def xt_admin_date(cls, raw_person):
# Extract administration
admin_datestring = Selector(text=raw_person).xpath(
'//td[1]/span/@title').extract()[0]
if ';' in admin_datestring:
admin_datestring = admin_datestring.split(";")[0]
if ',' in admin_datestring:
admin_datestring = admin_datestring.split(",")[0]
try:
if " - " in admin_datestring:
start_date = _clean(admin_datestring.split(' - ')[0])
end_date = _clean(admin_datestring.split(' - ')[1])
start_date = datetime.datetime.strptime(
start_date, "%d.%m.%Y").date()
end_date = datetime.datetime.strptime(
end_date, "%d.%m.%Y").date()
else:
start_date = datetime.datetime.strptime(
_clean(admin_datestring), "%d.%m.%Y").date()
end_date = None
except:
logger.error(
"Couldn't extract date from datestring {}".format(
admin_datestring))
import ipdb
ipdb.set_trace()
return (start_date, end_date)
def xt(cls, response):
ops = []
raw_ops = response.xpath(cls.XPATH).extract()
for raw_op in raw_ops[1:]:
op_sel = Selector(text=raw_op)
date = op_sel.xpath("//td[1]/text()").extract()
date = date[0]
url = op_sel.xpath("//td[2]/a/@href").extract()[0]
parl_id = u"({})".format(op_sel.xpath("//td[2]/a/text()").extract()[0])
title = op_sel.xpath("//td[3]/text()").extract()[0]
if title:
title = _clean(title).replace("*", ", ")
else:
title = None
email = op_sel.xpath("//td[3]/a/@href").extract()
if email:
email = email[0].replace("mailto:", "")
title = op_sel.xpath("//td[3]/a/text()").extract()[0]
else:
email = None
try:
date = datetime.datetime.strptime(_clean(date), "%d.%m.%Y").date()
except:
date = None
ops.append({"date": date, "url": url, "email": email, "title": title, "parl_id": parl_id})
return ops
def xt(cls, response):
ops = []
raw_ops = response.xpath(cls.XPATH).extract()
for raw_op in raw_ops[1:]:
op_sel = Selector(text=raw_op)
date = op_sel.xpath('//td[1]').xpath("normalize-space()").extract()[0]
url = op_sel.xpath('//td[2]/a/@href').extract()[0]
parl_id = u"({})".format(
op_sel.xpath('//td[3]/a').xpath('normalize-space()').extract()[0])
title = op_sel.xpath('//td[2]').xpath('normalize-space()').extract()[0]
if title:
title = _clean(title).replace("*", ", ")
else:
title = None
email = None
try:
date = datetime.datetime.strptime(
_clean(date), "%d.%m.%Y").date()
except:
date = None
ops.append({
'date': date,
'url': url,
'email': email,
'title': title,
'parl_id': parl_id
})
return ops
def xt(cls, response):
raw_signatures = response.xpath(cls.XPATH).extract()
signatures = []
for raw_signature in raw_signatures:
sig_sel = Selector(text=raw_signature)
signature_list = sig_sel.xpath('//td/text()').extract()
if len(signature_list) > 0:
full_name = _clean(signature_list[0])
if len(signature_list) > 1:
postal_code = _clean(signature_list[1])
else:
postal_code = u''
if len(signature_list) > 2:
location = _clean(signature_list[2])
else:
location = u''
if len(signature_list) > 3:
raw_date = time.strptime(_clean(signature_list[3]), '%d.%m.%Y')
date = datetime.date.fromtimestamp(time.mktime(raw_date))
else:
date = datetime.date.fromtimestamp(0)
signatures.append({
'full_name': full_name,
'postal_code': postal_code,
'location': location,
'date': date
})
return signatures
def xt(cls, step_selector):
title_selector = step_selector.xpath('//td[2]')[0]
# we have wortmeldungen!
if title_selector.xpath('//table'):
table_selector = title_selector.xpath('//table')[0]
raw_rows = [
Selector(text=raw_row) for raw_row in
table_selector.xpath('//tbody//tr').extract()
]
statements = []
# Extract statements data
for index, row_selector in enumerate(raw_rows):
try:
person_source_link = row_selector.xpath(
cls.XP_P_LINK).extract()[0]
person_name = row_selector.xpath(
cls.XP_P_NAME).extract()
statement_type = _clean(
row_selector.xpath(
cls.XP_T_TYPE).extract()[0])
protocol_link = row_selector.xpath(
cls.XP_PROT_LINK).extract()
protocol_text = _clean(
remove_tags(
row_selector.xpath(
cls.XP_PROT_TEXT).extract()[0],
'td a'))
statements.append({
'index':
index,
'person_source_link':
person_source_link,
'person_name':
person_name,
'statement_type':
statement_type,
'protocol_link':
protocol_link,
'protocol_text':
protocol_text,
})
except:
logger.error(
"Skipping statement '{}' due to extraction error"
.format(row_selector.extract()))
continue
title = {
'text': u'Wortmeldungen in der Debatte',
'statements': statements
}
else:
text = _clean(
remove_tags(
step_selector.xpath(cls.XPATH).extract()[0],
'td')).replace('<a href="',
'<a href="{}'.format(BASE_HOST))
title = {'text': text}
return title
def xt(cls, response):
bio = {
'birthdate': None,
'birthplace': '',
'deathdate': None,
'deathplace': '',
'occupation': ''
}
bio_data = response.xpath(cls.XPATH).extract()
if bio_data:
bio_data = bio_data[0]
else:
return bio
# Birth Data
for data in bio_data.split('<br>'):
birth = Selector(text=data)\
.xpath("//em[contains(text(),'Geb.')]/parent::*/text()")\
.extract()
if birth:
birth = birth[0]
bio['birthdate'] = _clean(birth.split(',')[0])
try:
bio['birthdate'] = datetime.datetime.strptime(
bio['birthdate'], "%d.%m.%Y").date()
except:
logger.error(
"Failed to parse birthdate: {}".format(
bio['birthdate']))
bio['birthdate'] = None
if len(birth.split(',')) > 1:
bio['birthplace'] = birth.split(',')[1]
# Death Data
death = Selector(text=data)\
.xpath("//em[contains(text(),'Verst.')]/parent::*/text()")\
.extract()
if death:
death = death[0]
bio['deathdate'] = _clean(death.split(',')[0])
try:
bio['deathdate'] = datetime.datetime.strptime(
bio['deathdate'], "%d.%m.%Y").date()
except:
logger.error(
"Failed to parse deathdate: {}".format(
bio['deathdate']))
bio['deathdate'] = None
if len(death.split(',')) > 1:
bio['deathplace'] = death.split(',')[1]
# Occupation
occupation = Selector(text=data)\
.xpath("//em[contains(text(),'Beruf')]/parent::*/text()")\
.extract()
if occupation:
occupation = occupation[0]
bio['occupation'] = occupation.split(',')[0]
return bio
def xt(cls, step_selector):
title_selector = step_selector.xpath('//td[2]')[0]
# we have wortmeldungen!
if title_selector.xpath('//table'):
table_selector = title_selector.xpath('//table')[0]
raw_rows = [
Selector(text=raw_row)
for raw_row
in table_selector.xpath('//tbody//tr').extract()]
statements = []
# Extract statements data
for index, row_selector in enumerate(raw_rows):
if(row_selector.xpath(cls.XP_P_LINK).extract()):
person_source_link = row_selector.xpath(
cls.XP_P_LINK).extract()[0]
else:
continue
person_name = row_selector.xpath(
cls.XP_P_NAME).extract()
if(row_selector.xpath(cls.XP_T_TYPE).extract()):
statement_type = _clean(
row_selector.xpath(cls.XP_T_TYPE).extract()[0])
else:
continue
protocol_link = row_selector.xpath(
cls.XP_PROT_LINK).extract()
if(row_selector.xpath(
cls.XP_PROT_TEXT).extract()):
protocol_text = _clean(
remove_tags(
row_selector.xpath(
cls.XP_PROT_TEXT).extract()[0],
'td a'))
else:
protocol_text = []
statements.append({
'index': index,
'person_source_link': person_source_link,
'person_name': person_name,
'statement_type': statement_type,
'protocol_link': protocol_link,
'protocol_text': protocol_text,
})
title = {
'text': u'Wortmeldungen in der Debatte',
'statements': statements
}
else:
text = _clean(
remove_tags(
step_selector.xpath(
cls.XPATH).extract()[0],
'td')).replace('<a href="', '<a href="{}'.format(BASE_HOST))
title = {'text': text}
return title
def xt(cls, response):
bio = {
'birthdate': None,
'birthplace': '',
'deathdate': None,
'deathplace': '',
'occupation': ''
}
bio_data = response.xpath(cls.XPATH).extract()
if bio_data:
bio_data = bio_data[0]
else:
return bio
# Birth Data
for data in bio_data.split('<br>'):
birth = Selector(text=data)\
.xpath("//em[contains(text(),'Geb.')]/parent::*/text()")\
.extract()
if birth:
birth = birth[0]
bio['birthdate'] = _clean(birth.split(',')[0])
try:
bio['birthdate'] = datetime.datetime.strptime(
bio['birthdate'], "%d.%m.%Y").date()
except:
logger.error("Failed to parse birthdate: {}".format(
bio['birthdate']))
bio['birthdate'] = None
if len(birth.split(',')) > 1:
bio['birthplace'] = birth.split(',')[1]
# Death Data
death = Selector(text=data)\
.xpath("//em[contains(text(),'Verst.')]/parent::*/text()")\
.extract()
if death:
death = death[0]
bio['deathdate'] = _clean(death.split(',')[0])
try:
bio['deathdate'] = datetime.datetime.strptime(
bio['deathdate'], "%d.%m.%Y").date()
except:
logger.error("Failed to parse deathdate: {}".format(
bio['deathdate']))
bio['deathdate'] = None
if len(death.split(',')) > 1:
bio['deathplace'] = death.split(',')[1]
# Occupation
occupation = Selector(text=data)\
.xpath("//em[contains(text(),'Beruf')]/parent::*/text()")\
.extract()
if occupation:
occupation = occupation[0]
bio['occupation'] = occupation.split(',')[0]
return bio
def xt(cls, response):
persons = []
raw_persons = response.xpath(cls.XPATH).extract()
for raw_person in raw_persons:
person = Selector(text=raw_person)
if person.xpath('//th'):
continue
source_link = person.xpath(
'//td//a/@href').extract()[0]
reversed_name = _clean(
Selector(
text=remove_tags(raw_person, 'img')
).xpath('//td//a/text()').extract()[0])
if ' siehe ' in reversed_name:
reversed_name = reversed_name.split(' siehe ')[1]
admin_title = person.xpath(
'//td[1]/span/text()').extract()
(admin_start_date, admin_end_date) = cls.xt_admin_date(
raw_person)
administration = {
'title': admin_title,
'start_date': admin_start_date,
'end_date': admin_end_date
}
# TODO EXTRACT DATE(S) FROM BUNDESMINISTERIUM td
# TODO ADD EITHER DATE(S) TO FUNCTION
try:
if person.xpath('//tr//td[3]/span/text()'):
function_short = person.xpath(
'//td[3]/span/text()').extract()[0]
function_title = person.xpath(
'//td[3]/span/@title').extract()[0]
elif person.xpath('//tr//td[3]/text()'):
function_short = _clean(person.xpath(
'//td[3]/text()').extract()[0])
function_title = ''
except:
import ipdb
ipdb.set_trace()
mandate = {
'short': function_short,
'title': function_title,
'administration': administration}
persons.append({
'source_link': source_link,
'reversed_name': reversed_name,
'mandate': mandate,
})
return persons
def xt(cls, response):
mandates_raw = response.xpath(cls.XPATH).extract()
mandates = []
for mandate in mandates_raw:
mandate = _clean(remove_tags(mandate, 'li'))
if "<div" in mandate and "</div>" in mandate:
mandate = _clean(
remove_tags(
Selector(
text=mandate).xpath("//div").extract()[0],
'div'))
function = mandate.split(u'<br>')[0].split(',')[0]
party = mandate.split(u'<br>')[0].split(',')[1]
# Start Date
try:
start_date = _clean(
mandate.split('<br>')[1].split(u'\u2013')[0])
start_date = datetime.datetime.strptime(
start_date, "%d.%m.%Y").date()
except:
logger.error(
u"Failed to parse mandate start date: {}".format(
start_date))
start_date = None
# End Date
try:
end_date = mandate.split('<br>')[1].split(u'\u2013')
if len(end_date) > 1 and end_date[1]:
end_date = datetime.datetime.strptime(
_clean(end_date[1]), "%d.%m.%Y").date()
else:
end_date = None
except:
logger.error(
u"Failed to parse mandate end date: {}".format(
end_date))
end_date = None
mandates.append({
'function': function,
'party': _clean(party),
'start_date': start_date,
'end_date': end_date,
})
return mandates
def xt(cls, response):
mandates_raw = response.xpath(cls.XPATH).extract()
mandates = []
for mandate in mandates_raw:
mandate = _clean(remove_tags(mandate, 'li'))
if "<div" in mandate and "</div>" in mandate:
mandate = _clean(remove_tags(
Selector(text=mandate).xpath("//div").extract()[0],
'div'))
function = mandate.split(u'<br>')[0].split(',')[0]
party = mandate.split(u'<br>')[0].split(',')[1]
# Start Date
try:
start_date = _clean(
mandate.split('<br>')[1].split(u'\u2013')[0])
start_date = datetime.datetime.strptime(
start_date, "%d.%m.%Y").date()
except:
logger.error(
u"Failed to parse mandate start date: {}".format(start_date))
start_date = None
# End Date
try:
end_date = mandate.split(
'<br>')[1].split(u'\u2013')
if len(end_date) > 1 and end_date[1]:
end_date = datetime.datetime.strptime(
_clean(end_date[1]), "%d.%m.%Y").date()
else:
end_date = None
except:
logger.error(
u"Failed to parse mandate end date: {}".format(end_date))
end_date = None
mandates.append({
'function': function,
'party': _clean(party),
'start_date': start_date,
'end_date': end_date,
})
return mandates
def xt(cls, response):
XPATH_BI_creator = cls.XPATH.format("Erstunterzeichner")
XPATH_PET_creator = cls.XPATH.format("eine Petition")
creators = []
raw_creators_list = response.xpath(XPATH_PET_creator).extract()
if len(raw_creators_list) > 0:
# PET started by members of parliament
for raw_creator in raw_creators_list:
creator_sel = Selector(text=raw_creator)
raw_parl_id_url = creator_sel.xpath("//a/@href").extract()
name = u''
parl_id = u''
if len(raw_parl_id_url) > 0:
raw_parl_id = raw_parl_id_url[0].split("/")
if len(raw_parl_id) > 1:
parl_id = raw_parl_id[2]
raw_name = creator_sel.xpath("//a/text()").extract()
if len(raw_name) > 0:
name = raw_name[0]
if parl_id != u'' and name != u'':
creators.append((parl_id, name))
else:
raw_creators_list = response.xpath(XPATH_BI_creator).extract()
if len(raw_creators_list) > 0:
# BI first signed by a person
name = _clean(raw_creators_list[0].split("\t")[1])
creators.append(("", name))
# VBG seem to have no visible "starter"
return creators
def xt(cls, response):
persons = []
raw_persons = response.xpath(cls.XPATH).extract()
for raw_person in raw_persons:
source_link = Selector(text=raw_person).xpath(
'//td//a/@href').extract()[0]
reversed_name = _clean(
Selector(text=raw_person).xpath('//td//a/text()').extract()[0])
if ' siehe ' in reversed_name:
reversed_name = reversed_name.split(' siehe ')[1]
mandates = []
party_spans = Selector(text=raw_person).xpath(
'//td[2]//span').extract()
for party_span in party_spans:
party_short = Selector(text=party_span).xpath(
'//span/text()').extract()[0]
party_title = Selector(text=party_span).xpath(
'//span/@title').extract()[0]
mandates.append(
{'short': party_short, 'title': party_title})
electoral_state = {
'short': Selector(text=raw_person).xpath('//td[last()]//span/text()').extract()[0],
'long': Selector(text=raw_person).xpath('//td[last()]//span/@title').extract()[0]}
persons.append({
'source_link': source_link,
'reversed_name': reversed_name,
'mandates': mandates,
'electoral_state': electoral_state,
})
return persons
def xt(cls, response):
persons = []
raw_persons = response.xpath(cls.XPATH).extract()
for raw_person in raw_persons:
person = Selector(text=raw_person)
if person.xpath('//th'):
continue
source_link = person.xpath(
'//td//a/@href').extract()[0]
reversed_name = _clean(
Selector(
text=remove_tags(raw_person, 'img')
).xpath('//td//a/text()').extract()[0])
(pres_start_date, pres_end_date) = cls.xt_pres_date(
raw_person)
mandate = {
'title': u'RechnungshofpräsidentIn',
'short': u'RH-PräsidentIn',
'start_date': pres_start_date,
'end_date': pres_end_date
}
persons.append({
'source_link': source_link,
'reversed_name': reversed_name,
'mandate': mandate,
})
return persons
def xt(cls, response):
XPATH_BI_creator = cls.XPATH.format("Erstunterzeichner")
XPATH_PET_creator = cls.XPATH.format("eine Petition")
creators = []
raw_creators_list = response.xpath(XPATH_PET_creator).extract()
if len(raw_creators_list) > 0:
# PET started by members of parliament
for raw_creator in raw_creators_list:
creator_sel = Selector(text=raw_creator)
raw_parl_id_url = creator_sel.xpath("//a/@href").extract()
name = u''
parl_id = u''
if len(raw_parl_id_url) > 0:
raw_parl_id = raw_parl_id_url[0].split("/")
if len(raw_parl_id) > 1:
parl_id = raw_parl_id[2]
raw_name = creator_sel.xpath("//a/text()").extract()
if len(raw_name) > 0:
name = raw_name[0]
if parl_id != u'' and name != u'':
creators.append((parl_id, name))
else:
raw_creators_list = response.xpath(XPATH_BI_creator).extract()
if len(raw_creators_list) > 0:
# BI first signed by a person
name = _clean(raw_creators_list[0].split("\t")[1])
creators.append(("", name))
# VBG seem to have no visible "starter"
return creators
def xt(cls, response):
persons = []
raw_persons = response.xpath(cls.XPATH).extract()
for raw_person in raw_persons:
source_link = Selector(text=raw_person).xpath(
'//td//a/@href').extract()[0]
reversed_name = _clean(
Selector(text=raw_person).xpath('//td//a/text()').extract()[0])
if ' siehe ' in reversed_name:
reversed_name = reversed_name.split(' siehe ')[1]
mandates = []
party_spans = Selector(text=raw_person).xpath(
'//td[2]//span').extract()
for party_span in party_spans:
party_short = Selector(text=party_span).xpath(
'//span/text()').extract()[0]
party_title = Selector(text=party_span).xpath(
'//span/@title').extract()[0]
mandates.append(
{'short': party_short, 'title': party_title})
electoral_state = {
'short': Selector(text=raw_person).xpath('//td[4]//span/text()').extract()[0],
'long': Selector(text=raw_person).xpath('//td[4]//span/@title').extract()[0]}
persons.append({
'source_link': source_link,
'reversed_name': reversed_name,
'mandates': mandates,
'electoral_state': electoral_state,
})
return persons
def xt(cls, step_selector):
title_selector = step_selector.xpath("//td[2]")[0]
# we have wortmeldungen!
if title_selector.xpath("//table"):
table_selector = title_selector.xpath("//table")[0]
raw_rows = [Selector(text=raw_row) for raw_row in table_selector.xpath("//tbody//tr").extract()]
statements = []
# Extract statements data
for index, row_selector in enumerate(raw_rows):
if row_selector.xpath(cls.XP_P_LINK).extract():
person_source_link = row_selector.xpath(cls.XP_P_LINK).extract()[0]
else:
continue
person_name = row_selector.xpath(cls.XP_P_NAME).extract()
if row_selector.xpath(cls.XP_T_TYPE).extract():
statement_type = _clean(row_selector.xpath(cls.XP_T_TYPE).extract()[0])
else:
continue
protocol_link = row_selector.xpath(cls.XP_PROT_LINK).extract()
if row_selector.xpath(cls.XP_PROT_TEXT).extract():
protocol_text = _clean(
remove_tags(row_selector.xpath(cls.XP_PROT_TEXT).extract()[0], "td a")
)
else:
protocol_text = []
statements.append(
{
"index": index,
"person_source_link": person_source_link,
"person_name": person_name,
"statement_type": statement_type,
"protocol_link": protocol_link,
"protocol_text": protocol_text,
}
)
title = {"text": u"Wortmeldungen in der Debatte", "statements": statements}
else:
text = _clean(remove_tags(step_selector.xpath(cls.XPATH).extract()[0], "td")).replace(
'<a href="', '<a href="{}'.format(BASE_HOST)
)
title = {"text": text}
return title
def xt(cls, response):
rows = response.xpath(cls.XPATH)
for row in rows:
raw_active = row.xpath('td[2]/text()').extract()
if len(raw_active) > 0:
active = _clean(raw_active[0])
if active == u'Aufl\xf6sung':
return False
return True
def xt(cls, response):
rows = response.xpath(cls.XPATH)
for row in rows:
raw_active = row.xpath('td[2]/text()').extract()
if len(raw_active) > 0:
active = _clean(raw_active[0])
if active == u'Aufl\xf6sung':
return False
return True
def xt(cls, response):
raw_signatures = response.xpath(cls.XPATH).extract()
signatures = []
for raw_signature in raw_signatures:
sig_sel = Selector(text=raw_signature)
signature_list = sig_sel.xpath('//td/text()').extract()
if len(signature_list) > 0:
full_name = _clean(signature_list[0])
if len(signature_list) > 1:
postal_code = _clean(signature_list[1])
else:
postal_code = u''
if len(signature_list) > 2:
location = _clean(signature_list[2])
else:
location = u''
if len(signature_list) > 3:
raw_date = time.strptime(_clean(signature_list[3]),
'%d.%m.%Y')
date = datetime.date.fromtimestamp(
time.mktime(raw_date))
else:
date = datetime.date.fromtimestamp(0)
signatures.append({
'full_name': full_name,
'postal_code': postal_code,
'location': location,
'date': date
})
return signatures
def xt(cls, response):
raw_laws = response.xpath(cls.XPATH_LAWS)
raw_reports = response.xpath(cls.XPATH_REPORTS)
raw_laws = raw_laws + raw_reports
laws = []
for raw_law in raw_laws:
raw_title = raw_law.xpath('text()').extract()
if len(raw_title) > 0:
law_title = _clean(raw_title[0])
else:
law_title = u''
raw_link = raw_law.xpath('@href').extract()
if len(raw_link) > 0:
law_link = raw_link[0]
law_llp, law_parl_id = COMITTEE.url_to_parlid(law_link)
law_link = "{}/{}".format(BASE_HOST, law_link)
else:
# without a link we can't get the necessary info
continue
if law_llp != u'' and law_parl_id != u'':
law = {
'title': law_title,
'source_link': law_link,
'parl_id': law_parl_id,
'llp': law_llp,
}
laws.append(law)
return laws
def xt(cls, response):
raw_laws = response.xpath(cls.XPATH_LAWS)
raw_reports = response.xpath(cls.XPATH_REPORTS)
raw_laws = raw_laws + raw_reports
laws = []
for raw_law in raw_laws:
raw_title = raw_law.xpath('text()').extract()
if len(raw_title) > 0:
law_title = _clean(raw_title[0])
else:
law_title = u''
raw_link = raw_law.xpath('@href').extract()
if len(raw_link) > 0:
law_link = raw_link[0]
law_llp, law_parl_id = COMITTEE.url_to_parlid(law_link)
law_link = "{}/{}".format(BASE_HOST, law_link)
else:
# without a link we can't get the necessary info
continue
if law_llp != u'' and law_parl_id != u'':
law = {
'title': law_title,
'source_link': law_link,
'parl_id': law_parl_id,
'llp': law_llp,
}
laws.append(law)
return laws
def xt(cls, response):
mandates_raw = response.xpath(cls.XPATH).extract()
mandates = []
for mandate in mandates_raw:
mandate = _clean(remove_tags(mandate, 'li'))
if "<div" in mandate and "</div>" in mandate:
mandate = _clean(remove_tags(
Selector(text=mandate).xpath("//div").extract()[0],
'div'))
function = mandate.split(u'<br>')[0].split(',')[0]
party = mandate.split(u'<br>')[0].split(',')[1] if ',' in mandate.split(u'<br />')[0] else ''
llp_raw = re.match(
'^.*\((.*)\. GP\).*$', function
)
function = re.sub(
'\((.*)\. GP\)','', function
).strip()
m_llp_roman_begin = \
m_llp_roman_end = \
llp_raw.group(1) if llp_raw else ''
if u'–' in m_llp_roman_begin:
m_llp_roman_begin,m_llp_roman_end = m_llp_roman_begin.split(u'–')
for llp in range(roman.fromRoman(m_llp_roman_begin.strip('. ')),
roman.fromRoman(m_llp_roman_end.strip('. '))+1
) if m_llp_roman_begin else [None]:
llp_roman = roman.toRoman(llp) if llp else None
# Start Date
try:
start_date = _clean(
mandate.split('<br>')[1].split(u'\u2013')[0])
start_date = datetime.datetime.strptime(
start_date, "%d.%m.%Y").date()
except:
logger.error(
u"Failed to parse mandate start date: {}".format(start_date))
start_date = None
# End Date
try:
end_date = mandate.split(
'<br>')[1].split(u'\u2013')
if len(end_date) > 1 and end_date[1]:
end_date = datetime.datetime.strptime(
_clean(end_date[1]), "%d.%m.%Y").date()
else:
end_date = None
except:
logger.error(
u"Failed to parse mandate end date: {}".format(end_date))
end_date = None
mandates.append({
'function': function,
'party': _clean(party),
'start_date': start_date,
'end_date': end_date,
'llp': llp,
'llp_roman': llp_roman,
})
return mandates
def xt(cls, response):
raw_memberships = response.xpath(cls.XPATH)
memberships = []
for raw_membership in raw_memberships:
raw_llp = raw_membership.xpath('a[1]/text()').extract()[1]
nrbr = u'Nationalrat'
comittee_llp = None
if nrbr in raw_llp:
comittee_llp = raw_llp.split()[-2][:-1]
else:
nrbr = u'Bundesrat'
tablerows = raw_membership.xpath('following-sibling::div[1]/table[1]/tbody/tr').extract()
last_function = u''
for row in tablerows:
row_sel = Selector(text=row)
raw_function = row_sel.xpath('//td[@class="biogr_am_funktext"]/text()').extract()
if len(raw_function) > 0:
function = _clean(raw_function[0])
# TODO: standardization of functions should be done on model level
last_function = function
else:
function = last_function
raw_comittee_link = row_sel.xpath('//td[@class="biogr_am_ausschuss"]/a/@href').extract()
if raw_comittee_link:
comittee_link = raw_comittee_link[0]
comittee_link = "{}/{}".format(BASE_HOST, comittee_link)
else:
comittee_link = u''
_,comittee_parl_id = COMITTEE.url_to_parlid(comittee_link)
raw_comitee_name = row_sel.xpath('//td[@class="biogr_am_ausschuss"]/a/text()').extract()
if len(raw_comitee_name) > 0:
comittee_name = _clean(raw_comitee_name[0])
else:
raw_comitee_name = row_sel.xpath('//td[@class="biogr_am_ausschuss"]/text()').extract()
if len(raw_comitee_name) > 0:
comittee_name = _clean(raw_comitee_name[0])
else:
comittee_name = u''
raw_dates = row_sel.xpath('//td[@class="biogr_am_vonbis"]/text()').extract()[0]
if raw_dates:
raw_dates = _clean(raw_dates)
# \u2013 == - (dash)
raw_dates = raw_dates.split(u'\u2013')
if len(raw_dates) > 0:
raw_from = raw_dates[0]
if raw_from is not u'':
raw_from = time.strptime(raw_from, '%d.%m.%Y')
date_from = datetime.datetime.fromtimestamp(time.mktime(raw_from))
else:
date_from = None
else:
date_from = None
if len(raw_dates) > 1:
raw_to = raw_dates[1]
if raw_to is not u'':
raw_to = time.strptime(raw_to, '%d.%m.%Y')
date_to = datetime.datetime.fromtimestamp(time.mktime(raw_to))
else:
date_to = None
else:
date_to = None
# we cant add the membership if the parl_id of the comitee is empty
if comittee_parl_id is not u'':
memberships.append({
'comittee':
{
'name': comittee_name,
'parl_id': comittee_parl_id,
'nrbr': nrbr,
'legislative_period': comittee_llp,
'source_link': comittee_link
},
'function': function,
'date_from': date_from,
'date_to': date_to
})
return memberships
def xt(cls, response):
raw_memberships = response.xpath(cls.XPATH)
memberships = []
for raw_membership in raw_memberships:
raw_llp = raw_membership.xpath('a[1]/text()').extract()[1]
nrbr = u'Nationalrat'
comittee_llp = None
if nrbr in raw_llp:
comittee_llp = raw_llp.split()[-2][:-1]
else:
nrbr = u'Bundesrat'
tablerows = raw_membership.xpath(
'following-sibling::div[1]/table[1]/tbody/tr').extract()
last_function = u''
for row in tablerows:
row_sel = Selector(text=row)
raw_function = row_sel.xpath(
'//td[@class="biogr_am_funktext"]/text()').extract()
if len(raw_function) > 0:
function = _clean(raw_function[0])
# TODO: standardization of functions should be done on model level
last_function = function
else:
function = last_function
raw_comittee_link = row_sel.xpath(
'//td[@class="biogr_am_ausschuss"]/a/@href').extract()
if raw_comittee_link:
comittee_link = raw_comittee_link[0]
comittee_link = "{}{}".format(BASE_HOST, comittee_link)
else:
comittee_link = u''
_, comittee_parl_id = COMITTEE.url_to_parlid(comittee_link)
raw_comitee_name = row_sel.xpath(
'//td[@class="biogr_am_ausschuss"]/a/text()').extract(
)
if len(raw_comitee_name) > 0:
comittee_name = _clean(raw_comitee_name[0])
else:
raw_comitee_name = row_sel.xpath(
'//td[@class="biogr_am_ausschuss"]/text()'
).extract()
if len(raw_comitee_name) > 0:
comittee_name = _clean(raw_comitee_name[0])
else:
comittee_name = u''
raw_dates = row_sel.xpath(
'//td[@class="biogr_am_vonbis"]/text()').extract()[0]
if raw_dates:
raw_dates = _clean(raw_dates)
# \u2013 == - (dash)
raw_dates = raw_dates.split(u'\u2013')
if len(raw_dates) > 0:
raw_from = raw_dates[0]
if raw_from is not u'':
raw_from = time.strptime(raw_from, '%d.%m.%Y')
date_from = datetime.datetime.fromtimestamp(
time.mktime(raw_from))
else:
date_from = None
else:
date_from = None
if len(raw_dates) > 1:
raw_to = raw_dates[1]
if raw_to is not u'':
raw_to = time.strptime(raw_to, '%d.%m.%Y')
date_to = datetime.datetime.fromtimestamp(
time.mktime(raw_to))
else:
date_to = None
else:
date_to = None
# we cant add the membership if the parl_id of the comitee is empty
if comittee_parl_id is not u'':
memberships.append({
'comittee': {
'name': comittee_name,
'parl_id': comittee_parl_id,
'nrbr': nrbr,
'legislative_period': comittee_llp,
'source_link': comittee_link
},
'function': function,
'date_from': date_from,
'date_to': date_to
})
return memberships
def xt(cls, response):
raw_meetings = response.xpath(cls.XPATH)
meetings = []
for raw_meeting in raw_meetings:
raw_header_row = raw_meeting.xpath('tr[@class="historyHeader"]')
raw_date = raw_header_row.xpath('td[1]/text()').extract()
if len(raw_date) > 0:
raw_date = _clean(raw_date[0])
if raw_date is not u'':
raw_date = time.strptime(raw_date, '%d.%m.%Y')
meeting_date = datetime.datetime.fromtimestamp(time.mktime(raw_date))
else:
meeting_date = None
else:
meeting_date = None
raw_number = raw_header_row.xpath('td[2]/em/a/text()').extract()
if len(raw_number) > 0 and u'Sitzung' in raw_number[0]:
meeting_number = raw_number[0].split()[0][:-1]
else:
continue # not a meeting
raw_document_urls = raw_header_row.xpath('td[2]/a/@href').extract()
html_link, pdf_link = u"", u""
for url in raw_document_urls:
if url.endswith('.pdf'):
pdf_link = url
if not pdf_link.startswith(BASE_HOST):
pdf_link = "{}/{}".format(BASE_HOST, pdf_link)
elif url.endswith('.html'):
html_link = url
if not html_link.startswith(BASE_HOST):
html_link = "{}/{}".format(BASE_HOST, html_link)
title = u'Tagesordnung der {}. Sitzung des {} am {}'\
.format(meeting_number, COMITTEE.NAME.xt(response), str(meeting_date.date()))
if html_link != u'' or pdf_link != u'':
meeting_document = {
'title': title,
'html_link': html_link,
'pdf_link': pdf_link
}
else:
meeting_document = None
raw_rows = raw_header_row.xpath('following-sibling::tr')
meeting_topics = []
for raw_row in raw_rows:
raw_topic_number = raw_row.xpath('td[1]/text()').extract()
if len(raw_topic_number) > 0:
topic_number_list = _clean(raw_topic_number[0]).split()
if len(topic_number_list) == 2 and topic_number_list[0] == u'TOP':
topic_number = int(topic_number_list[1])
else:
topic_number = 0
else:
topic_number = 0
raw_topic_text = raw_row.xpath('td[2]/text()').extract()
if len(raw_topic_text) > 0:
topic_text = _clean(raw_topic_text[0])
if topic_text.endswith('('):
topic_text = topic_text[:-1].rstrip()
else:
topic_text = u''
if len(raw_topic_text) > 1:
topic_comment = _clean(raw_topic_text[1])
if topic_comment.startswith(')'):
topic_comment = topic_comment[:-1].lstrip()
else:
topic_comment = u''
raw_topic_law_text = raw_row.xpath('td[2]/a/text()').extract()
if len(raw_topic_law_text) > 0:
topic_law_text = u'({})'.format(raw_topic_law_text[0])
else:
topic_law_text = u''
topic_text = u'{} {}'.format(topic_text,topic_law_text)
raw_topic_law_link = raw_row.xpath('td[2]/a/@href').extract()
if len(raw_topic_law_link) > 0:
topic_law_llp, topic_law_id = COMITTEE.url_to_parlid(raw_topic_law_link[0])
else:
topic_law_llp, topic_law_id = u'', u''
if topic_law_id != u'':
topic_law = {
'parl_id': topic_law_id,
'llp': topic_law_llp
}
else:
topic_law = None
topic = {
'number': topic_number,
'text': topic_text,
'comment': topic_comment,
'law': topic_law
}
meeting_topics.append(topic)
meeting = {
'number': meeting_number,
'date': meeting_date,
'agenda': meeting_document,
'topics': meeting_topics,
}
meetings.append(meeting)
return meetings
def xt(cls, response):
raw_meetings = response.xpath(cls.XPATH)
meetings = []
for raw_meeting in raw_meetings:
raw_header_row = raw_meeting.xpath(
'tr[@class="historyHeader"]')
raw_date = raw_header_row.xpath('td[1]/text()').extract()
if len(raw_date) > 0:
raw_date = _clean(raw_date[0])
if raw_date is not u'':
raw_date = time.strptime(raw_date, '%d.%m.%Y')
meeting_date = datetime.datetime.fromtimestamp(
time.mktime(raw_date))
else:
meeting_date = None
else:
meeting_date = None
raw_number = raw_header_row.xpath(
'td[2]/em/a/text()').extract()
if len(raw_number) > 0 and u'Sitzung' in raw_number[0]:
meeting_number = raw_number[0].split()[0][:-1]
else:
continue # not a meeting
raw_document_urls = raw_header_row.xpath(
'td[2]/a/@href').extract()
html_link, pdf_link = u"", u""
for url in raw_document_urls:
if url.endswith('.pdf'):
pdf_link = url
if not pdf_link.startswith(BASE_HOST):
pdf_link = "{}/{}".format(BASE_HOST, pdf_link)
elif url.endswith('.html'):
html_link = url
if not html_link.startswith(BASE_HOST):
html_link = "{}/{}".format(BASE_HOST, html_link)
title = u'Tagesordnung der {}. Sitzung des {} am {}'\
.format(meeting_number, COMITTEE.NAME.xt(response), str(meeting_date.date()))
if html_link != u'' or pdf_link != u'':
meeting_document = {
'title': title,
'html_link': html_link,
'pdf_link': pdf_link
}
else:
meeting_document = None
raw_rows = raw_header_row.xpath('following-sibling::tr')
meeting_topics = []
for raw_row in raw_rows:
raw_topic_number = raw_row.xpath('td[1]/text()').extract()
if len(raw_topic_number) > 0:
topic_number_list = _clean(raw_topic_number[0]).split()
if len(topic_number_list
) == 2 and topic_number_list[0] == u'TOP':
topic_number = int(topic_number_list[1])
else:
topic_number = 0
else:
topic_number = 0
raw_topic_text = raw_row.xpath('td[2]/text()').extract()
if len(raw_topic_text) > 0:
topic_text = _clean(raw_topic_text[0])
if topic_text.endswith('('):
topic_text = topic_text[:-1].rstrip()
else:
topic_text = u''
if len(raw_topic_text) > 1:
topic_comment = _clean(raw_topic_text[1])
if topic_comment.startswith(')'):
topic_comment = topic_comment[:-1].lstrip()
else:
topic_comment = u''
raw_topic_law_text = raw_row.xpath(
'td[2]/a/text()').extract()
if len(raw_topic_law_text) > 0:
topic_law_text = u'({})'.format(raw_topic_law_text[0])
else:
topic_law_text = u''
topic_text = u'{} {}'.format(topic_text, topic_law_text)
raw_topic_law_link = raw_row.xpath(
'td[2]/a/@href').extract()
if len(raw_topic_law_link) > 0:
topic_law_llp, topic_law_id = COMITTEE.url_to_parlid(
raw_topic_law_link[0])
else:
topic_law_llp, topic_law_id = u'', u''
if topic_law_id != u'':
topic_law = {
'parl_id': topic_law_id,
'llp': topic_law_llp
}
else:
topic_law = None
topic = {
'number': topic_number,
'text': topic_text,
'comment': topic_comment,
'law': topic_law
}
meeting_topics.append(topic)
meeting = {
'number': meeting_number,
'date': meeting_date,
'agenda': meeting_document,
'topics': meeting_topics,
}
meetings.append(meeting)
return meetings
