def get_id(self, collection, identifier, type=None):
if collection != "organizations":
existing = vpapi.getfirst(collection, where={'identifiers': {'$elemMatch': {'identifier': identifier}}})
else:
if type:
existing = vpapi.getfirst(collection, where={'identifiers': {'$elemMatch': {'identifier': identifier}}})
else:
existing = vpapi.getfirst(collection, where={'name': identifier})
if existing:
p_id = existing['id']
else:
p_id = "Not found"
return p_id
def scrape_membership(self):
# Returns chambers membership list with the basic information data
# for each member of every chamber for Moldova's parliament.
chamber_membership = []
print "\n\tScraping chambers membership from Moldova's parliament..."
mps_list = self.mps_list()
members = {}
membership_correction = self.membership_correction()
all_members = vpapi.getall("people")
for member in all_members:
members[member['identifiers'][0]['identifier']] = member['id']
chamber_id = vpapi.getfirst("organizations",
where={"identifiers": {
"$elemMatch": {
"identifier": "20", "scheme": "parlament.md"
}
}})
deputy_list_url = "http://www.parlament.md/StructuraParlamentului/" \
"Deputies/tabid/87/language/ro-RO/Default.aspx"
for member in mps_list:
p_id = members[member['identifier']]
role = membership_correction[member['membership'].encode('utf-8')]
chamber_membership_json = self.build_memberships_doc(p_id, chamber_id['id'], member['membership'],
role, deputy_list_url)
chamber_membership.append(chamber_membership_json)
print "\n\tScraping completed! \n\tScraped " + str(len(chamber_membership)) + " members of chambers \n"
return chamber_membership
def scrape_chamber(self):
# Iterates in every parliamentary group json document and
# returns the list with the json document structure that Visegrad+ API accepts
print "\n\tScraping chambers from Belarus Lowerhouse parliament..."
chambers = parser.chambers()
chambers_list = []
url = "http://house.gov.by/index.php/,10087,,,,2,,,0.html"
for chamber in chambers:
chamber_json = self.build_organization_doc("chamber", chambers[chamber]['name'], chamber,
chambers[chamber]['start_date'], chambers[chamber]['end_date'],
url, "", "")
if chamber == "2":
del chamber_json['dissolution_date']
del chamber_json['contact_details']
del chamber_json['parent_id']
existing = vpapi.getfirst("organizations", where={'identifiers': {'$elemMatch': chamber_json['identifiers'][0]}})
if not existing:
resp = vpapi.post("organizations", chamber_json)
else:
# update by PUT is preferred over PATCH to correctly remove properties that no longer exist now
resp = vpapi.put("organizations", existing['id'], chamber_json, effective_date=self.effective_date())
if resp["_status"] != "OK":
raise Exception("Invalid status code")
chambers_list.append(chamber_json)
print "\n\tScraping completed! \n\tScraped " + str(len(chambers_list)) + " chambers"
return chambers_list
def scrape_chamber(self):
# Scrapes chambers and Returns the list of chambers with all the information needed for each
url = "http://www.parliament.am/deputies.php?sel=ful&ord=photo&show_session=5&lang=arm&enc=utf8"
soup = scrape.download_html_file(url)
chambers_list = []
print "\n\tScraping chambers from Armenia's parliament...\n"
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
all_options = soup.find("select", {"name": "show_session"}).findAll("option")
for each_option in pbar(all_options):
identifier = each_option.get('value')
name = each_option.get_text()
url = "http://www.parliament.am/deputies.php?lang=arm&sel=&ord=&show_session=" + identifier
if "100" not in identifier:
founding_date = self.terms[identifier]["start_date"]
dissolution_date = self.terms[identifier]["end_date"]
chamber_json = self.build_organization_doc("chamber", name, identifier, founding_date,
dissolution_date, url, "", "")
del chamber_json['contact_details']
del chamber_json['parent_id']
if identifier == "5":
del chamber_json['dissolution_date']
existing = vpapi.getfirst("organizations", where={'identifiers': {'$elemMatch': chamber_json['identifiers'][0]}})
if not existing:
resp = vpapi.post("organizations", chamber_json)
else:
resp = vpapi.put("organizations", existing['id'], chamber_json, effective_date=self.effective_date())
if resp["_status"] != "OK":
raise Exception("Invalid status code")
chambers_list.append(chamber_json)
print "\n\tScraping completed! \n\tScraped " + str(len(chambers_list)) + " chambers"
return chambers_list
def scrape_parliamentary_group_membership(self):
# Returns parliamentary groups membership list with the basic information data
# for each member of every parliamentary group for Armenia's parliament.
print "\n\tScraping parliamentary groups membership from Armenia's parliament...\n"
chambers = {}
groups = {}
members = {}
memberships = self.membership_correction()
all_chambers = vpapi.getall("organizations", where={"classification": "chamber"})
for chamber in all_chambers:
chambers[chamber['identifiers'][0]["identifier"]] = chamber['id']
all_groups = vpapi.getall('organizations', where={"classification": "parliamentary group"})
for group in all_groups:
groups[group['sources'][0]['url']] = group['id']
all_members = vpapi.getall("people")
for member in all_members:
members[member['name']] = member['id']
parties_membership = []
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
for term in pbar(list(reversed(sorted(self.terms.keys())))):
url = "http://www.parliament.am/deputies.php?lang=arm&sel=factions&SubscribeEmail=&show_session=" + str(term)
soup = scrape.download_html_file(url)
for each_div in soup.findAll('div', {"class": "content"}):
party_name = each_div.find("center").find("b").get_text()
party_name_ordered = party_name.replace(" ", " ")
exist = vpapi.getfirst("organizations", where={'name': party_name_ordered,
"parent_id": chambers[str(term)]})
if exist:
o_id = exist['id']
for each_tr in each_div.find('table', {"style": "margin-top:10px; margin-bottom:10px;"}).findAll('tr'):
if each_tr.has_attr('bgcolor'):
continue
else:
td_array = each_tr.findAll('td')
names = td_array[0].find('a').get_text().split(' ')
first_name = names[1]
last_name = names[0]
middle_name = names[2]
name_ordered = "%s %s %s" % (first_name, middle_name, last_name)
membership = each_tr.find('span', {'class': "news_date"}).get_text()
if membership == "":
membership = "անդամ".decode('utf-8')
else:
membership = membership[1:len(membership)-1]
role = memberships[membership.encode('utf-8')]
if name_ordered in members:
p_id = members[name_ordered]
party_membership_json = self.build_memberships_doc(p_id, o_id, membership, role, url)
parties_membership.append(party_membership_json)
print "\n\tScraping completed! \n\tScraped " + str(len(parties_membership)) + " members of parliamentary groups"
return parties_membership
def get_index(self, collection, sort, vote_events):
last_event = vpapi.getfirst(collection, sort=sort)
if last_event:
index = (
next(index for (index, d) in enumerate(vote_events) if d["identifier"] == last_event["identifier"]) + 1
)
else:
index = 0
return index
def extract_person_idendifier(text):
match = re.findall('/sqw/detail.sqw\?id=([0-9]{1,})', text)
if match:
ex = vpapi.getfirst('people', where={'id': match[0]})
if ex:
return match[0]
else:
return None
else:
return None
def scrape_chamber_changes_and_save(term):
"""Scrape list of changes of memberships in the parliament chamber
and save (or update) the respective memberships.
If an MP referred by the membership does not exist, scrape and save him/her.
"""
change_list = parse.change_list(term)
oid = get_chamber_id(term)
for change in reversed(change_list['_items']):
logging.info('Scraping mandate change of `%s` at %s' % (change['poslanec']['meno'], change['dátum']))
# if MP is not scraped yet, scrape and save him
existing = vpapi.getfirst('people',
where={'identifiers': {'$elemMatch': {'identifier': change['poslanec']['id'], 'scheme': 'nrsr.sk'}}},
projection={'id': 1})
if existing:
pid = existing['id']
else:
p = Person.scrape(change['poslanec']['id'], term)
pid = p.save()
# create or update the membership
m = Membership()
m.label = 'Poslanec Národnej rady SR'
m.role = 'member'
m.person_id = pid
m.organization_id = oid
m.sources = [{
'url': change_list['url'],
'note': 'Zmeny v poslaneckom zbore na webe NRSR'
}]
if change['zmena'] in ('Mandát vykonávaný (aktívny poslanec)', 'Mandát náhradníka vykonávaný'):
m.start_date = sk_to_utc(change['dátum'])
m.save()
# close previous membership of Izák, Jaroslav (9. 9. 2008 - 20. 5. 2009 - 12. 6. 2010)
if term == '4' and change['poslanec']['meno'] == 'Izák, Jaroslav' and change['dátum'] == '20. 5. 2009':
del m.start_date
m.end_date = change['dátum']
m.save()
elif change['zmena'] in ('Mandát zaniknutý', 'Mandát sa neuplatňuje', 'Mandát náhradníka zaniknutý'):
m.end_date = sk_to_utc(change['dátum'])
# only close an existing membership (counterexample: Érsek, Árpád, 27. 9. 2010 - 10. 3. 2012)
existing_only = True
# except an inaccuracy in source data for Šimko, Ivan (15. 10. 2002 - 15. 10. 2002)
if term == '3' and change['poslanec']['meno'] == 'Šimko, Ivan':
existing_only = False
m.save(existing_only)
elif change['zmena'] in ('Mandát nadobudnutý vo voľbách', 'Mandát náhradníka získaný'):
pass
else:
raise RuntimeError("unknown change '%s' of a membership in chamber" % change['zmena'])
logging.info('Scraped %s mandate changes' % len(change_list['_items']))
def save(self):
scraped = self.__dict__
existing = vpapi.getfirst('people', where={'identifiers': {'$elemMatch': self.identifiers[0]}})
if not existing:
resp = vpapi.post('people', scraped)
else:
# update by PUT is preferred over PATCH to correctly remove properties that no longer exist now
resp = vpapi.put('people', existing['id'], scraped, effective_date=effective_date)
if resp['_status'] != 'OK':
raise Exception(self.name, resp)
return resp['id']
def get_or_create(self, endpoint, item, refresh=False, where_keys=None):
sort = []
embed = []
where = {}
if where_keys:
for key in where_keys:
where[key] = item[key]
elif endpoint == 'memberships':
where = {
'person_id': item['person_id'],
'organization_id': item['organization_id']
}
where['start_date'] = item.get('start_date', {"$exists": False})
sort = [('start_date', -1)]
elif endpoint in ('motions', 'speeches'):
where = {'sources.url': item['sources'][0]['url']}
elif endpoint == 'vote-events':
embed = ['votes']
if 'motion_id' in item:
where = {'motion_id': item['motion_id']}
else:
where = {'start_date': item['start_date']}
elif endpoint == 'votes':
where = {
'vote_event_id': item['vote_event_id'],
'voter_id': item['voter_id'],
}
elif endpoint == 'events':
where = {'identifier': item['identifier']}
else:
where = {
'identifiers': {'$elemMatch': item['identifiers'][0]}}
created = False
resp = vpapi.getfirst(endpoint, where=where, sort=sort)
if not resp:
resp = vpapi.post(endpoint, item)
created = True
self.log('Created %s' % resp['_links']['self']['href'], DEBUG)
else:
pk = resp['id']
resp = vpapi.put("%s/%s" % (endpoint, pk), item)
self.log('Updated %s' % resp['_links']['self']['href'], DEBUG)
if resp['_status'] != 'OK':
raise Exception(resp)
if refresh:
resp = vpapi.get(
resp['_links']['self']['href'], sort=sort, embed=embed)
resp['_created'] = created
return resp
def parliamentary_group_membership(self):
# Returns parliamentary groups membership list with all needed information data
# for each member of every parliamentary group for Belarus Lower house parliament.
party_membership_list = []
roles = self.membership_correction()
party = self.parliamentary_groups()
url = party['url']
soup = scrape.download_html_file(url)
party = soup.find("h1").get_text()
existing_party = vpapi.getfirst("organizations", where={"name": party})
if existing_party:
for each_tr in soup.find("table", {"width": "595"}).findAll('tr')[1:]:
td_array = each_tr.findAll('td')
name = td_array[1].get_text().strip()
if "кіраўнік групы" in name.encode('utf-8'):
name = name.encode('utf-8').replace("кіраўнік групы", "").strip()
name = name[:len(name) - 4]
membership = "кіраўнік групы".decode('utf-8')
else:
membership = "Член".decode('utf-8')
names = name.split(" ")
first_name = names[1]
last_name = names[0]
name_ordered = last_name + ", " + first_name
existing = vpapi.getfirst("people", where={'sort_name': name_ordered})
if existing:
p_id = existing['id']
if existing_party['id'] and p_id:
party_membership_json = {
"organization_id": existing_party['id'],
"person_id": p_id,
"url": url,
"membership": membership,
"role": roles[membership.encode('utf-8')]
}
party_membership_list.append(party_membership_json)
return party_membership_list
def get_or_create(resource, item, key=None):
"""Unless the item already exists in the resource (identified by
`key` fields) create it. Return id of the item and a bool whether
the item was newly created or not. If key is not given, all fields
of the item are used as a key.
"""
if key is None:
key = item.keys()
query = {field: item[field] for field in key}
existing = vpapi.getfirst(resource, where=query)
if existing:
return existing['id'], False
resp = vpapi.post(resource, item)
return resp['id'], True
def parliamentary_groups(self):
# Scrapes parliamentary groups and Returns the list json doc of parliamentary group
url = "http://house.gov.by/index.php/,17543,,,,2,,,0.html"
index_start = url.index("/,") + 2
index_end = url.index(",,,,2")
identifier = url[index_start:index_end]
chamber = vpapi.getfirst("organizations", where={'identifiers': {'$elemMatch': {"identifier": "2", "scheme": "house.by"}}})
soup = scrape.download_html_file(url)
party_name = soup.find('h1').get_text()
party_json = {
"name": party_name,
"url": url,
"identifier": identifier,
"parent_id": chamber['id']
}
return party_json
def scrape_committee(self):
# Scrapes committee groups and Returns the list of
# committee groups with all the information needed for each.
print "\n\tScraping parliamentary committees from Moldova's parliament..."
committees = self.committee_list()
chamber_id = vpapi.getfirst("organizations",
where={"identifiers": {
"$elemMatch": {
"identifier": "20", "scheme": "parlament.md"
}
}})
committees_list = []
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
for committee in pbar(committees):
soup = scrape.download_html_file(committee['url'])
email_tag = soup.find("span", {"id": "dnn_ctr486_ViewCommissionPermanent_ctrlViewCommissionType_lblCommissionContacts"}).find('a')
phone = soup.find("span", {"id": "dnn_ctr486_ViewCommissionPermanent_ctrlViewCommissionType_lblCommissionContacts"}).find('p')
if phone.get_text().strip() != "":
phone_number = phone.get_text()[6:].strip()
else:
phone_number = None
if email_tag:
email = email_tag.get_text()
else:
email = None
committee_json = self.build_organization_doc("committe", committee['name'], committee['identifier'],
"", "", committee['url'], email, chamber_id['id'], )
del committee_json['founding_date']
del committee_json['dissolution_date']
if not email:
del committee_json['contact_details']
elif not phone_number:
del committee_json['contact_details']
else:
committee_json['contact_details'].append({
"label": "Tel.",
"type": "tel",
"value": phone_number
})
committees_list.append(committee_json)
print "\n\tScraping completed! \n\tScraped " + str(len(committees_list)) + " committees"
return committees_list
def scrape_chamber(self):
# Scrapes chambers and Returns the list of chambers with all the information needed for each
url = "http://www.parlament.md/Parlamentarismul%C3%AEnRepublicaMoldova/" \
"Istorie%C8%99ievolu%C8%9Bie/tabid/96/language/ro-RO/Default.aspx"
chambers_to_fix = {"XII": "12", "XIII": "13", "XIV": "14", "XV": "15", "XVI": "16", "XVII": "17",
"XVIII": "18", "XIX": "19", "XX": "20"}
chambers = []
soup = scrape.download_html_file(url)
print "\n\tScraping chambers from Moldova's parliament..."
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
for each_a in pbar(soup.find('div', {"class": "LocalizedContent"}).findAll('a')):
name = each_a.get_text().strip()
if name != "":
url = "http://www.parlament.md" + each_a.get('href')
if "(" in name:
chamber_roman = name[name.index('X'):name.index('(')].replace('-a', "").strip()
chamber_identifier = chambers_to_fix[chamber_roman]
founding_date = self.terms[chamber_identifier]['start_date']
dissolution_date = self.terms[chamber_identifier]['end_date']
else:
chamber_roman = name[-6:len(name)-3].strip()
chamber_identifier = chambers_to_fix[chamber_roman]
founding_date = self.terms[chamber_identifier]['start_date']
dissolution_date = self.terms[chamber_identifier]['end_date']
chamber_json = self.build_organization_doc("chamber", name, chamber_identifier, founding_date,
dissolution_date, url, "", "")
del chamber_json['contact_details']
del chamber_json['parent_id']
if chamber_identifier == "20":
del chamber_json['dissolution_date']
existing = vpapi.getfirst("organizations", where={'identifiers': {'$elemMatch': chamber_json['identifiers'][0]}})
if not existing:
resp = vpapi.post("organizations", chamber_json)
else:
# update by PUT is preferred over PATCH to correctly remove properties that no longer exist now
resp = vpapi.put("organizations", existing['id'], chamber_json, effective_date=self.effective_date())
if resp["_status"] != "OK":
raise Exception("Invalid status code")
chambers.append(chamber_json)
print "\n\tScraping completed! \n\tScraped " + str(len(chambers)) + " chambers"
return chambers
def scrape_membership(self):
# Iterates in chamber member json document and
# returns the list with the json document structure that Visegrad+ API accepts
print "\n\tScraping membership's data from Belarus Lowerhouese parliament..."
mp_list = parser.mps_list()
chamber_membership_list = []
members = {}
url = "http://house.gov.by/index.php/,17041,,,,2,,,0.html"
all_members = vpapi.getall("people")
for person in all_members:
members[person['identifiers'][0]['identifier']] = person['id']
chamber = vpapi.getfirst("organizations", where={"identifiers": {"$elemMatch": {"identifier": "2", "scheme": "house.by"}}})
for member in mp_list:
p_id = members[member['identifier']]
o_id = chamber['id']
chamber_membership_json = self.build_memberships_doc(p_id, o_id, member['membership'], member['role'], url)
chamber_membership_list.append(chamber_membership_json)
print "\n\tScraping completed! \n\tScraped " + str(len(chamber_membership_list)) + " members"
return chamber_membership_list
def committee_list(self):
# Returns the list of committee groups with basic information for each
url = "http://house.gov.by/index.php/,17052,,,,2,,,0.html"
soup = scrape.download_html_file(url)
committees = []
chamber = vpapi.getfirst("organizations", where={'identifiers': {'$elemMatch': {"identifier": "2", "scheme": "house.by"}}})
for each_div in soup.findAll('div', {"style": "margin-left:0px; padding-bottom: 1px;"}):
name = each_div.find('a').get_text().strip()
url = each_div.find('a').get('href')
index_start = url.index("/,") + 2
index_end = url.index(",,,,2")
identifier = url[index_start:index_end]
committee_json = {
"name": name,
"url": url,
"identifier": identifier,
"parent_id": chamber['id']
}
committees.append(committee_json)
return committees
def scrape_committee_members(self):
# Iterates in every committee member json doc and returns the
# list with the json document structure that Visegrad+ API accepts
print "\n\tScraping committee groups from Belarus Upperhouse parliament...\n"
members = {}
committee_membership = []
all_members = vpapi.getall("people")
for member in all_members:
if member['identifiers'][0]['identifier'] not in members:
members[member['identifiers'][0]['identifier']] = member['id']
else:
continue
committee_membership_list = parser.committee_membership()
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
print "\n\tProcessing members of committee groups from Belarus Upperhouse parliament...\n"
for member in pbar(committee_membership_list):
if member['identifier'] in members:
p_id = members[member['identifier']]
else:
p_id = None
existing = vpapi.getfirst("organizations", where={"name": member['committee_name'], "parent_id": member['committee_parent_id']})
if existing:
o_id = existing['id']
else:
o_id = None
if p_id and o_id:
committee_membership_json = self.build_memberships_doc(p_id, o_id, member['membership'],
member['role'], member['url'])
committee_membership.append(committee_membership_json)
else:
continue
print "\n\tScraping completed! \n\tScraped " + str(len(committee_membership)) + " members"
return committee_membership
def scrape_parliamentary_groups(self):
# Scrapes parliamentary groups and Returns the list of
# parliamentary groups with all the information needed for each
chamber_id = vpapi.getfirst("organizations",
where={"identifiers": {
"$elemMatch": {
"identifier": "20", "scheme": "parlament.md"
}
}})
parties_list = self.parliamentary_group_list()
parties = []
print "\n\tScraping parliamentary groups from Moldova's parliament..."
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
for party in pbar(parties_list):
founding_date = self.terms["20"]["start_date"]
party_json = self.build_organization_doc("parliamentary group", party['name'],
party['identifier'], founding_date, "",
party['url'], "", chamber_id['id'])
del party_json['contact_details']
del party_json['dissolution_date']
parties.append(party_json)
return parties
def scrape_new_debates(term):
"""Scrape and save speeches from debates of the given term, one
of those newer terms where transcripts of debates are published
in parts assigned to individual speakers.
Returns number of scraped speeches.
"""
debate_part_kinds = {
'Uvádzajúci uvádza bod': 'speech',
'Vstup predsedajúceho': 'speech',
'Vystúpenie spoločného spravodajcu': 'speech',
'Vystúpenie': 'speech',
'Vystúpenie v rozprave': 'speech',
'Vystúpenie s faktickou poznámkou': 'speech',
'Vystúpenie s procedurálnym návrhom': 'speech',
'Prednesenie otázky': 'question',
'Zodpovedanie otázky': 'answer',
'Doplňujúca otázka / reakcia zadávajúceho': 'question',
'Prednesenie interpelácie': 'question',
'Odpoveď na interpeláciu': 'answer',
'scene': 'scene'
}
def insert_speech(kind):
"""Insert a speech entity for the given debate part kind
and data from parent scope variables and update end date
of the corresponding session and sitting. Delete `text`
variable."""
nonlocal text, last_speech_enddatetime
if not text: return
speech = {
'text': text.strip().replace('[', '(').replace(']', ')'),
'date': start_datetime,
'type': debate_part_kinds.get(kind, 'speech'),
'position': len(speeches) + 1,
'event_id': sitting_id,
'sources' : [{
'url': dpart_url,
'note': 'Prepis časti debaty na webe NRSR'
}]
}
if dpart_video:
speech['video'] = dpart_video
if kind != 'scene':
speech['creator_id'] = speaker_id
speech['attribution_text'] = attribution.strip()
speeches.append(speech)
text = ''
if end_datetime > session_end_date:
vpapi.patch('events', session_id, {'end_date': end_datetime})
if end_datetime > sitting_end_date:
vpapi.patch('events', sitting_id, {'end_date': end_datetime})
last_speech_enddatetime = datetime.strptime(end_datetime, '%Y-%m-%dT%H:%M:%S')
logging.info('Scraping debates of term `%s`' % term)
chamber_id = get_chamber_id(term)
# prepare mapping from MP's name to id
people = vpapi.getall('people', projection={'name': 1})
mps = {mp['name']: mp['id'] for mp in people}
# load name corrections
with open(os.path.join(CONF_DIR, 'name_corrections.json'), encoding='utf8') as f:
name_corrections = json.load(f)
# scraping will start since the most recent sitting start date
last_sitting = vpapi.getfirst('events',
where={'type': 'sitting', 'organization_id': chamber_id},
sort='-start_date')
since_date = last_sitting['start_date'][:10] if last_sitting else None
# scrape list of debate parts
debate_parts = parse.new_debates_list(term, since_date)
speech_count = 0
session_name = ''
speeches = []
for dp in debate_parts['_items']:
# stop at very recent debate parts (may be incomplete)
start_datetime = sk_to_utc('%s %s' % (dp['dátum'], dp['trvanie']['od']))
sd = datetime.strptime(start_datetime, '%Y-%m-%dT%H:%M:%S')
if datetime.utcnow() - sd < timedelta(days=5):
break
# skip already scraped debate parts
existing = vpapi.getfirst('speeches', where={'sources.url': dp['prepis']['url']})
if existing: continue
logging.info('Scraping debate part %s %s-%s (id=%s)' %
(dp['dátum'], dp['trvanie']['od'], dp['trvanie']['do'], dp['prepis']['id']))
dpart = parse.debate_of_terms56(dp['prepis']['id'])
if not dpart['riadky']: continue
end_datetime = sk_to_utc('%s %s' % (dp['dátum'], dp['trvanie']['do']))
dpart_kind = dp['druh']
dpart_url = dp['prepis']['url']
dpart_video = dp['video']['url'] if 'video' in dp else None
if not session_name.startswith('%s. ' % dp['schôdza']):
# create new session event
session_name = '%s. schôdza' % dp['schôdza']
session = {
'name': session_name,
'identifier': dp['schôdza'],
'organization_id': chamber_id,
'type': 'session',
'start_date': start_datetime,
'end_date': end_datetime,
}
key = ('organization_id', 'type', 'identifier')
session_id, _ = get_or_create('events', session, key)
session_end_date = end_datetime
# find the last moment of the last sitting of this session
session_last_sitting = vpapi.getfirst('events',
where={'type': 'sitting', 'parent_id': session_id},
sort='-start_date')
if session_last_sitting:
last_speech_enddatetime = datetime.strptime(session_last_sitting['end_date'], '%Y-%m-%dT%H:%M:%S')
sitting_identifier = session_last_sitting['identifier']
sitting_id = session_last_sitting['id']
sitting_end_date = session_last_sitting['end_date']
else:
last_speech_enddatetime = datetime.min
sitting_identifier = '0'
if sd - last_speech_enddatetime > timedelta(hours=5):
# create new sitting event
sitting_identifier = str(int(sitting_identifier) + 1)
sitting_name = '%s. deň rokovania, %s' % (sitting_identifier, dp['dátum'])
sitting = {
'name': sitting_name,
'identifier': sitting_identifier,
'organization_id': chamber_id,
'type': 'sitting',
'start_date': start_datetime,
'end_date': end_datetime,
'parent_id': session_id,
}
key = ('parent_id', 'type', 'identifier')
sitting_id, _ = get_or_create('events', sitting, key)
sitting_end_date = end_datetime
# save speeches of the previous sitting
if len(speeches) > 0:
vpapi.post('speeches', speeches)
speech_count += len(speeches)
if dp != debate_parts['_items'][0]:
logging.info('Scraped %s speeches from previous sitting' % len(speeches))
speeches = []
# add the first speaker name that is sometimes missing
first_speaker = '<strong>%s, %s</strong>' % (dp['osoba']['meno'], dp['osoba']['funkcia'])
dpart['riadky'].insert(0, first_speaker)
# extract speeches from the debate part
text = ''
within_scene = False
for par in dpart['riadky']:
if not par: continue
par = par.replace('\n', ' ').strip()
# skip eventual speech number
if re.match('^(\d+)\.$', par): continue
# convert brackets to parentheses
par = re.sub(r'\[(.*?)\]', r'(\1)', par)
# convert all inner nested parentheses to brackets
n = 1
while n >= 1:
(par, n) = re.subn(r'\((.*?)\((\.*?)\)(.*?)\)', r'(\1[\2]\3)', par, flags=re.DOTALL)
# process eventual multiparagraph scene
if par.startswith('(') and par.count('(') > par.count(')'):
# save eventual previous speech
insert_speech(dpart_kind)
text = '<p>%s</p>' % lxml.html.fromstring(par[1:]).text_content()
within_scene = True
continue
if within_scene:
if par.endswith(')') and par.count(')') > par.count('('):
text += '\n\n<p>%s</p>' % lxml.html.fromstring(par[:-1]).text_content()
insert_speech('scene')
within_scene = False
else:
text += '\n\n<p>%s</p>' % lxml.html.fromstring(par).text_content()
continue
# process eventual new speaker
# format `Doe, John, foreign minister`
speech_start_pattern = r'<strong>(\w+), (\w+\.?)( (\w+\.?))?, (.*)</strong>'
sp = re.match(speech_start_pattern, par, re.DOTALL)
if sp:
# save eventual previous speech
insert_speech(dpart_kind)
# identify speaker
name = '%s %s' % (sp.group(2), sp.group(1))
if (sp.group(4)):
name = name.replace(' ', ' %s ' % sp.group(4))
attribution = sp.group(5)
if name in name_corrections:
name = name_corrections[name]
if len(name) == 0: continue
speaker_id = mps.get(name)
# create unknown speakers
if not speaker_id:
logging.warn('Speaker `%s, %s` not found, creating new Person' % (name, attribution))
name_parts = re.match(r'(\w+\.?)( (\w+\.?))? (\w+)', name)
person = {
'name': name,
'family_name': name_parts.group(4),
'given_name': name_parts.group(1)
}
person['sort_name'] = '%s, %s' % (person['family_name'], person['given_name'])
if name_parts.group(3):
person['additional_name'] = name_parts.group(3)
person['sort_name'] += ' %s' % person['additional_name']
resp = vpapi.post('people', person)
speaker_id = resp['id']
mps[name] = speaker_id
continue
# remove HTML tags
par = lxml.html.fromstring(par).text_content()
# process eventual scene in this paragraph
scene_pattern = r'(.*?)\(\s*([\d%s][^\(\)]{2,}[\.?!“])\s*\)(.*)$' % scrapeutils.CS_UPPERS
while True:
scene = re.match(scene_pattern, par, re.DOTALL)
if not scene: break
if scene.group(1):
text += '\n\n<p>%s</p>' % scene.group(1).strip()
insert_speech(dpart_kind)
text = '<p>%s</p>' % scene.group(2).strip()
insert_speech('scene')
par = scene.group(3)
if par:
text += '\n\n<p>%s</p>' % par
insert_speech(dpart_kind)
if len(speeches) > 0:
vpapi.post('speeches', speeches)
logging.info('Scraped %s speeches' % len(speeches))
speech_count += len(speeches)
logging.info('Scraped %s speeches in total' % speech_count)
def get_chamber_id(term):
"""Return chamber id of the given term."""
chamber = vpapi.getfirst('organizations', where={
'classification': 'chamber',
'identifiers': {'$elemMatch': {'identifier': term, 'scheme': 'nrsr.sk'}}})
return chamber['id'] if chamber else None
def scrape_motions(term):
"""Scrape and save motions from the given term that are not scraped
yet starting from the oldest ones. One Motion item, one VoteEvent
item and many Vote items are created for each scraped motion detail
page.
Returns number of scraped motions.
"""
logging.info('Scraping motions of term `%s`' % term)
# prepare mappings from source identifier to id for MPs and parliamentary groups
chamber_id = get_chamber_id(term)
people = vpapi.getall('people', projection={'identifiers': 1})
mps = {mp['identifiers'][0]['identifier']: mp['id'] for mp in people if 'identifiers' in mp}
orgs = vpapi.getall('organizations', where={'classification': 'parliamentary group', 'parent_id': chamber_id})
parl_groups = {c['name']: c['id'] for c in orgs}
# add differently spelled parliamentary groups
group_corrections = {
'2': {
'Klub HZDS': 'Klub ĽS-HZDS',
'Klub SMK': 'Klub SMK-MKP',
'Klub Nezávislí': 'Klub Nezávislý',
},
'3': {
'Klub HZDS': 'Klub ĽS-HZDS',
'Klub SDKÚ': 'Klub SDKÚ-DS',
'Klub Smer': 'Klub SMER-SD',
'Klub Smer-SD': 'Klub SMER-SD',
'Klub KNP': 'Klub nezávislých poslancov NR SR',
'Klub Nezávislý': 'Klub nezávislých poslancov NR SR',
},
}
for k, v in group_corrections.get(term, {}).items():
parl_groups[k] = parl_groups[v]
# prepare list of sessions that are not completely scraped yet
sessions_to_scrape = []
session_list = parse.session_list(term)
for session in session_list['_items']:
motions = parse.session(session['číslo'], term)
if len(motions['_items']) == 0: continue
last_motion_id = motions['_items'][-1]['id']
m_url = 'http://www.nrsr.sk/web/Default.aspx?sid=schodze/hlasovanie/hlasklub&ID=%s' % last_motion_id
existing = vpapi.getfirst('motions', where={'sources.url': m_url})
if existing: break
sessions_to_scrape.append((session, motions))
# scrape motions from those sessions
scraped_motions_count = 0
for s, motions in reversed(sessions_to_scrape):
logging.info('Scraping session `%s`' % s['názov'])
# insert the session event unless it already exists
session = {
'name': s['názov'],
'identifier': s['číslo'],
'organization_id': chamber_id,
'type': 'session',
}
try:
session['start_date'] = sk_to_utc(s['trvanie']) + 'T00:00:00'
session['end_date'] = session['start_date']
except ValueError:
# multiday session contains votes; dates are set by debates scraping
pass
key = ('organization_id', 'type', 'identifier')
session_id, _ = get_or_create('events', session, key)
for i, m in enumerate(motions['_items']):
# check if the motion is already present
m_id = re.search(r'ID=(\d+)', m['url']['výsledok']).group(1)
# we not use directly m['url']['kluby'] because it is not always present
m_url = 'http://www.nrsr.sk/web/Default.aspx?sid=schodze/hlasovanie/hlasklub&ID=%s' % m_id
existing = vpapi.getfirst('motions', where={'sources.url': m_url})
if existing: continue
try:
motion_id = None
vote_event_id = None
# insert motion
logging.info('Scraping motion %s of %s (voted at %s)' % (i+1, len(motions['_items']), m['dátum']))
parsed_motion = parse.motion(m['id'])
motion = {
'organization_id': chamber_id,
'legislative_session_id': session_id,
'identifier': parsed_motion['číslo'],
'text': parsed_motion['názov'],
'date': sk_to_utc(m['dátum']),
'sources': [{
'url': parsed_motion['url'],
'note': 'Hlasovanie na webe NRSR'
}],
}
if 'výsledok' in parsed_motion:
motion['result'] = 'pass' if parsed_motion['výsledok'] == 'Návrh prešiel' else 'fail'
resp = vpapi.post('motions', motion)
motion_id = resp['id']
# insert vote event
vote_event = {
'motion_id': motion_id,
'organization_id': chamber_id,
'legislative_session_id': session_id,
'identifier': parsed_motion['číslo'],
'start_date': motion['date'],
'sources': [{
'url': parsed_motion['url'],
'note': 'Hlasovanie na webe NRSR'
}],
}
if 'výsledok' in parsed_motion:
vote_event['result'] = motion['result']
if 'súčty' in parsed_motion:
options = {
'yes': '[z] za',
'no': '[p] proti',
'abstain': '[?] zdržalo sa',
'absent': '[0] neprítomní',
'not voting': '[n] nehlasovalo'
}
vote_event['counts'] = [
{'option': o, 'value': int(parsed_motion['súčty'][s])}
for o, s in options.items() if parsed_motion['súčty'][s] != ''
]
if len(vote_event['counts']) == 0:
del vote_event['counts']
resp = vpapi.post('vote-events', vote_event)
vote_event_id = resp['id']
# insert votes
if 'hlasy' in parsed_motion and len(parsed_motion['hlasy']) > 0:
vote_options = {
'z': 'yes',
'p': 'no',
'?': 'abstain',
'n': 'not voting',
'0': 'absent'
}
votes = []
for v in parsed_motion['hlasy']:
# skip MPs not applying their mandate
if v['hlas'] == '-': continue
pg = normalize_parlgroup_name(v['klub'])
votes.append({
'vote_event_id': vote_event_id,
'option': vote_options[v['hlas']],
'voter_id': mps.get(v['id']),
'group_id': parl_groups.get(pg),
})
if len(votes) > 0:
resp = vpapi.post('votes', votes)
# delete incomplete data if insertion of the motion, vote event or votes failed
except:
if motion_id:
vpapi.delete('motions', motion_id)
if vote_event_id:
vpapi.delete('vote-events', vote_event_id)
raise
scraped_motions_count += 1
logging.info('Scraped %s motions of term `%s`' % (scraped_motions_count, term))
return scraped_motions_count
def scrape_from_group_and_save(group_type, id, term):
"""Scrape memberships in a given group and save (or update) them.
If group or MP referred by the membership does not exist, scrape
and save it/him/her.
"""
group = parse.group(group_type, id)
# if group is not scraped yet, scrape and save it
g = vpapi.getfirst('organizations',
where={
'classification': group_type,
'identifiers': {'$elemMatch': {'identifier': id, 'scheme': 'nrsr.sk'}}},
projection={'id': 1})
if g:
oid = g['id']
else:
o = Organization.scrape(group_type, id)
oid = o.save()
roles = {
'člen': 'member',
'členka': 'member',
'predseda': 'chairman',
'predsedníčka': 'chairwoman',
'podpredseda': 'vice-chairman',
'podpredsedníčka': 'vice-chairwoman',
'vedúci': 'chairman',
'vedúca': 'chairwoman',
'náhradník': 'substitute',
'náhradníčka': 'substitute',
'overovateľ': 'verifier',
'overovateľka': 'verifier',
'poverený vedením klubu': 'chairman',
'podpredseda poverený vedením výboru': 'vice-chairman',
'náhradný člen': 'substitute',
'náhradná členka': 'substitute',
}
for member in group['členovia']:
logging.info('Scraping membership of `%s`' % member['meno'])
# if member MP is not scraped yet, scrape and save him
existing = vpapi.getfirst('people',
where={'identifiers': {'$elemMatch': {'identifier': member['id'], 'scheme': 'nrsr.sk'}}},
projection={'id': 1})
if existing:
pid = existing['id']
else:
p = Person.scrape(member['id'], term)
pid = p.save()
m = Membership()
m.person_id = pid
m.organization_id = oid
m.sources = [{
'url': group['url'],
'note': 'Profil na webe NRSR'
}]
# create or update all periods of the membership
for period in member['obdobia']:
if period.get('rola'):
m.label = period['rola'].capitalize() + ' v skupine ' + group['názov']
m.role = roles[period['rola'].lower()]
else:
m.label = 'V skupine ' + group['názov']
if period.get('od'):
m.start_date = sk_to_utc(period.get('od'))
if period.get('do'):
m.end_date = sk_to_utc(period.get('do'))
m.save()
for attr in ('role', 'start_date', 'end_date'):
if hasattr(m, attr):
delattr(m, attr)
logging.info('Scraped %s memberships' % len(group['členovia']))
# close all open memberships in this group that were not updated
logging.info('Closing not updated open memberships')
present = datetime.utcnow() - timedelta(minutes=10)
query = {
'organization_id': oid,
'$or': [{'end_date': {'$exists': False}}, {'end_date': {'$in': [None, '']}}],
'updated_at': {'$lt': present.isoformat()}
}
to_close = vpapi.getall('memberships', where=query)
for m in to_close:
vpapi.patch('memberships', m['id'], {'end_date': datestring_add(effective_date, -1)})
def scrape(countries, people, votes):
global effective_date
effective_date = date.today().isoformat()
# execute MP's bio data.
georgia = georgia_scraper.GeorgiaScraper()
armenia = armenia_scraper.ArmeniaScraper()
ukraine = ukraine_scraper.UkraineScraper()
belarus_lowerhouse = belarus_lowerhouse_scraper.BelarusLowerhouseScraper()
belarus_upperhouse = belarus_upperhouse_scraper.BelarusUpperhouseScraper()
moldova = moldova_scraper.MoldovaScraper()
references = {"georgia": georgia, "armenia": armenia, "ukraine": ukraine,
"belarus-lowerhouse": belarus_lowerhouse, "moldova": moldova,
"belarus-upperhouse": belarus_upperhouse}
countries_array = []
if countries == "all":
for key in references:
countries_array.append(key)
else:
countries_array = countries.split(',')
indexes = []
for country in countries_array:
if country.lower() not in references:
indexes.append(countries_array.index(country))
if len(indexes) > 0:
countries_array.pop(indexes)
with open(os.path.join(BASE_DIR, 'access.json')) as f:
creds = json.load(f)
if len(countries_array) > 0:
for item in sorted(countries_array):
if internet_on(): # scrape and post data from parliaments if there's internet connection
print "\n\tPosting and updating data from %s parliament" % item
print "\tThis may take a few minutes..."
vpapi.parliament(creds[item.lower()]['parliament'])
vpapi.timezone(creds[item.lower()]['timezone'])
vpapi.authorize(creds[item.lower()]['api_user'], creds[item.lower()]['password'])
if people == "yes":
members = references[item.lower()].scrape_mp_bio_data()
chamber = references[item.lower()].scrape_chamber()
parliamentary_groups = references[item.lower()].scrape_parliamentary_groups()
committee = references[item.lower()].scrape_committee()
data_collections = {
"a-people": members,
"b-chamber": chamber,
"c-parliamentary_groups": parliamentary_groups,
"d-committe": committee
}
# inserts data for each data collection in Visegrad+ Api
for collection in sorted(set(data_collections)):
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
print "\n\tPosting and updating data to the Visegrad+ from %s data collection\n\n" % \
collection[2:]
if len(data_collections[collection]) > 0:
for json_doc in pbar(data_collections[collection]):
if collection == "a-people":
where_condition = {'identifiers': {'$elemMatch': json_doc['identifiers'][0]}}
collection_of_data = "people"
elif collection == "c-parliamentary_groups" or collection == "d-committe":
if item.lower() == "armenia" or item.lower() == "belarus-upperhouse"\
or item.lower() == "ukraine":
where_condition = {'name': json_doc['name'], "parent_id": json_doc['parent_id']}
else:
where_condition = {'name': json_doc['name']}
collection_of_data = "organizations"
elif collection == "b-chamber":
where_condition = {'identifiers': {'$elemMatch': json_doc['identifiers'][0]}}
collection_of_data = "organizations"
existing = vpapi.getfirst(collection_of_data, where=where_condition)
if not existing:
resp = vpapi.post(collection_of_data, json_doc)
else:
json_obj_id = existing['id']
items_to_delete = ["created_at", "updated_at", "_links", "id"]
for item_delete in items_to_delete:
del existing[item_delete]
if json.loads(json.dumps(json_doc)) == existing:
continue
else:
resp = vpapi.put(collection_of_data, json_obj_id, json_doc, effective_date=effective_date)
# update by PUT is preferred over PATCH to correctly remove properties that no longer exist now
if resp["_status"] != "OK":
raise Exception("Invalid status code")
print "\n\tFinished Posting and updating data from %s data collection\n" % collection[2:]
if item.lower() != "georgia":
memberships = {
"chambers": references[item.lower()].scrape_membership(),
"parliamentary_groups": references[item.lower()].scrape_parliamentary_group_membership(),
"committees": references[item.lower()].scrape_committee_members()
}
elif item.lower() == "georgia":
memberships = {
"chambers": references[item.lower()].scrape_membership()
}
for data_collection in memberships:
widgets_stat = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
prog_bar = ProgressBar(widgets=widgets_stat)
if len(memberships[data_collection]) > 0:
print "\n\tPosting and updating data from %s membership data collection\n" % data_collection
for json_doc in prog_bar(memberships[data_collection]):
existing = vpapi.getfirst("memberships", where={'organization_id': json_doc['organization_id'],
"person_id": json_doc['person_id']})
if not existing:
resp = vpapi.post("memberships", json_doc)
else:
json_obj_id = existing['id']
items_to_delete = ["created_at", "updated_at", "_links", "id"]
for item_delete in items_to_delete:
del existing[item_delete]
if json.loads(json.dumps(json_doc)) == existing:
continue
else:
resp = vpapi.put("memberships", json_obj_id, json_doc, effective_date=effective_date)
if resp["_status"] != "OK":
raise Exception("Invalid status code")
print "\n\tFinished Posted and updated data from %s membership data collection\n" % data_collection
else:
print "\n\tThere is no data from %s membership data collection\n" % data_collection
continue
if votes == "yes":
if item.lower() == "ukraine":
events = references[item.lower()].scrape_events()
try:
if len(events) > 0:
widgets_events = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar_events = ProgressBar(widgets=widgets_events)
for json_doc in pbar_events(events):
existing_event = vpapi.getfirst("events", where={'identifier': json_doc['identifier']})
if not existing_event:
resp = vpapi.post("events", json_doc)
else:
resp = vpapi.put("events", json_doc['id'], json_doc, effective_date=effective_date)
if resp["_status"] != "OK":
raise Exception("Invalid status code")
print "\n\tFinished Posting and updating data from events data collection"
else:
print "\n\tThere are no new events"
except BaseException as ex:
print ex.message
else:
print "\tThere's not any event to post from %s parliament" % item
motions_vote_events = references[item.lower()].vote_events()
voting_results = references[item.lower()].scrape_votes()
try:
if len(voting_results) > 0:
resp = vpapi.post("votes", voting_results)
if resp["_status"] != "OK":
raise Exception("Invalid status code")
print "\n\tFinished Posting and updating data from votes data collection"
except BaseException as ex:
print ex.message
elif item.lower() == "georgia":
voting_data_collections = {
"amotions": references[item.lower()].motions(),
"bvote-events": references[item.lower()].vote_events(),
}
votes = references[item.lower()].scrape_votes()
for collection in sorted(voting_data_collections):
try:
if len(voting_data_collections[collection]) > 0:
resp = vpapi.post(collection[1:], voting_data_collections[collection])
if resp["_status"] != "OK":
raise Exception("Invalid status code")
print "\n\tFinished Posting and updating data from %s data collection" % collection[1:]
except BaseException as ex:
print ex.message
print "\n\tPosting voting records from Georgia Parliament\n"
try:
if len(votes) > 0:
vpapi.post("votes", votes)
print "\n\tFinished Posting and updating data from votes data collection"
except BaseException as ex:
print ex.message
else:
print "\n\tThere are no voting records for %s" % item
vpapi.deauthorize()
else:
print "\n\tInternet connection problems for %s official parliament web page" % item
continue
else:
print "\n\tInvalid country/ies added"
def export_speeches(self):
speeches = self.load_json('speeches')
people = {}
prefix_regex = re.compile(
ur'(pred\u015bedavaju\u0107i )|(pred\u015bednik )|\
(generalni sekretar )', re.U)
for p in vpapi.getall('people'):
name = self.normalize_name(p['name'])
people[name] = p['id']
for speech in speeches:
session_id = speech.get('event_id')
speech['event_id'] = self.events_ids[session_id]
url = speech['sources'][0]['url']
if url.endswith('.pdf'):
parsed_speeches = self.download_pdf(url)
for n, s in enumerate(parsed_speeches):
text_speech = speech.copy()
text_speech['text'] = s['text']
text_speech['position'] = n + 1
text_speech['type'] = 'speech'
creator = self.normalize_name(s['creator'])
creator = prefix_regex.sub('', creator)
if creator in people:
text_speech['creator_id'] = people[creator]
else:
creator_id = None
for name in people:
if name in creator:
creator_id = people[name]
break
if creator_id is None:
resp = vpapi.getfirst('people',
where={
'name': {
'$regex':
s['creator'],
'options': 'i'
}
})
if resp is None:
self.log(
'Person "%(creator)s" not found. \
Creating one' % s, WARNING)
item = {
'name': s['creator'],
'sources': text_speech['sources']
}
resp = vpapi.post('people', item)
creator_id = resp['id']
people[creator] = creator_id
text_speech['creator_id'] = creator_id
self.get_or_create('speeches',
text_speech,
where_keys=['event_id', 'position'])
else:
self.get_or_create('speeches', speech)
def laws(self):
# Returns the list of the json structure of motions and
# vote events document with all the information data needed for both.
laws_url = "http://votes.parliament.ge/en/search/passed_laws?sEcho=1&iColumns=7&sColumns=&iDisplayStart=0" \
"&iDisplayLength=3000000&mDataProp_0=0&mDataProp_1=1&mDataProp_2=2&mDataProp_3=3&mDataProp_4=4" \
"&mDataProp_5=5&mDataProp_6=6&sSearch=&bRegex=false&sSearch_0=&bRegex_0=false" \
"&bSearchable_0=true&sSearch_1=&bRegex_1=false&bSearchable_1=true&sSearch_2=" \
"&bRegex_2=false&bSearchable_2=true&sSearch_3=&bRegex_3=false&bSearchable_3=true&sSearch_4=" \
"&bRegex_4=false&bSearchable_4=true&sSearch_5=&bRegex_5=false&bSearchable_5=true&sSearch_6=" \
"&bRegex_6=false&bSearchable_6=true&iSortCol_0=0&sSortDir_0=desc&iSortingCols=1&bSortable_0=true" \
"&bSortable_1=true&bSortable_2=true&bSortable_3=true&bSortable_4=true&bSortable_5=true" \
"&bSortable_6=true&parliament=1&start_date=&end_date=&_=1440146282982"
result = urlopen(laws_url).read()
json_result = json.loads(result)
laws_array = []
last_item = vpapi.getfirst("vote-events", sort="-start_date")
index_counter = 0
if last_item:
law_url = "/en/laws/" + last_item['id']
for element in json_result['aaData']:
soup = BeautifulSoup(element[1], "html.parser")
url_soup = soup.find('a').get('href')
if law_url == url_soup:
break
index_counter += 1
else:
index_counter = len(json_result['aaData'])
existing = vpapi.getfirst("organizations", where={"identifiers": {"$elemMatch": {"identifier": "8", "scheme": "parliament.ge"}}})
if existing:
organization_id = existing['id']
widgets = [' Progress: ', Percentage(), ' ', Bar(marker='#', left='[', right=']'),
' ', ETA(), " - Processed: ", Counter(), ' items ']
pbar = ProgressBar(widgets=widgets)
if len(json_result['aaData'][:index_counter]) > 0:
for item in pbar(json_result['aaData'][:index_counter]):
soup = BeautifulSoup(item[1], 'html.parser')
api_name = soup.get_text()
url = "http://votes.parliament.ge" + soup.find('a').get('href')
index_of_id = url.index('laws/')
index = index_of_id + 5
motion_id = url[index:]
date = self.local_to_utc(item[0] + " 04:00")
votes_for = item[3]
votes_against = item[4]
# votes_abstain = str(item[5])
votes_not_present = item[6]
json_motion = {
"date": date,
"start_date": date,
"sources": [{
"url": url,
"note": "ვებგვერდი"
}],
"id": motion_id,
"identifier": motion_id,
"motion_id": motion_id,
"organization_id": organization_id,
"text": api_name,
"result": "pass",
"counts": [
{
"option": "yes",
"value": votes_for
},
{
"option": "no",
"value": votes_against
},
{
"option": "absent",
"value": votes_not_present
}
]
}
laws_array.append(json_motion)
return laws_array
def get_latest_item(self, endpoint, time_key):
return vpapi.getfirst(endpoint, sort='-%s' % time_key)
def export_speeches(self):
speeches = self.load_json('speeches')
people = {}
prefix_regex = re.compile(
ur'(pred\u015bedavaju\u0107i )|(pred\u015bednik )|\
(generalni sekretar )', re.U)
for p in vpapi.getall('people'):
name = self.normalize_name(p['name'])
people[name] = p['id']
for speech in speeches:
session_id = speech.get('event_id')
speech['event_id'] = self.events_ids[session_id]
url = speech['sources'][0]['url']
if url.endswith('.pdf'):
parsed_speeches = self.download_pdf(url)
for n, s in enumerate(parsed_speeches):
text_speech = speech.copy()
text_speech['text'] = s['text']
text_speech['position'] = n + 1
text_speech['type'] = 'speech'
creator = self.normalize_name(s['creator'])
creator = prefix_regex.sub('', creator)
if creator in people:
text_speech['creator_id'] = people[creator]
else:
creator_id = None
for name in people:
if name in creator:
creator_id = people[name]
break
if creator_id is None:
resp = vpapi.getfirst(
'people', where={
'name': {
'$regex': s['creator'],
'options': 'i'
}
}
)
if resp is None:
self.log('Person "%(creator)s" not found. \
Creating one' % s, WARNING)
item = {
'name': s['creator'],
'sources': text_speech['sources']
}
resp = vpapi.post('people', item)
creator_id = resp['id']
people[creator] = creator_id
text_speech['creator_id'] = creator_id
self.get_or_create(
'speeches',
text_speech,
where_keys=['event_id', 'position']
)
else:
self.get_or_create('speeches', speech)
def vote_events(self):
print "\n\n\tScraping Motions and Vote Events data from Ukraine's parliament..."
vote_events = parser.vote_events_list()
index_vote_events = self.get_index("vote-events", "-start_date", vote_events)
index_motions = self.get_index("motions", "-date", vote_events)
index = min(index_vote_events, index_motions)
voting_events = []
motions = []
if len(vote_events) > 0:
print "\n\n\tPosting Motions and Vote events data to the Visegrad+ API from Ukraine's parliament..."
if len(vote_events[index:]) > 0:
widgets = [
" Progress: ",
Percentage(),
" ",
Bar(marker="#", left="[", right="]"),
" ",
ETA(),
" - Processed: ",
Counter(),
" events ",
]
pbar = ProgressBar(widgets=widgets)
for motion in pbar(vote_events[index:]):
json_motion = self.build_json_motion(
motion["date"][:19],
motion["sources"][0]["url"],
motion["id"],
motion["legislative_session_id"],
motion["organization_id"],
motion["text"],
motion["result"],
)
motions.append(json_motion)
existing = vpapi.getfirst("motions", where={"identifier": json_motion["identifier"]})
if not existing:
vpapi.post("motions", json_motion)
else:
continue
json_vote_event = self.build_vote_event_json(
motion["date"][:19],
motion["legislative_session_id"],
motion["id"],
motion["organization_id"],
motion["result"],
motion["counts"],
)
voting_events.append(json_vote_event)
existing1 = vpapi.getfirst("vote-events", where={"id": json_vote_event["id"]})
if not existing1:
vpapi.post("vote-events", json_vote_event)
else:
continue
print "\n\tFinished posting motions and vote events data."
print "\tScraped %s motions and vote events" % str(len(vote_events[index:]) * 2)
else:
print "\n\tThere are no new motion and vote events data."
else:
print "\n\tThere are no new motions or vote events."
return motions, voting_events
