def get_history(self):
i = 0
ignore = False
for tr in self.doc.xpath(
'//table[@cellpadding="1"]/tr') + self.doc.xpath(
'//tr[@class="rowClick" or @class="rowHover"]'):
tds = tr.findall('td')
if len(tds) < 4:
continue
# if there are more than 4 columns, ignore the first one.
i = min(len(tds) - 4, 1)
if tr.attrib.get('class', '') == 'DataGridHeader':
if tds[2].text == u'Titulaire':
ignore = True
else:
ignore = False
continue
if ignore:
continue
# Remove useless details
detail = tr.cssselect('div.detail')
if len(detail) > 0:
detail[0].drop_tree()
t = Transaction()
date = u''.join([txt.strip() for txt in tds[i + 0].itertext()])
raw = u' '.join([txt.strip() for txt in tds[i + 1].itertext()])
debit = u''.join([txt.strip() for txt in tds[-2].itertext()])
credit = u''.join([txt.strip() for txt in tds[-1].itertext()])
t.parse(date, re.sub(r'[ ]+', ' ', raw))
card_debit_date = self.doc.xpath(
u'//span[@id="MM_HISTORIQUE_CB_m_TableTitle3_lblTitle"] | //label[contains(text(), "débiter le")]'
)
if card_debit_date:
t.rdate = Date(dayfirst=True).filter(date)
m = re.search('(\d{2}\/\d{2}\/\d{4})', card_debit_date[0].text)
assert m
t.date = Date(dayfirst=True).filter(m.group(1))
if t.date is NotAvailable:
continue
if 'tot dif' in t.raw.lower():
t.deleted = True
t.set_amount(credit, debit)
yield t
i += 1
