class item(ItemElement):
klass = Account
def condition(self):
return self.page.accounts_list_condition(self.el)
class Type(Filter):
def filter(self, label):
for pattern, actype in AccountsPage.TYPES.items():
if pattern in label:
return actype
return Account.TYPE_UNKNOWN
obj__title = CleanText('td[@class="ColonneLibelle"][2]')
obj__nature = CleanText('td[@class="ColonneLibelle"][3]')
obj_label = Format('%s %s', Field('_title'), Field('_nature'))
obj_currency = FrenchTransaction.Currency(
'./td[@class="ColonneCode"]')
obj_id = CleanText('td[@class="ColonneLibelle"][1]')
obj__link = Link('td[@class="ColonneLibelle"][1]/a',
default=NotAvailable)
obj__rib_link = Link('.//a[contains(@href, "rib.jsp")]')
obj_type = Type(Field('label'))
obj_balance = CleanDecimal('td[@class="ColonneNumerique"]/nobr',
replace_dots=True)
class item(ItemElement):
klass = Recipe
obj_id = Regexp(Link('.//a[has-class("bu_cuisine_recette_img")]'),
r'.*\/(.*)$')
obj_url = Link('.//a[has-class("bu_cuisine_recette_img")]')
obj_title = CleanText(
'.//p[has-class("bu_cuisine_recette_title")]')
obj_short_description = NotAvailable
obj_author = NotAvailable
obj_ingredients = NotAvailable
class obj_picture(ItemElement):
klass = BaseImage
obj_url = NotAvailable
def obj_thumbnail(self):
style = Attr(
'.//a[has-class("bu_cuisine_recette_img")]/span',
'style')(self)
return Thumbnail(
style.replace("background-image:url(",
"").rstrip(");"))
obj_instructions = NotAvailable
obj_preparation_time = CleanDecimal(
'(.//span[has-class("bu_cuisine_recette_carnet_duree")])[1]')
obj_cooking_time = CleanDecimal(
'(.//span[has-class("bu_cuisine_recette_carnet_duree")])[2]',
default=0)
obj_nb_person = NotAvailable
def obj_coming(self):
if Field('type')(self) == Account.TYPE_CHECKING and Field('balance')(self) != 0:
# When the balance is 0, we get a website unavailable on the history page
# and the following navigation is broken
has_coming = False
coming = 0
details_page = self.page.browser.open(Field('url')(self))
coming_op_link = Link('//a[contains(text(), "Opérations à venir")]', default=NotAvailable)(details_page.page.doc)
if coming_op_link:
coming_op_link = Regexp(Link('//a[contains(text(), "Opérations à venir")]'), r'../(.*)')(details_page.page.doc)
coming_operations = self.page.browser.open(self.page.browser.BASEURL + '/voscomptes/canalXHTML/CCP/' + coming_op_link)
else:
coming_op_link = Link('//a[contains(text(), "Opérations en cours")]')(details_page.page.doc)
coming_operations = self.page.browser.open(coming_op_link)
if CleanText('//span[@id="amount_total"]')(coming_operations.page.doc):
has_coming = True
coming += CleanDecimal('//span[@id="amount_total"]', replace_dots=True)(coming_operations.page.doc)
if CleanText(u'.//dt[contains(., "Débit différé à débiter")]')(self):
has_coming = True
coming += CleanDecimal(u'.//dt[contains(., "Débit différé à débiter")]/following-sibling::dd[1]',
replace_dots=True)(self)
return coming if has_coming else NotAvailable
return NotAvailable
class item(ItemElement):
klass = Message
obj_id = Regexp(Link('./div/div/small/a', default=''),
'/.+/status/(.+)',
default=None)
obj_title = Regexp(
CleanText(
'./div[@class="content"]/div/p[has-class("tweet-text")]',
replace=[('@ ', '@'), ('# ', '#'),
('http:// ', 'http://')]), '(.{50}|.+).+')
obj_content = CleanText(
'./div[@class="content"]/div/p[has-class("tweet-text")]',
replace=[('@ ', '@'), ('# ', '#'), ('http:// ', 'http://')])
obj_sender = Regexp(Link('./div/div/small/a', default=''),
'/(.+)/status/.+',
default=None)
obj_date = DatetimeFromTimestamp(
Attr(
'./div/div[@class="stream-item-header"]/small/a/span | ./div/div[@class="ProfileTweet-authorDetails"]/span/a/span',
'data-time'))
def validate(self, obj):
return obj.id is not None
class item(ItemElement):
klass = Suggestion
obj_id = Regexp(Link('./td[2]/a'), '/([0-9]+)')
obj_name = CleanText('./td[2]/a/text()')
obj_description = Format('Genre: %s - Country: %s',
CleanText('./td[3]/text()'),
CleanText('./td[4]/text()'))
obj_url = Link('./td[2]/a')
def obj_url(self):
url = Link(u'./a', default=NotAvailable)(self)
if not url:
url = Regexp(Attr(u'.//span', 'onclick', default=''), r'\'(https.*)\'', default=NotAvailable)(self)
if url:
if 'CreditRenouvelable' in url:
url = Link(u'.//a[contains(text(), "espace de gestion crédit renouvelable")]')(self.el)
return urljoin(self.page.url, url)
return url
class item(ItemElement):
klass = BaseCalendarEvent
obj_url = Link('./a[1]')
obj_id = Regexp(Link('./a[1]'), r'\?(\d+)')
obj_summary = CleanText('./a[1]')
obj_start_date = Date(CleanText('./span[1]'))
obj_category = CATEGORIES.CONCERT
obj_status = STATUS.CONFIRMED
def next_page(self):
if Link(u'//a[text()="▶"]')(self) is not None:
form = self.page.get_form(xpath='//*[@id="history_form"]')
form['page'] = CleanDecimal(Link(u'//a[text()="▶"]'))(self)
return requests.Request("POST", self.page.url, data=form)
def obj_url(self):
async_page = Async('details').loaded_page(self)
url = Link(
'//a[contains(@href, "download")]|//a[contains(@href, "generated_invoices")]',
default=NotAvailable)(async_page.doc)
if not url:
url = Link(
'//a[contains(text(), "Imprimer un récapitulatif de commande")]'
)(async_page.doc)
return url
class item(ItemElement):
klass = BaseCalendarEvent
obj_url = Link('./div[@class="bbox"]/h1/a')
obj_id = Regexp(Link('./div[@class="bbox"]/h1/a'), r'aspx\?(.+)')
obj_location = CleanText('./div[@class="bbox"]/span/a')
obj_start_date = DateTime(Attr('.//time', 'datetime'))
obj_summary = Regexp(Attr('./div[@class="bbox"]/h1/a', 'title'),
r'details of (.+)')
obj_category = CATEGORIES.CONCERT
obj_status = STATUS.CONFIRMED
class item(ItemElement):
klass = Subscription
obj_id = Regexp(Link('.'), r'\bidContrat=(\d+)', default='')
obj__page = Regexp(Link('.'), r'\bpage=([^&]+)', default='')
obj_label = CleanText('.')
def validate(self, obj):
# unsubscripted contracts may still be there, skip them else
# facture-historique could yield wrong bills
return bool(obj.id) and obj._page != 'nec-tdb-ouvert'
class item(ItemElement):
klass = Housing
def condition(self):
return Regexp(Link('./div[has-class("box-header")]/a[@class="title-item"]'), '/annonces/(.*)', default=None)(self)
obj_id = Regexp(Link('./div[has-class("box-header")]/a[@class="title-item"]'), '/annonces/(.*)')
obj_title = CleanText('./div[has-class("box-header")]/a[@class="title-item"]')
obj_area = CleanDecimal(Regexp(CleanText('./div[has-class("box-header")]/a/span[@class="h1"]'),
'(.*?)(\d*) m\xb2(.*?)', '\\2'), default=NotAvailable)
obj_cost = CleanDecimal(CleanText('./div[has-class("box-header")]/a/span[@class="price"]'),
replace_dots=True, default=Decimal(0))
obj_currency = Regexp(CleanText('./div[has-class("box-header")]/a/span[@class="price"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'), default=u'€')
obj_utilities = UTILITIES.UNKNOWN
def obj_date(self):
_date = Regexp(CleanText('./div[has-class("box-header")]/p[@class="date"]'),
'.* / (.*)')(self)
return parse_french_date(_date)
obj_station = CleanText('./div[@class="box-body"]/div/div/p[@class="item-transports"]', default=NotAvailable)
obj_location = CleanText('./div[@class="box-body"]/div/div/p[@class="item-description"]/strong')
obj_text = CleanText('./div[@class="box-body"]/div/div/p[@class="item-description"]')
obj_rooms = CleanDecimal(
'./div[@class="box-body"]/div/div/div[@class="clearfix"]/ul[has-class("item-summary")]/li[1]/strong',
default=NotAvailable
)
obj_price_per_meter = PricePerMeterFilter()
def obj_bedrooms(self):
rooms_bedrooms_area = XPath(
'./div[@class="box-body"]/div/div/div[@class="clearfix"]/ul[has-class("item-summary")]/li'
)(self)
if len(rooms_bedrooms_area) > 2:
return CleanDecimal(
'./div[@class="box-body"]/div/div/div[@class="clearfix"]/ul[has-class("item-summary")]/li[2]/strong',
default=NotAvailable
)(self)
else:
return NotAvailable
obj_url = Format(
u'http://www.pap.fr%s',
Link(
'./div[@class="box-body"]/div/div/div[@class="clearfix"]/div[@class="float-right"]/a'
)
)
def obj_photos(self):
photos = []
for img in XPath('./div[@class="box-body"]/div/div/a/img/@src')(self):
photos.append(HousingPhoto(u'%s' % img))
return photos
class item(ItemElement):
klass = Bill
load_details = Link('.//a[contains(text(), "détails")]') & MyAsyncLoad
obj_id = Format('%s_%s', Env('subid'), Field('label'))
obj_url = Async('details') & Link('//a[span[contains(text(), "Télécharger la facture")]]', default=NotAvailable)
obj_date = Date(CleanText('./div[contains(@class, "cell-date")]'), dayfirst=True)
obj_format = 'pdf'
obj_label = Regexp(CleanText('./div[contains(@class, "cell-nb-order")]'), r' (.*)')
obj_type = DocumentTypes.BILL
obj_price = CleanDecimal(CleanText('./div[contains(@class, "cell-value")]'), replace_dots=(' ', '€'))
obj_currency = 'EUR'
class item(ItemElement):
klass = Account
load_details = Attr('.//a', 'href') & AsyncLoad
obj__link_id = Async(
'details', Link('//li/a[contains(text(), "Mouvements")]'))
obj__link_inv = Link('./td[1]/a')
obj_id = CleanText('./td[2]', replace=[(' ', '')])
obj_label = CleanText('./td[1]')
obj_balance = CleanDecimal('./td[3]', replace_dots=True)
obj_currency = FrenchTransaction.Currency('./td[4]')
obj__card_links = []
obj_type = Account.TYPE_LIFE_INSURANCE
obj__is_inv = True
def populate(self, accounts):
cards = []
for account in accounts:
for li in self.doc.xpath('//li[@class="nav-category"]'):
title = CleanText().filter(li.xpath('./h3'))
for a in li.xpath('./ul/li//a'):
label = CleanText().filter(
a.xpath('.//span[@class="nav-category__name"]'))
balance_el = a.xpath(
'.//span[@class="nav-category__value"]')
balance = CleanDecimal(
replace_dots=True,
default=NotAvailable).filter(balance_el)
if 'CARTE' in label and balance:
acc = Account()
acc.balance = balance
acc.label = label
acc.currency = FrenchTransaction.Currency().filter(
balance_el)
acc._link = Link().filter(a.xpath('.'))
acc._history_page = acc._link
acc.id = acc._webid = Regexp(
pattern='([^=]+)$').filter(Link().filter(
a.xpath('.')))
acc.type = Account.TYPE_CARD
if not acc in cards:
cards.append(acc)
elif account.label == label and account.balance == balance:
if not account.type:
account.type = AccountsPage.ACCOUNT_TYPES.get(
title, Account.TYPE_UNKNOWN)
if account.type == Account.TYPE_LOAN:
account._history_page = None
elif account.type in (Account.TYPE_LIFE_INSURANCE,
Account.TYPE_MARKET):
account._history_page = re.sub(
'/$', '',
Link().filter(a.xpath('.')))
elif '/compte/cav' in a.attrib[
'href'] or not 'titulaire' in self.url:
account._history_page = self.browser.other_transactions
else:
account._history_page = self.browser.budget_transactions
account._webid = Attr(
None, 'data-account-label').filter(
a.xpath(
'.//span[@class="nav-category__name"]'))
accounts.extend(cards)
class get_history(ListElement):
item_xpath = '//table[@id="ContentTable_datas"]//tr[@class]'
next_page = Link('//table[@id="tgDecorationFoot"]//b/following-sibling::a[1]')
class item(ItemElement):
klass = FrenchTransaction
obj_rdate = FrenchTransaction.Date(CleanText('./td[1]'))
obj_date = FrenchTransaction.Date(CleanText('./td[3]'))
obj_raw = FrenchTransaction.Raw(CleanText('./td[2]'))
_obj_amnt = FrenchTransaction.Amount(CleanText('./td[5]'), replace_dots=False)
obj_original_amount = FrenchTransaction.Amount(CleanText('./td[4]'), replace_dots=False)
obj_original_currency = FrenchTransaction.Currency(CleanText('./td[4]'))
obj_commission = FrenchTransaction.Amount(CleanText('./td[6]'), replace_dots=False)
def obj__coming(self):
if Field('date')(self) >= date.today():
return True
return
def obj_amount(self):
if not Field('obj_commission'):
return Field('_obj_amnt')
else:
return CleanDecimal(replace_dots=False).filter(self.el.xpath('./td[5]')) - CleanDecimal(replace_dots=False).filter(self.el.xpath('./td[6]'))
class iter_history(TableElement):
item_xpath = '//table[@class="Tableau"]/tr[td[not(has-class("enteteTableau"))]]'
head_xpath = '//table[@class="Tableau"]/tr[td[has-class("enteteTableau")]]/td'
col_date = u'Date d\'effet'
col_raw = u'Nature du mouvement'
col_amount = u'Montant brut'
next_page = Link('//a[contains(@href, "PageSuivante")]', default=None)
class item(ItemElement):
klass = Transaction
load_details = Link('./td/a', default=None) & AsyncLoad
obj_date = Date(CleanText(TableCell('date')), dayfirst=True)
obj_raw = Transaction.Raw(TableCell('raw'))
obj_amount = CleanDecimal(TableCell('amount'), replace_dots=True)
obj_investments = Env('investments')
def parse(self, el):
try:
page = Async('details').loaded_page(self)
except AttributeError:
page = None
self.env['investments'] = list(page.get_investments(
)) if page and 'numMvt' in page.url else []
class iter_history(TableElement):
item_xpath = '//table/tbody/tr'
head_xpath = '//table/thead/tr/th'
col_label = 'Nature'
col_amount = 'Montant'
col_date = 'Date d\'effet'
next_page = Link('//li[@class="pagination__next"]/a')
class item(ItemElement):
klass = Transaction
obj_date = Date(CleanText(TableCell('date')), dayfirst=True)
obj_raw = Transaction.Raw(CleanText(TableCell('label')))
obj_amount = CleanDecimal(TableCell('amount'),
replace_dots=True,
default=NotAvailable)
obj__is_coming = False
def parse(self, el):
if el.xpath('./td[2]/a'):
m = re.search(
'(\d+)',
el.xpath('./td[2]/a')[0].get(
'data-modal-alert-behavior', ''))
if m:
self.env['account']._history_pages.append((Field('raw')(self),\
self.page.browser.open('%s%s%s' % (self.page.url.split('mouvements')[0], 'mouvement/', m.group(1))).page))
raise SkipItem()
class iter_prices(ListElement):
item_xpath = '//div[@class="adContainer "]'
next_page = Link(
'//section[@class="pagination"]/ul/li[@class="last"]/a')
class item(ItemElement):
klass = Price
obj_id = CleanText('./p/a/@data-annid')
obj_cost = CleanDecimal('./a/div/div/div/div[@class="fieldPrice"]')
obj_currency = Regexp(
CleanText('./a/div/div/div/div[@class="fieldPrice"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'),
default=u'€')
obj_message = Format(
'%s / %s / %s', CleanText('./a/div/div/h3'),
CleanText('./a/div/div/div/div[@class="fieldYear"]'),
CleanText('./a/div/div/div/div[@class="fieldMileage"]'))
obj_url = Format('http://www.lacentrale.fr%s',
CleanText('./a/@href'))
obj_product = LaCentraleProduct()
def obj_shop(self):
shop = Shop(CleanText('./p/a/@data-annid')(self))
return shop
def obj_code(self):
link = Link(TableCell('label')(self)[0].xpath('./a'),
default=NotAvailable)(self)
if not link:
return NotAvailable
return Regexp(pattern='isin=([A-Z\d]+)&?',
default=NotAvailable).filter(link)
def companies_link(self):
companies_link = []
for tr in self.doc.xpath(
'//table[@summary="Liste des titulaires de contrats cartes"]//tr'
):
companies_link.append(Link(tr.xpath('.//a'))(self))
return companies_link
class iter_history(ListElement):
item_xpath = '''//table[@class="ca-table"][caption[span[b[contains(text(), "HISTORIQUE DES OPERATIONS")]]]]
//tr[contains(@class, "ligne-")]'''
next_page = Link(
'//a[@class="liennavigationcorpspage"][img[@alt="Page suivante"]]',
default=None)
class item(ItemElement):
klass = Transaction
def fill_env(self, page, parent=None):
# This *Element's parent has only the dateguesser in its env and we want to
# use the same object, not copy it.
self.env = parent.env
def obj_date(self):
# Dates in the first column may appear as '12/01/2019' or '12/01'
date = CleanText('./td[1]/font//text()')(self)
if len(date) == 10:
return Date(CleanText('./td[1]/font//text()'),
dayfirst=True)(self)
elif len(date) == 5:
# Date has no indicated year.
return DateGuesser(CleanText('./td[1]//text()'),
Env('date_guesser'))(self)
obj_raw = Transaction.Raw(CleanText('./td[2]/font//text()'))
obj_amount = CleanDecimal.French('./td[3]/font//text()')
obj_rdate = Field('date')
class item(ItemElement):
klass = Account
obj__owner = Regexp(
CleanText('./td[1]/text()', replace=[(' ', '')]),
'Titulaire:(.*)')
obj_id = Format(
'%s%s',
Regexp(CleanText('./td[1]/a', replace=[(' ', '')]), '([\d]+)'),
Field('_owner'))
obj_label = Field('_owner')
obj_balance = NotAvailable
obj_currency = FrenchTransaction.Currency('./td[2]')
obj__link_id = Link('./td[1]/a')
obj_type = Account.TYPE_CARD
obj__card_links = []
obj__is_inv = False
obj__is_webid = False
def parse(self, el):
account = [
acc for acc in self.env['accounts']
if acc.id == Field('id')(self)
]
if account:
account[0]._card_links.append(Field('_link_id')(self))
raise SkipItem()
class get_operations(ListElement):
item_xpath = '//table[has-class("style-operations")]/tbody//tr'
next_page = Link(
'//div[@class="m-table-paginator full-width-xs"]//a[@id="next-page"]'
)
class item(ItemElement):
klass = Transaction
def condition(self):
if 'tr-section' in self.el.attrib.get('class', ''):
return False
elif 'tr-trigger' in self.el.attrib.get('class', ''):
return True
return False
def obj_date(self):
return Transaction.Date(
Regexp(
CleanText(
'./preceding::tr[has-class("tr-section")][1]/th'),
r'(\d+/\d+/\d+)'))(self)
obj_raw = Transaction.Raw(
Format(
'%s %s', CleanText('./td[1]'),
CleanText(
'./following-sibling::tr[contains(@class, "tr-more")]/td/p[1]/span'
)))
obj_amount = MyDecimal('./td[2]', replace_dots=True)
class item(ItemElement):
def klass(self):
return BiplanCalendarEventConcert() if self.env['is_concert'] else BiplanCalendarEventTheatre()
def condition(self):
return (self.el.xpath('./div') and CleanText('./div/a/img/@src')(self)[-1] != '/')
def validate(self, obj):
return (self.is_valid_event(obj, self.env['city'], self.env['categories']) and
self.is_event_in_valid_period(obj.start_date, self.env['date_from'], self.env['date_to']))
def is_valid_event(self, event, city, categories):
if city and city != '' and city.upper() != event.city.upper():
return False
if categories and len(categories) > 0 and event.category not in categories:
return False
return True
def is_event_in_valid_period(self, event_date, date_from, date_to):
if event_date >= date_from:
if not date_to:
return True
else:
if event_date <= date_to:
return True
return False
obj_id = Regexp(Link('./div/a'), '/(.*?).html')
obj_start_date = CombineDate(BiplanDate('div/div/b'), StartTime('div/div/b'))
obj_end_date = CombineDate(BiplanDate('div/div/b'), EndTime('.'))
obj_price = BiplanPrice('div/div/b')
obj_summary = CleanText("div/div/div/a/strong")
class item(ItemElement):
klass = Account
def obj_id(self):
area_id = Regexp(CleanText('//span[@class="CelMnTiersT1"]'),
r'\((\d+)\)',
default='')(self)
acc_id = Regexp(CleanText('./td[1]'), r'(\d+)\s*(\d+)',
r'\1\2')(self)
if area_id:
return '%s.%s' % (area_id, acc_id)
return acc_id
def obj__formdata(self):
js = Attr('./td/a[1]', 'onclick', default=None)(self)
if js is None:
return
args = re.search(r'\((.*)\)', js).group(1).split(',')
form = args[0].strip().split('.')[1]
idx = args[2].strip()
idroot = args[4].strip().replace("'", "")
return (form, idx, idroot)
obj_url = Link('./td/a[1]', default=None)
class iter_history(TableElement):
item_xpath = '//table[@class="Tableau"]/tr[td[not(has-class("enteteTableau"))]]'
head_xpath = '//table[@class="Tableau"]/tr[td[has-class("enteteTableau")]]/td'
col_date = u'Date d\'effet'
col_raw = u'Nature du mouvement'
col_amount = u'Montant brut'
next_page = Link('//a[contains(@href, "PageSuivante")]', default=None)
class item(ItemElement):
klass = Transaction
obj_date = Date(CleanText(TableCell('date')), dayfirst=True)
obj_raw = Transaction.Raw(TableCell('raw'))
obj_amount = CleanDecimal(TableCell('amount'), replace_dots=True)
obj__detail = Env('detail')
def obj_id(self):
try:
return Regexp(Link('./td/a', default=None), 'numMvt=(\d+)', default=None)(self)
except TypeError:
return NotAvailable
def parse(self, el):
link = Link('./td/a', default=None)(self)
page = self.page.browser.async_open(link) if link else None
self.env['detail'] = page
def get_account_details(self, account_id):
balance = CleanDecimal.French(
'//a[div[div[span[span[contains(text(), "%s")]]]]]/div[1]/div[2]/span/span'
% account_id,
default=NotAvailable)(self.doc)
currency = Currency(
'//a[div[div[span[span[contains(text(), "%s")]]]]]/div[1]/div[2]/span/span'
% account_id,
default=NotAvailable)(self.doc)
label = CleanText(
'//a[div[div[span[span[contains(text(), "%s")]]]]]/div[1]/div[1]/span/span'
% account_id,
default=NotAvailable)(self.doc)
url = Link('//a[div[div[span[span[contains(text(), "%s")]]]]]' %
account_id,
default=None)(self.doc)
if url:
account_url = 'https://bgpi-gestionprivee.credit-agricole.fr' + url
else:
account_url = None
return balance, currency, label, account_url
class item(ItemElement):
klass = Housing
def condition(self):
return Regexp(Link('./div[@class="header-annonce"]/a'), '/annonces/(.*)', default=None)(self)
obj_id = Regexp(Link('./div[@class="header-annonce"]/a'), '/annonces/(.*)')
obj_title = CleanText('./div[@class="header-annonce"]/a')
obj_area = CleanDecimal(Regexp(CleanText('./div[@class="header-annonce"]/a/span[@class="desc"]'),
'(.*?)(\d*) m\xb2(.*?)', '\\2'), default=NotAvailable)
obj_cost = CleanDecimal(CleanText('./div[@class="header-annonce"]/a/span[@class="prix"]'),
replace_dots=(',', '.'), default=Decimal(0))
obj_currency = Regexp(CleanText('./div[@class="header-annonce"]/a/span[@class="prix"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'), default=u'€')
def obj_date(self):
_date = CleanText('./div[@class="header-annonce"]/span[@class="date"]')(self)
return parse_french_date(_date)
obj_station = CleanText('./div/div/div[@cladd=metro]', default=NotAvailable)
obj_location = CleanText('./div[@class="clearfix"]/div/a/span/img/@alt')
obj_text = CleanText('./div[@class="clearfix"]/div[@class="description clearfix"]/p')
def obj_photos(self):
photos = []
for img in XPath('//div[@class="vignette-annonce"]/a/span/img/@src')(self):
photos.append(HousingPhoto(u'%s' % img))
return photos
def populate(self, accounts):
cards = []
for account in accounts:
for li in self.doc.xpath('//li[@class="nav-category"]'):
title = CleanText().filter(li.xpath('./h3'))
for a in li.xpath('./ul/li//a'):
label = CleanText().filter(a.xpath('.//span[@class="nav-category__name"]'))
balance_el = a.xpath('.//span[@class="nav-category__value"]')
balance = CleanDecimal(replace_dots=True, default=NotAvailable).filter(balance_el)
if 'CARTE' in label and not empty(balance):
acc = Account()
acc.balance = balance
acc.label = label
acc.currency = FrenchTransaction.Currency().filter(balance_el)
acc.url = urljoin(self.url, Link().filter(a.xpath('.')))
acc._history_page = acc.url
try:
acc.id = acc._webid = Regexp(pattern='carte/(.*)$').filter(Link().filter(a.xpath('.')))
except RegexpError:
# Those are external cards, ie: amex cards
continue
acc.type = Account.TYPE_CARD
if not acc in cards:
cards.append(acc)
elif account.label == label and account.balance == balance:
if not account.type:
account.type = AccountsPage.ACCOUNT_TYPES.get(title, Account.TYPE_UNKNOWN)
account._webid = Attr(None, 'data-account-label').filter(a.xpath('.//span[@class="nav-category__name"]'))
if cards:
self.browser.go_cards_number(cards[0].url)
if self.browser.cards.is_here():
self.browser.page.populate_cards_number(cards)
accounts.extend(cards)
