writeTransactions(self, listOfList): | |
with open(self.csv_location, 'w') as csvfile: | |
fieldnames = ['Date', 'Amount', 'Description', 'Reference'] | |
writer = csv.DictWriter(csvfile, fieldnames=fieldnames) | |
writer.writeheader() | |
for list in listOfList: | |
writer.writerow(self.makeLine(list)) | |
def go(self): | |
parsed = self.parseTransactions() | |
self.writeTransactions(parsed) | |
ParseStandardBankPDF("2015", 'Statement number 24.pdf').go() | |
ParseStandardBankPDF("2015", 'Statement number 25.pdf').go() | |
ParseStandardBankPDF("2015", 'Statement number 26.pdf').go() | |
ParseStandardBankPDF("2015", 'Statement number 27.pdf').go() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment