def parse(self, el):
raw = self.extract.match(el.text).group("html")
raw = raw.replace('\\"', '"').replace('\\n',
'').replace('\\/', '/')
parsed = lxml.html.fromstring(raw)
self.env['name'] = CleanText(
'.//span[@class="popUpTitleBold"]')(parsed)
self.env['object'] = CleanText(
'.//span[@class="popUpTitleNormal"]')(parsed).strip(' /')
url = Attr('.//div[@class="popUpMsDiagramm"]/img',
'src')(parsed)
self.env['id'] = url.split('_')[1]
for tr in parsed.xpath('.//tr'):
td = tr.xpath('.//td')
if len(td) == 1 and "Datum" in td[0].text:
l = td[0].text.split()[1:3]
self.env['datetime'] = "%s %s" % (l[0], l[1])
elif len(td) == 2:
if "Wasserstand" in td[0].text:
self.env['levelvalue'] = td[1].text.split()[0]
elif "Durchfluss" in td[0].text:
self.env['flowvalue'] = td[1].text.split()[0]
elif "Tendenz" in td[0].text:
try:
self.env['forecast'] = Attr('img', 'src')(
td[1]).split("/")[-1]
except ParseError:
self.env['forecast'] = None
# TODO
self.env['alarm'] = None
def parse(self, el):
raw = self.extract.match(el.text).group("html")
raw = raw.replace('\\"', '"').replace('\\n', '').replace('\\/', '/')
parsed = lxml.html.fromstring(raw)
self.env['name'] = CleanText('.//span[@class="popUpTitleBold"]')(parsed)
self.env['object'] = CleanText('.//span[@class="popUpTitleNormal"]')(parsed).strip(' /')
url = Attr('.//div[@class="popUpMsDiagramm"]/img', 'src')(parsed)
self.env['id'] = url.split('_')[1]
for tr in parsed.xpath('.//tr'):
td = tr.xpath('.//td')
if len(td) == 1 and "Datum" in td[0].text:
l = td[0].text.split()[1:3]
self.env['datetime'] = "%s %s" % (l[0], l[1])
elif len(td) == 2:
if "Wasserstand" in td[0].text:
self.env['levelvalue'] = td[1].text.split()[0]
elif "Durchfluss" in td[0].text:
self.env['flowvalue'] = td[1].text.split()[0]
elif "Tendenz" in td[0].text:
try:
self.env['forecast'] = Attr('img', 'src')(td[1]).split("/")[-1]
except ParseError:
self.env['forecast'] = None
# TODO
self.env['alarm'] = None
def get_history_jid(self):
if self.is_asv:
# Assurance Vie, we do not support this kind of account.
return None
span = Attr('//span[starts-with(@id, "index:j_id")]', 'id')(self.doc)
return span.split(':')[1]
def get_history_jid(self):
span = self.doc.xpath('//span[@id="index:panelASV"]')
if len(span) > 1:
# Assurance Vie, we do not support this kind of account.
return None
span = Attr('//span[starts-with(@id, "index:j_id")]', 'id')(self.doc)
jid = span.split(':')[1]
return jid
def obj_DPE(self):
DPE = Attr(
'//div[has-class("energy-box")]//div[has-class("energy-rank")]',
'class',
default=""
)(self)
if DPE:
DPE = [x.replace("energy-rank-", "").upper()
for x in DPE.split() if x.startswith("energy-rank-")][0]
return getattr(ENERGY_CLASS, DPE, NotAvailable)
def obj_DPE(self):
DPE = Attr(
'//div[has-class("energy-box")]//div[has-class("energy-rank")]',
'class',
default=""
)(self)
if DPE:
DPE = [x.replace("energy-rank-", "").upper()
for x in DPE.split() if x.startswith("energy-rank-")][0]
return getattr(ENERGY_CLASS, DPE, NotAvailable)
def obj_details(self):
GES = Attr(
'//div[has-class("energy-box")]//div[has-class("rank")]',
'class',
default=None
)(self)
if GES:
GES = [x.replace("rank-", "").upper()
for x in GES.split() if x.startswith("rank-")][0]
else:
GES = NotAvailable
return {
"GES": GES
}
def obj_level(self):
classes = Attr(u'//*[@class="lignes"]//div[@id="%s"]' %
self.env[u'line'],
attr='class')(self)
classes = classes.split()
if u"perturb_critique_trav" in classes:
return CRITICAL_AND_WORK
elif u"perturb_critique" in classes:
return CRITICAL
elif u"perturb_alerte_trav" in classes:
return ALERT_AND_WORK
elif u"perturb_alerte" in classes:
return ALERT
elif u"perturb_normal_trav" in classes:
return NORMAL_AND_WORK
elif u"perturb_normal" in classes:
return NORMAL
def obj_level(self):
classes = Attr(
u'//*[@class="lignes"]//div[@id="%s"]' % self.env[u'line'],
attr='class'
)(self)
classes = classes.split()
if u"perturb_critique_trav" in classes:
return CRITICAL_AND_WORK
elif u"perturb_critique" in classes:
return CRITICAL
elif u"perturb_alerte_trav" in classes:
return ALERT_AND_WORK
elif u"perturb_alerte" in classes:
return ALERT
elif u"perturb_normal_trav" in classes:
return NORMAL_AND_WORK
elif u"perturb_normal" in classes:
return NORMAL
def parse(self, el):
# Trying to find vdate and unitvalue
unitvalue, vdate = None, None
for span in TableCell('label')(self)[0].xpath('.//span'):
if unitvalue is None:
unitvalue = Regexp(CleanText('.'), '^([\d,]+)$',
default=None)(span)
if vdate is None:
vdate = None if any(x in CleanText('./parent::div')(span) for x in [u"échéance", "Maturity"]) else \
Regexp(CleanText('.'), '^([\d\/]+)$', default=None)(span)
self.env['unitvalue'] = MyDecimal().filter(
unitvalue) if unitvalue else NotAvailable
self.env['vdate'] = Date(
dayfirst=True).filter(vdate) if vdate else NotAvailable
page = None
link_id = Attr(u'.//a[contains(@title, "détail du fonds")]',
'id',
default=None)(self)
inv_id = Attr('.//a[contains(@id, "linkpdf")]', 'id',
default=None)(self)
if link_id and inv_id:
form = self.page.get_form('//div[@id="operation"]//form')
form['idFonds'] = inv_id.split('-', 1)[-1]
form['org.richfaces.ajax.component'] = form[link_id] = link_id
page = self.page.browser.open(form['javax.faces.encodedURL'],
data=dict(form)).page
if "hsbc.fr" in self.page.browser.BASEURL: # special space for HSBC
m = re.search('fundid=(\w+).+SH=(\w+)',
CleanText('//complete', default="")(page.doc))
if m: # had to put full url to skip redirections.
page = page.browser.open(
'https://www.assetmanagement.hsbc.com/feedRequest?feed_data=gfcFundData&cod=FR&client=FCPE&fId=%s&SH=%s&lId=fr'
% m.groups()).page
elif "consulteroperations" not in self.page.browser.url: # not on history
url = Regexp(CleanText('//complete'),
r"openUrlFichesFonds\('(.*?)'\)",
default=NotAvailable)(page.doc)
if url is NotAvailable:
# redirection to a useless graphplot page with url like /portal/salarie-sg/fichefonds?idFonds=XXX&source=/portal/salarie-sg/monepargne/mesavoirs
assert CleanText('//redirect/@url')(page.doc)
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
if url.startswith('http://docfinder.is.bnpparibas-ip.com/'):
# pdf... http://docfinder.is.bnpparibas-ip.com/api/files/040d05b3-1776-4991-aa49-f0cd8717dab8/1536
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
match = re.match(
r'http://www.cpr-am.fr/fr/fonds_detail.php\?isin=([A-Z0-9]+)',
url)
match = match or re.match(
r'http://www.cpr-am.fr/particuliers/product/view/([A-Z0-9]+)',
url)
if match:
self.env['code'] = match.group(1)
self.env['code_type'] = Investment.CODE_TYPE_ISIN
return
page = self.page.browser.open(url).page
try:
self.env['code'] = page.get_code()
self.env['code_type'] = page.CODE_TYPE
# Handle page is None and page has not get_code method
except AttributeError:
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
def get_history_jid(self):
span = Attr('//*[starts-with(@id, "index:j_id")]', 'id')(self.doc)
jid = span.split(':')[1]
return jid
def get_history_jid(self):
span = Attr('//*[starts-with(@id, "index:j_id")]', 'id')(self.doc)
jid = span.split(':')[1]
return jid
def parse(self, el):
# Trying to find vdate and unitvalue
unitvalue, vdate = None, None
for span in TableCell('label')(self)[0].xpath('.//span'):
if unitvalue is None:
unitvalue = Regexp(CleanText('.'), '^([\d,]+)$', default=None)(span)
if vdate is None:
vdate = None if any(x in CleanText('./parent::div')(span) for x in [u"échéance", "Maturity"]) else \
Regexp(CleanText('.'), '^([\d\/]+)$', default=None)(span)
self.env['unitvalue'] = MyDecimal().filter(unitvalue) if unitvalue else NotAvailable
self.env['vdate'] = Date(dayfirst=True).filter(vdate) if vdate else NotAvailable
page = None
link_id = Attr(u'.//a[contains(@title, "détail du fonds")]', 'id', default=None)(self)
inv_id = Attr('.//a[contains(@id, "linkpdf")]', 'id', default=None)(self)
if link_id and inv_id:
form = self.page.get_form('//div[@id="operation"]//form')
form['idFonds'] = inv_id.split('-', 1)[-1]
form['org.richfaces.ajax.component'] = form[link_id] = link_id
page = self.page.browser.open(form['javax.faces.encodedURL'], data=dict(form)).page
if "hsbc.fr" in self.page.browser.BASEURL: # special space for HSBC
m = re.search('fundid=(\w+).+SH=(\w+)', CleanText('//complete', default="")(page.doc))
if m: # had to put full url to skip redirections.
page = page.browser.open('https://www.assetmanagement.hsbc.com/feedRequest?feed_data=gfcFundData&cod=FR&client=FCPE&fId=%s&SH=%s&lId=fr' % m.groups()).page
elif "consulteroperations" not in self.page.browser.url: # not on history
url = Regexp(CleanText('//complete'), r"openUrlFichesFonds\('(.*?)',true|false\).*", default=NotAvailable)(page.doc)
if url is NotAvailable:
# redirection to a useless graphplot page with url like /portal/salarie-sg/fichefonds?idFonds=XXX&source=/portal/salarie-sg/monepargne/mesavoirs
# or on bnp, look for plot display function in a script
assert CleanText('//redirect/@url')(page.doc) or CleanText('//script[contains(text(), "afficherGraphique")]')(page.doc)
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
useless_urls = (
# pdf... http://docfinder.is.bnpparibas-ip.com/api/files/040d05b3-1776-4991-aa49-f0cd8717dab8/1536
'http://docfinder.is.bnpparibas-ip.com/',
# Redirection to a useless page with url like "https://epargne-salariale.axa-im.fr/fr/"
'https://epargne-salariale.axa-im.fr/fr/',
)
for useless_url in useless_urls:
if url.startswith(useless_url):
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
match = re.match(r'http://www.cpr-am.fr/fr/fonds_detail.php\?isin=([A-Z0-9]+)', url)
match = match or re.match(r'http://www.cpr-am.fr/particuliers/product/view/([A-Z0-9]+)', url)
if match:
self.env['code'] = match.group(1)
self.env['code_type'] = Investment.CODE_TYPE_ISIN
return
if url.startswith('http://fr.swisslife-am.com/fr/'):
self.page.browser.session.cookies.set('location', 'fr')
self.page.browser.session.cookies.set('prof', 'undefined')
page = self.page.browser.open(url).page
try:
self.env['code'] = page.get_code()
self.env['code_type'] = page.CODE_TYPE
# Handle page is None and page has not get_code method
except AttributeError:
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
def parse(self, el):
# Trying to find vdate and unitvalue
unitvalue, vdate = None, None
for span in TableCell('label')(self)[0].xpath('.//span'):
if unitvalue is None:
unitvalue = Regexp(CleanText('.'), '^([\d,]+)$', default=None)(span)
if vdate is None:
vdate = None if any(x in CleanText('./parent::div')(span) for x in ["échéance", "Maturity"]) else \
Regexp(CleanText('.'), '^([\d\/]+)$', default=None)(span)
self.env['unitvalue'] = MyDecimal().filter(unitvalue) if unitvalue else NotAvailable
self.env['vdate'] = Date(dayfirst=True).filter(vdate) if vdate else NotAvailable
self.env['_link'] = None
self.env['asset_category'] = NotAvailable
page = None
link_id = Attr(u'.//a[contains(@title, "détail du fonds")]', 'id', default=None)(self)
inv_id = Attr('.//a[contains(@id, "linkpdf")]', 'id', default=None)(self)
if link_id and inv_id:
form = self.page.get_form('//div[@id="operation"]//form')
form['idFonds'] = inv_id.split('-', 1)[-1]
form['org.richfaces.ajax.component'] = form[link_id] = link_id
page = self.page.browser.open(form['javax.faces.encodedURL'], data=dict(form)).page
if 'hsbc.fr' in self.page.browser.BASEURL:
# Special space for HSBC, does not contain any information related to performances.
m = re.search(r'fundid=(\w+).+SH=(\w+)', CleanText('//complete', default='')(page.doc))
if m: # had to put full url to skip redirections.
page = page.browser.open('https://www.assetmanagement.hsbc.com/feedRequest?feed_data=gfcFundData&cod=FR&client=FCPE&fId=%s&SH=%s&lId=fr' % m.groups()).page
elif not self.page.browser.history.is_here():
url = page.get_invest_url()
if empty(url):
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
# URLs used in browser.py to access investments performance history:
if url.startswith('https://optimisermon.epargne-retraite-entreprises'):
# This URL can be used to access the BNP Wealth API to fetch investment performance and ISIN code
self.env['_link'] = url
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
elif (url.startswith('http://sggestion-ede.com/product') or
url.startswith('https://www.lyxorfunds.com/part') or
url.startswith('https://www.societegeneralegestion.fr') or
url.startswith('http://www.etoile-gestion.com/productsheet')):
self.env['_link'] = url
# Try to fetch ISIN code from URL with re.match
match = re.match(r'http://www.cpr-am.fr/fr/fonds_detail.php\?isin=([A-Z0-9]+)', url)
match = match or re.match(r'http://www.cpr-am.fr/particuliers/product/view/([A-Z0-9]+)', url)
if match:
self.env['code'] = match.group(1)
if is_isin_valid(match.group(1)):
self.env['code_type'] = Investment.CODE_TYPE_ISIN
else:
self.env['code_type'] = Investment.CODE_TYPE_AMF
return
# Try to fetch ISIN code from URL with re.search
m = re.search(r'&ISIN=([^&]+)', url)
m = m or re.search(r'&isin=([^&]+)', url)
m = m or re.search(r'&codeIsin=([^&]+)', url)
m = m or re.search(r'lyxorfunds\.com/part/([^/]+)', url)
if m:
self.env['code'] = m.group(1)
if is_isin_valid(m.group(1)):
self.env['code_type'] = Investment.CODE_TYPE_ISIN
else:
self.env['code_type'] = Investment.CODE_TYPE_AMF
return
useless_urls = (
# pdf... http://docfinder.is.bnpparibas-ip.com/api/files/040d05b3-1776-4991-aa49-f0cd8717dab8/1536
'http://docfinder.is.bnpparibas-ip.com/',
# The AXA website displays performance graphs but everything is calculated using JS scripts.
# There is an API but it only contains risk data and performances per year, not 1-3-5 years.
'https://epargne-salariale.axa-im.fr/fr/',
# Redirection to the Rothschild Gestion website, which doesn't exist anymore...
'https://www.rothschildgestion.com',
# URL to the Morningstar website does not contain any useful information
'http://doc.morningstar.com',
)
for useless_url in useless_urls:
if url.startswith(useless_url):
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
return
if url.startswith('http://fr.swisslife-am.com/fr/'):
self.page.browser.session.cookies.set('location', 'fr')
self.page.browser.session.cookies.set('prof', 'undefined')
try:
page = self.page.browser.open(url).page
except HTTPNotFound:
# Some pages lead to a 404 so we must avoid unnecessary crash
self.logger.warning('URL %s was not found, investment details will be skipped.', url)
if isinstance(page, CodePage):
self.env['code'] = page.get_code()
self.env['code_type'] = page.CODE_TYPE
self.env['asset_category'] = page.get_asset_category()
else:
# The page is not handled and does not have a get_code method.
self.env['code'] = NotAvailable
self.env['code_type'] = NotAvailable
self.env['asset_category'] = NotAvailable
