def on_loaded(self):
self.operations = []
for tr in self.document.getiterator('tr'):
if tr.attrib.get('class', '') == 'hdoc1' or tr.attrib.get('class', '') == 'hdotc1':
tds = tr.findall('td')
if len(tds) != 4:
continue
date = u''
date = tds[0].text
label = u''
label += tds[1].text
label = label.replace(u'\xa0', u'')
for child in tds[1].getchildren():
if child.text: label += child.text
if child.tail: label += child.tail
if tds[1].tail: label += tds[1].tail
label = label.strip()
amount = tds[2].text.replace('.', '').replace(',', '.')
# if we don't have exactly one '.', this is not a floatm try the next
operation = Operation(len(self.operations))
if amount.count('.') != 1:
amount = tds[3].text.replace('.', '').replace(',', '.')
operation.amount = float(amount)
else:
operation.amount = - float(amount)
operation.date = date
operation.label = label
self.operations.append(operation)
def get_history(self):
mvt_table = self.document.xpath("//table[@id='mouvements']", smart_strings=False)[0]
mvt_ligne = mvt_table.xpath("./tbody/tr")
operations = []
for mvt in mvt_ligne:
operation = Operation(len(operations))
operation.date = mvt.xpath("./td")[0].text
tp = mvt.xpath("./td")[1]
operation.label = remove_extra_spaces(remove_html_tags(self.browser.parser.tostring(tp)))
r = re.compile(r"\d+")
tp = mvt.xpath("./td/span")
amount = None
for t in tp:
if r.search(t.text):
amount = t.text
amount = "".join(amount.replace(".", "").replace(",", ".").split())
if amount[0] == "-":
operation.amount = -float(amount[1:])
else:
operation.amount = float(amount)
operations.append(operation)
return operations
def get_history(self):
index = 0
for tr in self.document.getiterator('tr'):
first_td = tr.getchildren()[0]
if first_td.attrib.get('class', '') == 'i g' or first_td.attrib.get('class', '') == 'p g':
operation = Operation(index)
index += 1
operation.date = first_td.text
operation.label = tr.getchildren()[2].text.replace('\n',' ')
if len(tr.getchildren()[3].text) > 2:
s = tr.getchildren()[3].text
elif len(tr.getchildren()[4].text) > 2:
s = tr.getchildren()[4].text
else:
s = "0"
balance = u''
for c in s:
if c.isdigit() or c == "-":
balance += c
if c == ',':
balance += '.'
operation.amount = float(balance)
yield operation
def on_loaded(self):
self.operations = []
for tr in self.document.getiterator('tr'):
if tr.attrib.get('class', '') == 'hdoc1' or tr.attrib.get('class', '') == 'hdotc1':
tds = tr.findall('td')
if len(tds) != 3:
continue
date = tds[0].getchildren()[0].attrib.get('name', '')
label = u''
label += tds[1].text or u''
label = label.replace(u'\xa0', u'')
for child in tds[1].getchildren():
if child.text: label += child.text
if child.tail: label += child.tail
if tds[1].tail: label += tds[1].tail
label = label.strip()
amount = tds[2].text.replace('.', '').replace(',', '.')
operation = Operation(len(self.operations))
operation.date = date
operation.label = label
operation.amount = float(amount)
self.operations.append(operation)
def get_history(self, start_index = 0):
"""
Returns the history of a specific account. Note that this function
expects the current page page to be the one dedicated to this history.
"""
# tested on CA Lorraine, Paris, Toulouse
# avoir parsing the page as an account-dedicated page if it is not the case
if not self.is_account_page():
return
index = start_index
operation = False
body_elmt_list = self.document.xpath('/html/body/*')
# type of separator used in the page
separators = 'hr'
# How many <hr> elements do we have under the <body>?
sep_expected = len(self.document.xpath('/html/body/hr'))
if (not sep_expected):
# no <hr>? Then how many class-less <div> used as separators instead?
sep_expected = len(self.document.xpath('/html/body/div[not(@class) and not(@style)]'))
separators = 'div'
# the interesting divs are after the <hr> elements
interesting_divs = []
right_div_count = 0
left_div_count = 0
sep_found = 0
for body_elmt in body_elmt_list:
if (separators == 'hr' and body_elmt.tag == 'hr'):
sep_found += 1
elif (separators == 'div' and body_elmt.tag == 'div' and body_elmt.get('class', 'nope') == 'nope'):
sep_found += 1
elif (sep_found >= sep_expected and body_elmt.tag == 'div'):
# we just want <div> with dv class and a style attribute
if (body_elmt.get('class', '') != 'dv'):
continue
if (body_elmt.get('style', 'nope') == 'nope'):
continue
interesting_divs.append(body_elmt)
if (self.is_right_aligned_div(body_elmt)):
right_div_count += 1
else:
left_div_count += 1
# So, how are data laid out?
toulouse_way_of_life = (left_div_count == 2 * right_div_count)
# we'll have: one left-aligned div for the date, one right-aligned
# div for the amount, and one left-aligned div for the label. Each time.
if (not toulouse_way_of_life):
for body_elmt in interesting_divs:
if (self.is_right_aligned_div(body_elmt)):
# this is the second line of an operation entry, displaying the amount
data = self.extract_text(body_elmt).replace(',', '.').replace(' ', '')
matches = re.findall('^(-?[0-9]+\.[0-9]{2}).*$', data)
operation.amount = float(matches[0]) if (matches) else 0.0
yield operation
else:
# this is the first line of an operation entry, displaying the date and label
data = self.extract_text(body_elmt)
matches = re.findall('^([012][0-9]|3[01])/(0[1-9]|1[012]).(.+)$', data)
operation = Operation(index)
index += 1
if (matches):
operation.date = u'%s/%s' % (matches[0][0], matches[0][1])
operation.label = u'%s' % matches[0][2]
else:
operation.date = u'01/01'
operation.label = u'Unknown'
else:
for i in range(0, len(interesting_divs)/3):
operation = Operation(index)
index += 1
# amount
data = self.extract_text(interesting_divs[(i*3)+1]).replace(',', '.').replace(' ', '')
matches = re.findall('^(-?[0-9]+\.[0-9]{2}).*$', data)
operation.amount = float(matches[0]) if (matches) else 0.0
# date
data = self.extract_text(interesting_divs[i*3])
matches = re.findall('^([012][0-9]|3[01])/(0[1-9]|1[012])', data)
operation.date = u'%s/%s' % (matches[0][0], matches[0][1]) if (matches) else u'01/01'
#label
data = self.extract_text(interesting_divs[(i*3)+2])
data = re.sub(' +', ' ', data)
operation.label = u'%s' % data
yield operation