def obj_coming(self):
page = self.page.browser.open(
BrowserURL('account_coming',
identifiant=Field('iban'))(self)).page
coming = Eval(
lambda x, y: x / 10**y,
CleanDecimal(
Dict('infoOperationsAvenir/cumulTotal/montant',
default='0')),
CleanDecimal(
Dict('infoOperationsAvenir/cumulTotal/nb_dec',
default='0')))(page.doc)
# this so that card coming transactions aren't accounted twice in the total incoming amount
for el in Dict('infoOperationsAvenir/natures')(page.doc):
if Dict('nature/libelle')(
el) == "Factures / Retraits cartes":
coming_carte = Eval(
lambda x, y: x / 10**y,
CleanDecimal(
Dict('cumulNatureMere/montant', default='0')),
CleanDecimal(
Dict('cumulNatureMere/nb_dec',
default='0')))(el)
coming -= coming_carte
break
return coming
class get_price(ItemElement):
klass = Price
obj_id = Env('_id')
obj_cost = CleanDecimal(
'//div[@class="mainInfos"]/div/p[@class="gpfzj"]')
obj_currency = Regexp(
CleanText('//div[@class="mainInfos"]/div/p[@class="gpfzj"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'),
default=u'€')
obj_message = Format('%s %s',
CleanText('//div[@class="mainInfos"]/div/div/h1'),
CleanText('//div[@class="mainInfos"]/div/div/p'))
obj_url = BrowserURL('advert_page', _id=Env('_id'))
def obj_shop(self):
shop = Shop(Env('_id')(self))
shop.name = Regexp(
CleanText(
'(//div[@xtcz="contacter_le_vendeur"]/div/ul/li)[1]'),
'Nom : (.*)')(self)
shop.location = JSVar(CleanText('//script'), var='tooltip')(self)
shop.info = CleanText(
'//div[@xtcz="contacter_le_vendeur"]/div/ul/li[has-class("printPhone")]'
)(self)
return shop
obj_product = LaCentraleProduct()
def next_page(self):
page = Regexp(CleanText('//link[@rel="next"]/@href', default=''),
'.*pg=(\d*)',
default=None)(self)
return BrowserURL('adv_search',
search=Env('search'),
page=int(page))(self)
class item(ItemElement):
klass = Subscription
load_details = BrowserURL('profilpage') & AsyncLoad
obj_subscriber = Env('subscriber')
obj_label = Env('subid')
obj_id = obj_label
obj__multi = Env('multi')
def parse(self, el):
subscriber = Async(
'details',
CleanText(
u'//span[contains(text(), "prénom / nom")]/following-sibling::span[1]'
))(self)
self.env['subscriber'] = subscriber if subscriber else \
Async('details', Format('%s %s %s', \
CleanText(u'//*[contains(text(), "civilité")]/following-sibling::*[1]'), \
CleanText(u'//*[contains(text(), "prénom")]/following-sibling::*[1]'), \
CleanText(u'//*[text() = "nom :"]/following-sibling::*[1]')))(self)
subid = Regexp(Attr('.', 'ecareurl', default="None"),
'idContrat=(\d+)',
default=None)(self)
self.env[
'subid'] = subid if subid else self.page.browser.username
self.env['multi'] = True if subid else False
# Prevent from available account but no added in customer area
if subid and not json.loads(
self.page.browser.open(Attr(
'.', 'ecareurl')(self)).content)['html']:
raise SkipItem()
class item(ItemElement):
klass = Bill
obj_date = Date(Dict('dueDate'),
parse_func=parse_french_date,
default=NotAvailable)
obj_price = CleanDecimal(Dict('amountIncludingTax'))
obj_format = 'pdf'
def obj_label(self):
return 'Facture du %s' % Field('date')(self)
def obj_id(self):
return '%s_%s' % (Env('subid')(self),
Field('date')(self).strftime('%d%m%Y'))
def get_params(self):
params = {
'billid': Dict('id')(self),
'billDate': Dict('dueDate')(self)
}
return urlencode(params)
obj_url = BrowserURL('doc_api_pro',
subid=Env('subid'),
dir=Dict('documents/0/mainDir'),
fact_type=Dict('documents/0/subDir'),
billparams=get_params)
obj__is_v2 = False
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_description = CleanText(
Join('\n',
'//div[@id="annonce-detail"]/p[@class="text"]',
textCleaner=CleanHTML))
obj_id = Env('_id')
obj_url = BrowserURL('advert_page', _id=Env('_id'))
obj_publication_date = Date(Regexp(
CleanText('//div[@id="annonce-detail"]/p[@class="infos"]'),
'(\d{2}/\d{2}/\d{4})',
default=NotAvailable),
default=NotAvailable)
obj_title = CleanText('//div[@id="annonce"]/div/div/h1')
obj_society_name = CleanText('//section[@class="entp-resume"]/h1/a')
obj_contract_type = CleanText(
'//dl[@class="infos-annonce"]/dt[span[@class="picto picto-contrat-grey"]]/following-sibling::dd[1]'
)
obj_place = CleanText(
'//dl[@class="infos-annonce"]/dt[span[@class="picto picto-geolocalisation-grey"]]/following-sibling::dd[1]'
)
obj_pay = CleanText(
'//div[@id="annonce-detail"]/p[@class="infos"]/preceding-sibling::p[1]',
replace=[('Salaire : ', '')])
class get_housing(ItemElement):
klass = Housing
def parse(self, el):
details = dict()
self.env['area'] = NotAvailable
for item in el.xpath('//div[@class="line"]/h2'):
if 'Surface' in CleanText('./span[@class="property"]')(item):
self.env['area'] = CleanDecimal(
Regexp(CleanText('./span[@class="value"]'), '(.*)m.*'),
replace_dots=(',', '.'))(item)
else:
key = u'%s' % CleanText('./span[@class="property"]')(item)
if 'GES' in key or 'Classe' in key:
details[key] = CleanText(
'./span[@class="value"]/noscript/a')(item)
else:
details[key] = CleanText('./span[@class="value"]')(
item)
self.env['details'] = details
obj_id = Env('_id')
obj_title = CleanText('//title')
obj_cost = CleanDecimal('//h2[@itemprop="price"]/@content',
default=Decimal(0))
obj_currency = Regexp(
CleanText('//h2[@itemprop="price"]/span[@class="value"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'),
default=u'€')
obj_text = CleanText('//meta[@name="description"]/@content')
obj_location = CleanText('//span[@itemprop="address"]')
obj_details = Env('details')
obj_area = Env('area')
obj_url = BrowserURL('housing', _id=Env('_id'))
def obj_date(self):
_date = Regexp(
CleanText('//p[has-class("line")]', replace=[(u'à', '')]),
'.*Mise en ligne le (.*)')(self)
for fr, en in DATE_TRANSLATE_FR:
_date = fr.sub(en, _date)
self.env['tmp'] = _date
return DateTime(Env('tmp'), LinearDateGuesser())(self)
def obj_photos(self):
items = re.findall(r'images\[\d\]\s*=\s*"([\w/\.]*\.jpg)";',
CleanText('//script')(self))
photos = [HousingPhoto(u'http:%s' % item) for item in items]
if not photos:
img = CleanText('//meta[@itemprop="image"]/@content',
default=None)(self)
if img:
photos.append(HousingPhoto(img))
return photos
def obj_type(self):
url = BrowserURL('housing', _id=Env('_id'))(self)
if 'colocation' in url:
return POSTS_TYPES.SHARING
elif 'location' in url:
isFurnished = False
for li in XPath('//ul[@itemprop="description"]/li')(self):
label = CleanText('./span[has-class("criteria-label")]')(
li)
if label.lower() == "meublé":
isFurnished = (
CleanText('./span[has-class("criteria-value")]')(
li).lower() == 'oui')
if isFurnished:
return POSTS_TYPES.FURNISHED_RENT
else:
return POSTS_TYPES.RENT
elif 'vente' in url:
offertype = Attr(
'//button[has-class("offer-contact-vertical-phone")][1]',
'data-offertransactiontype')(self)
if offertype == '4':
return POSTS_TYPES.VIAGER
else:
return POSTS_TYPES.SALE
return NotAvailable
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_id = Env('id')
obj_url = BrowserURL('advert_page', id=Env('id'))
obj_society_name = CleanText(
'//td[@class="Contenu"]/table[4]/tr[1]/td[1]/a')
obj_title = CleanText('//td[@class="Titre15"]')
obj_description = Format(
'%s\n%s',
Join(
'\n',
u'//td[@class="Contenu"]/table[3]/tr[td/text()="Détails :"]/following-sibling::tr',
textCleaner=CleanHTML),
CleanHTML('//td[@class="Contenu"]/table[2]'))
obj_job_name = CleanText(
u'//td[@class="Contenu"]/table[3]/tr/td[text()="Poste :"]/following-sibling::td',
replace=[(u'-- Indifférent --', u'')])
obj_contract_type = CleanText(CleanHTML(
u'//td[@class="Contenu"]/table[3]/tr/td[text()="Contrat :"]/following-sibling::td',
default=u''),
replace=[(u'-- Indifférent --', u'')])
obj_pay = CleanText(
u'//td[@class="Contenu"]/table[3]/tr/td[contains(text(), "Rémunération")]/following-sibling::td',
default=u'')
obj_place = CleanText(
u'//td[@class="Contenu"]/table[3]/tr/td[contains(text(), "Région")]/following-sibling::td',
default=u'',
replace=[(u'-- Indifférent --', u''),
(u'Lieu de travail : ', u'')])
class get_housing(ItemElement):
klass = Housing
obj_id = Env('_id')
obj_title = Dict('characteristics/titleWithTransaction')
obj_location = Format('%s %s %s', Dict('location/address'),
Dict('location/postalCode'), Dict('location/cityLabel'))
obj_cost = TypeDecimal(Dict('characteristics/price'))
obj_currency = u'€'
obj_text = CleanHTML(Dict('characteristics/description'))
obj_url = BrowserURL('housing_html', _id=Env('_id'))
obj_area = TypeDecimal(Dict('characteristics/area'))
obj_date = FromTimestamp(Dict('characteristics/date'))
def obj_photos(self):
photos = []
for img in Dict('characteristics/images')(self):
m = re.search('http://thbr\.figarocms\.net.*(http://.*)', img.get('xl'))
if m:
photos.append(HousingPhoto(m.group(1)))
else:
photos.append(HousingPhoto(img.get('xl')))
return photos
def obj_details(self):
details = {}
details['fees'] = Dict('characteristics/fees')(self)
details['bedrooms'] = Dict('characteristics/bedroomCount')(self)
details['energy'] = Dict('characteristics/energyConsumptionCategory')(self)
rooms = Dict('characteristics/roomCount')(self)
if len(rooms):
details['rooms'] = rooms[0]
details['available'] = Dict('characteristics/available', default=NotAvailable)(self)
return details
class get_housing(ItemElement):
klass = Housing
obj_id = Env('_id')
obj_title = CleanText('//h1[@itemprop="name"]')
obj_location = CleanText('//span[@class="informations-localisation"]')
obj_cost = CleanDecimal('//span[@itemprop="price"]')
obj_currency = Regexp(CleanText('//span[@itemprop="price"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'), default=u'€')
obj_text = CleanHTML('//div[@itemprop="description"]')
obj_url = BrowserURL('housing', _id=Env('_id'))
obj_area = CleanDecimal(Regexp(CleanText('//h1[@itemprop="name"]'),
'(.*?)(\d*) m2(.*?)', '\\2'), default=NotAvailable)
def obj_photos(self):
photos = []
for img in XPath('//a[@class="thumbnail-link"]/img[@itemprop="image"]')(self):
url = Regexp(CleanText('./@src'), 'http://thbr\.figarocms\.net.*(http://.*)')(img)
photos.append(HousingPhoto(url))
return photos
def obj_details(self):
details = dict()
for item in XPath('//div[@class="features clearfix"]/ul/li')(self):
key = CleanText('./span[@class="name"]')(item)
value = CleanText('./span[@class="value"]')(item)
if value and key:
details[key] = value
key = CleanText('//div[@class="title-dpe clearfix"]')(self)
value = CleanText('//div[@class="energy-consumption"]')(self)
if value and key:
details[key] = value
return details
def next_page(self):
page = int(Env('page')(self)) + 1
return BrowserURL('webservice',
page=page,
lang=Env('lang'),
method_name=Env('method_name'),
pattern=Env('pattern'))(self)
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_id = Env('id')
obj_url = BrowserURL('advert', id=Env('id'))
obj_title = CleanText('//div[@class="modal-body"]/h2')
obj_job_name = CleanText('//div[@class="modal-body"]/h2')
obj_description = CleanText('//div[has-class("description")]/p')
obj_society_name = CleanText('//div[@class="media-body"]/h4')
obj_experience = Join(
u'- ',
'//h4[contains(text(), "Exp")]/following-sibling::ul[has-class("skill-list")][1]/li',
newline=True,
addBefore='\n- ')
obj_formation = Join(
u'- ',
'//h4[contains(text(), "For")]/following-sibling::ul[has-class("skill-list")][1]/li',
newline=True,
addBefore='\n- ')
obj_place = CleanText(
'//div[@class="modal-body"]/h2/following-sibling::p[1]')
obj_publication_date = PoleEmploiDate(
CleanText('//div[@class="modal-body"]/h2/following-sibling::p[2]'))
def parse(self, el):
for el in XPath('//dl[@class="icon-group"]/dt')(el):
dt = CleanText('.')(el)
if dt == u'Type de contrat':
self.obj.contract_type = CleanText(
'./following-sibling::dd[1]')(el)
elif dt == u'Salaire':
self.obj.pay = Regexp(
CleanText('./following-sibling::dd[1]'),
u'Salaire : (.*)')(el)
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_id = Env('_id')
obj_url = BrowserURL('advert_page', _id=Env('_id'))
obj_title = CleanText('//div[@class="infos-lieu"]/h1')
obj_place = CleanText('//div[@class="infos-lieu"]/h2')
obj_publication_date = Date(
Regexp(CleanText('//div[@class="info-agency"]'),
'.*Date de l\'annonce :(.*)',
default=''))
obj_job_name = CleanText('//div[@class="infos-lieu"]/h1')
obj_description = Format(
'\n%s%s', CleanHTML('//article[@id="post-description"]/div'),
CleanHTML('//article[@id="poste"]'))
obj_contract_type = Regexp(
CleanText('//article[@id="poste"]/div/ul/li'),
'Contrat : (\w*)',
default=NotAvailable)
obj_pay = Regexp(CleanText('//article[@id="poste"]/div/ul/li'),
'Salaire : (.*) par mois',
default=NotAvailable)
obj_experience = Regexp(CleanText('//article[@id="poste"]/div/ul/li'),
u'Expérience : (.* ans)',
default=NotAvailable)
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_id = Env('_id')
obj_url = BrowserURL('expired_advert', _id=Env('_id'))
obj_title = CleanText('//div[@role="alert"]')
obj_description = CleanText('//div[@role="alert"]')
class get_housing(ItemElement):
klass = Housing
obj_id = Env('_id')
obj_title = CleanText('//section[class="wctrl"]/h1')
obj_cost = CleanDecimal('//span[@class="i prix"]',
replace_dots=(',', '.'),
default=Decimal(0))
obj_currency = Regexp(CleanText('//span[@class="i prix"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'),
default='')
obj_text = CleanHTML('//article[@class="bloc description"]/p')
obj_location = CleanText('//span[@class="i ville"]')
obj_area = CleanDecimal(Regexp(CleanText('//span[@class="i"]'),
'.*/(.*) m.*',
default=NotAvailable),
default=NotAvailable)
obj_url = BrowserURL('housing', _id=Env('_id'))
obj_phone = CleanText('//input[@id="hftel"]/@value')
obj_date = datetime.now
def obj_details(self):
details = {}
for detail in self.el.xpath('//span[@class="i small"]'):
item = detail.text.split(':')
details[item[0]] = item[1]
return details
def obj_photos(self):
photos = []
for img in self.el.xpath('//ul[@id="ulPhotos"]/li/img/@src'):
url = u'http://www.entreparticuliers.com/%s' % img
photos.append(HousingPhoto(url))
return photos
def obj_coming(self):
if Field('type')(self) == Account.TYPE_CHECKING:
has_coming = False
coming = 0
coming_operations = self.page.browser.open(
BrowserURL('par_account_checking_coming',
accountId=Field('id'))(self))
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 get_housing(ItemElement):
klass = Housing
obj_id = Env('_id')
obj_title = CleanText(CleanHTML('//meta[@itemprop="name"]/@content'))
obj_area = CleanDecimal(Regexp(
CleanText(CleanHTML('//meta[@itemprop="name"]/@content')),
'(.*?)(\d*) m\xb2(.*?)', '\\2'),
default=NotAvailable)
obj_cost = CleanDecimal('//*[@itemprop="price"]')
obj_currency = Regexp(CleanText('//*[@itemprop="price"]'),
'.*([%s%s%s])' % (u'€', u'$', u'£'),
default=u'€')
obj_date = Date(
Regexp(
CleanText(
'//p[@class="offer-description-notes"]|//p[has-class("darkergrey")]'
), u'.* Mis à jour : (\d{2}/\d{2}/\d{4}).*'))
obj_text = CleanHTML(
'//div[@class="offer-description-text"]|//div[has-class("offer-description")]'
)
obj_location = CleanText('//*[@itemprop="address"]')
obj_url = BrowserURL('housing', _id=Env('_id'))
def obj_photos(self):
photos = []
for img in XPath(
'//div[@class="carousel-content"]/ul/li/a/img/@src|//div[@class="carousel"]/ul/li/a/img/@src'
)(self):
photos.append(HousingPhoto(u'%s' % img))
return photos
def obj_details(self):
details = {}
energy = CleanText(
'//div[has-class("energy-summary")]/span[@class="section-label"]|//div[has-class("energy-summary")]/div/span[@class="section-label"]',
default='')(self)
energy_value = CleanText(
'//div[has-class("energy-summary")]/span[@class="energy-msg"]',
default='')(self)
if energy and energy_value:
details[energy] = energy_value
greenhouse = CleanText(
'//div[has-class("greenhouse-summary")]/span[@class="section-label"]|//div[has-class("greenhouse-summary")]/div/span[@class="section-label"]',
default='')(self)
greenhouse_value = CleanText(
'//div[has-class("greenhouse-summary")]/span[@class="energy-msg"]',
default='')(self)
if greenhouse and greenhouse_value:
details[greenhouse] = greenhouse_value
for li in XPath('//ul[@itemprop="description"]/li')(self):
label = CleanText('./div[has-class("criteria-label")]')(li)
value = CleanText('./div[has-class("criteria-value")]')(li)
details[label] = value
return details
def next_page(self):
link = Link('//a[@title="Page suivante" and @href]',
default=None)(self)
if link:
form = self.page.get_form('//form')
form['__EVENTTARGET'] = re.search(
'PostBackOptions[^\w]+([^"]+)', link).group(1)
return requests.Request("POST", BrowserURL('investment', \
page=None)(self).replace('None', form.url), data=dict(form))
def next_page(self):
str_results = Regexp(
CleanText('//div[@class="m_resultats_recherche_titre"]'),
'.* - (\d* / \d*) .*')(self)
results = str_results.split('/')
if int(results[0]) - int(results[1]) < 10:
return BrowserURL('search',
pattern=Env('pattern'),
start=int(results[0]))(self)
def next_page(self):
page = Regexp(CleanText('//a[@data-action="fetch"]/@href',
default=''),
'.*page=(\d*)',
default=None)(self)
if page:
return BrowserURL('adv_search',
search=Env('search'),
page=int(page))(self)
class get_event(ItemElement):
klass = HybrideCalendarEvent
obj_id = Decode(Env('_id'))
obj_start_date = Date(CleanText('//span[@class="itemDateCreated"]'))
obj_summary = CleanText('//h2[@class="itemTitle"]')
obj_description = Format('%s\n%s',
CleanHTML('//div[@class="itemIntroText"]'),
CleanHTML('//div[@class="itemFullText"]'))
obj_url = BrowserURL('event_page', _id=Env('_id'))
def next_page(self):
page_nb = Dict('navigation/pagination/page')(self)
max_results = Dict('navigation/pagination/maxResults')(self)
results_per_page = Dict('navigation/pagination/resultsPerPage')(
self)
if int(max_results) / int(results_per_page) > int(page_nb):
return BrowserURL('search',
query=Env('query'),
page_number=int(page_nb) + 1)(self)
def next_page(self):
first_item = Regexp(CleanText(
'//li[@class="suiv"]/a/@data-searchparams', default=None),
'b=(.*)&q=.*',
default=None)
if first_item(self):
return BrowserURL('search_page',
pattern=Env('pattern'),
type=Env('type'),
first_item=first_item)(self)
def obj_coming(self):
page = self.page.browser.open(
BrowserURL('account_coming',
identifiant=Field('iban'))(self)).page
return Eval(
lambda x, y: x / 10**y,
CleanDecimal(
Dict('infoOperationsAvenir/cumulTotal/montant',
default='0')),
CleanDecimal(
Dict('infoOperationsAvenir/cumulTotal/nb_dec',
default='0')))(page.doc)
def load_details(self):
a = self.el.xpath(
".//img[@src='../../images/commun/loupe.png']")
if len(a) > 0:
values = a[0].get('onclick').replace('OpenDetailOperation(', '') \
.replace(')', '').replace(' ', '').replace("'", '').split(',')
keys = ["nummvt", "&numads", "dtmvt", "typmvt"]
data = dict(zip(keys, values))
url = BrowserURL('history_detail')(self)
r = self.page.browser.async_open(url=url, data=data)
return r
return None
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_id = Env('_id')
obj_url = BrowserURL('advert', _id=Env('_id'))
obj_title = CleanText('//h2')
obj_description = Format('%s\n%s',
CleanHTML('//div[@id="JobDescription"]'),
CleanText('//dl'))
obj_contract_type = CleanText('(//dl/dt[text()="Type de contrat"]/following-sibling::dd)[1]')
obj_society_name = CleanText('//div[@data-jsux="aboutCompany"]/div/dl/dd')
obj_place = CleanText('//h3')
obj_publication_date = MonsterDate(CleanText('(//dl/dt[starts-with(text(),"Publi")]/following-sibling::dd)[1]'))
class get_lyrics(ItemElement):
klass = SongLyrics
def condition(self):
return not XPath('//div[has-class("lyric-no-data")]')(self)
obj_id = Env('id')
obj_url = BrowserURL('songLyrics', id=Env('id'))
obj_content = CleanHTML('//pre[@id="lyric-body-text"]',
default=NotAvailable)
obj_title = CleanText('//h2[@id="lyric-title-text"]')
obj_artist = CleanText('//h3[@class="lyric-artist"]/a[1]',
default=NotAvailable)
class item(ItemElement):
klass = Document
obj_id = Format("%s_%s", Env("subscription"), Dict("periode"))
obj_format = "pdf"
obj_date = Date(Format("%s-12-31", Dict("periode")))
obj_label = Format("Attestation fiscale %s", Dict("periode"))
obj_type = DocumentTypes.OTHER
obj_url = BrowserURL(
"tax_certificate_download",
employer=Env("employer"),
year=Dict("periode"),
)
class get_job_advert(ItemElement):
klass = BaseJobAdvert
obj_id = Env('_id')
obj_url = BrowserURL('advert', _id=Env('_id'))
obj_title = CleanText('//div[@id="jobcopy"]/h1[@itemprop="title"]')
obj_description = CleanHTML('//div[@id="jobBodyContent"]')
obj_contract_type = Join('%s ',
'//dd[starts-with(@class, "multipledd")]')
obj_society_name = CleanText('//dd[@itemprop="hiringOrganization"]')
obj_place = CleanText('//span[@itemprop="jobLocation"]')
obj_pay = CleanText('//span[@itemprop="baseSalary"]')
obj_formation = CleanText('//span[@itemprop="educationRequirements"]')
obj_experience = CleanText('//span[@itemprop="qualifications"]')