def scrape_assem_bills(self, chamber, insert, session, year):
doc_type = {1: 'bill', 3: 'resolution', 5: 'concurrent resolution',
6: 'joint resolution'}
for docnum, bill_type in doc_type.iteritems():
parentpage_url = 'http://www.leg.state.nv.us/Session/%s/Reports/HistListBills.cfm?DoctypeID=%s' % (insert, docnum)
links = self.scrape_links(parentpage_url)
count = 0
for link in links:
count = count + 1
page_path = 'http://www.leg.state.nv.us/Session/%s/Reports/%s' % (insert, link)
with self.urlopen(page_path) as page:
page = page.decode("utf8").replace(u"\xa0", " ")
root = lxml.html.fromstring(page)
bill_id = root.xpath('string(/html/body/div[@id="content"]/table[1]/tr[1]/td[1]/font)')
title = root.xpath('string(/html/body/div[@id="content"]/table[1]/tr[5]/td)')
if insert.find('Special') != -1:
session = insert
bill = Bill(session, chamber, bill_id, title,
type=bill_type)
bill_text = root.xpath("string(/html/body/div[@id='content']/table[6]/tr/td[2]/a/@href)")
text_url = "http://www.leg.state.nv.us" + bill_text
bill.add_version("Bill Text", text_url)
primary, secondary = self.scrape_sponsors(page)
if primary[0] == 'By:':
primary.pop(0)
if primary[0] == 'ElectionsProceduresEthicsand':
primary[0] = 'Elections Procedures Ethics and'
full_name = ''
for part_name in primary:
full_name = full_name + part_name + " "
bill.add_sponsor('primary', full_name)
else:
for leg in primary:
bill.add_sponsor('primary', leg)
for leg in secondary:
bill.add_sponsor('cosponsor', leg)
minutes_count = 2
for mr in root.xpath('//table[4]/tr/td[3]/a'):
minutes = mr.xpath("string(@href)")
minutes_url = "http://www.leg.state.nv.us" + minutes
minutes_date_path = "string(//table[4]/tr[%s]/td[2])" % minutes_count
minutes_date = mr.xpath(minutes_date_path).split()
minutes_date = minutes_date[0] + minutes_date[1] + minutes_date[2] + " Minutes"
bill.add_document(minutes_date, minutes_url)
minutes_count = minutes_count + 1
self.scrape_actions(root, bill, "lower")
self.scrape_votes(page, bill, insert, year)
bill.add_source(page_path)
self.save_bill(bill)
def scrape_bill(self, session, chamber, bill_id, url):
with self.urlopen(url) as page:
page = lxml.html.fromstring(page)
page.make_links_absolute(url)
title = page.xpath("//br")[8].tail
if not title:
return
title = title.strip()
bill = Bill(session, chamber, bill_id, title)
bill.add_source(url)
action_link = page.xpath("//a[contains(@href, 'getActions')]")[0]
self.scrape_actions(bill, action_link.attrib['href'])
version_path = "//a[contains(., '%s')]"
for version_type in ('Introduced Bill', 'House Bill',
'Senate Bill', 'Engrossed Bill',
'Enrolled Act'):
path = version_path % version_type
links = page.xpath(path)
if links:
bill.add_version(version_type, links[0].attrib['href'])
for doc_link in page.xpath("//a[contains(@href, 'FISCAL')]"):
num = doc_link.text.strip().split("(")[0]
bill.add_document("Fiscal Impact Statement #%s" % num,
doc_link.attrib['href'])
bill['subjects'] = self.subjects[bill_id]
self.save_bill(bill)
def parse_bill(self, chamber, session, special, link):
bill_num = link.text.strip()
bill_type = re.search('type=(B|R|)', link.attrib['href']).group(1)
bill_id = "%s%s %s" % (bill_abbr(chamber), bill_type, bill_num)
url = info_url(chamber, session, special, bill_type, bill_num)
with self.urlopen(url) as page:
page = lxml.html.fromstring(page)
page.make_links_absolute(url)
title = page.xpath(
"//td[text() = 'Short Title:']/following-sibling::td")[0]
title = title.text.strip()
bill = Bill(session, chamber, bill_id, title)
bill.add_source(url)
self.parse_bill_versions(bill, page)
self.parse_history(bill, history_url(chamber, session, special,
bill_type, bill_num))
self.parse_votes(bill, vote_url(chamber, session, special,
bill_type, bill_num))
self.save_bill(bill)
def scrape_session_2009(self, chamber, session):
url, type = bills_url(chamber)
with self.urlopen(url) as page_html:
page = lxml.html.fromstring(page_html)
for element, attribute, link, pos in page.iterlinks():
if re.search("billtype=" + type + "&billnumber=[0-9]+", link) != None:
bill_page_url = bill_url(link)
with self.urlopen(bill_page_url) as bill_page_str:
bill_page = lxml.html.fromstring(bill_page_str)
splitted_link = link.split("=")
bill_number = splitted_link[-1]
bill_id = bill_page.cssselect('a[class="headerlink"]')
bill_id = bill_id[0]
bill_id = bill_id.text_content()
bill_title = bill_page.cssselect('td[style="color:Black"]')
bill_title = bill_title[0]
bill_title = bill_title.text_content()
bill = Bill(session, chamber, bill_id, bill_title)
bill.add_source(bill_page_url)
actions_table_list = bill_page.cssselect('table[rules="all"]')
actions_table = actions_table_list[0]
action_elements = actions_table.cssselect('tr')
# first element is not an action element
action_elements.pop(0)
for ae in action_elements:
action_element_parts = ae.cssselect('td')
action_date = action_element_parts[0]
actor_house = action_element_parts[1]
action_text = action_element_parts[2]
# look for acting comittees
match = re.search("(committee\(s\)|committee) on ([A-Z]{3}(/|-)[A-Z]{3}|[A-Z]{3})", action_text.text_content())
if(match != None):
actor = match.group(0)
elif(actor_house == 'H'):
actor = "lower"
elif (actor_house == 'S'):
actor = "upper"
else:
actor = chamber
action_date = dt.datetime.strptime(action_date.text_content(), '%m/%d/%Y')
if (re.search("The votes were as follows", action_text.text_content()) != None):
self.scrape_votes(action_text.text_content(), bill_page_url, actor_house, action_date, bill)
bill.add_action(actor, action_text, action_date)
with self.urlopen(versions_page_url(type, bill_number)) as versions_page_html:
versions_page = lxml.html.fromstring(versions_page_html)
versions_elements = versions_page.cssselect('span[class="searchtitle"]')
for ve in versions_elements:
element_text = ve.text_content()
version_name = element_text.rstrip("_.HTM")
bill.add_version(version_name, bill_version_url(element_text))
def scrape(self, chamber, session):
self.validate_session(session)
if chamber == 'upper':
other_chamber = 'lower'
bill_id = 'SB 1'
else:
other_chamber = 'upper'
bill_id = 'HB 1'
b1 = Bill(session, chamber, bill_id, 'A super bill')
b1.add_source('http://example.com/')
b1.add_version('As Introduced', 'http://example.com/SB1.html')
b1.add_document('Google', 'http://google.com')
b1.add_sponsor('primary', 'Bob Smith')
b1.add_sponsor('secondary', 'Johnson, Sally')
d1 = datetime.datetime.strptime('1/29/2010', '%m/%d/%Y')
v1 = Vote('upper', d1, 'Final passage', True, 2, 0, 0)
v1.yes('Smith')
v1.yes('Johnson')
d2 = datetime.datetime.strptime('1/30/2010', '%m/%d/%Y')
v2 = Vote('lower', d2, 'Final passage', False, 0, 1, 1)
v2.no('Bob Smith')
v2.other('S. Johnson')
b1.add_vote(v1)
b1.add_vote(v2)
b1.add_action(chamber, 'introduced', d1)
b1.add_action(chamber, 'read first time', d2)
b1.add_action(other_chamber, 'introduced', d2)
self.save_bill(b1)
def scrape_bill(self, chamber, session, bill_id, url):
with self.urlopen(url) as page:
page = lxml.html.fromstring(page)
page.make_links_absolute(url)
header = page.xpath('//h3/br')[0].tail.replace(' ', ' ')
title, primary_sponsor = header.split(' -- ')
if bill_id.startswith('H.B.') or bill_id.startswith('S.B.'):
bill_type = ['bill']
elif bill_id.startswith('H.C.R.') or bill_id.startswith('S.C.R.'):
bill_type = ['concurrent resolution']
elif bill_id.startswith('H.J.R.') or bill_id.startswith('S.J.R.'):
bill_type = ['joint resolution']
bill = Bill(session, chamber, bill_id, title, type=bill_type)
bill.add_sponsor('primary', primary_sponsor)
bill.add_source(url)
status_link = page.xpath('//a[contains(@href, "billsta")]')[0]
self.parse_status(bill, status_link.attrib['href'])
for link in page.xpath(
'//a[contains(@href, "bills/") and text() = "HTML"]'):
name = link.getprevious().tail.strip()
bill.add_version(name, link.attrib['href'])
self.save_bill(bill)
def scrape(self, chamber, session):
year = year_from_session(session)
url = bills_url(year)
with self.urlopen(url) as bills_page_html:
bills_page = lxml.html.fromstring(bills_page_html)
table_rows = bills_page.cssselect('tr')
# Eliminate empty rows
table_rows = table_rows[0:len(table_rows):2]
for row in table_rows:
row_elements = row.cssselect('td')
bill_document = row_elements[0]
bill_document.make_links_absolute(base_url())
element, attribute, link, pos = bill_document.iterlinks().next()
bill_id = element.text_content().rstrip('.pdf')
bill_document_link = link
title_and_sponsors = row_elements[1]
title_match = re.search('([A-Z][a-z]+.+[a-z])[A-Z]', title_and_sponsors.text_content())
sponsors_match = re.search('[a-z]([A-Z]+.+)', title_and_sponsors.text_content())
title = title_match.group(1)
sponsors = sponsors_match.group(1)
separated_sponsors = sponsors.split('--')
bill = Bill(session, chamber, bill_id, title)
bill.add_version('current', bill_document_link)
if separated_sponsors[1] == '(NONE)':
bill.add_sponsor('primary', separated_sponsors[0])
else:
bill.add_sponsor('cosponsor', separated_sponsors[0])
bill.add_sponsor('cosponsor', separated_sponsors[1])
versions_page_element = row_elements[2]
versions_page_element.make_links_absolute(base_url())
element, attribute, link, pos = versions_page_element.iterlinks().next()
bill.add_source(link)
self.scrape_versions(link, bill)
actions_page_element = row_elements[3]
element, attribute, link, pos = actions_page_element.iterlinks().next()
frame_link = base_url() + link.split('?Open&target=')[1]
self.scrape_actions(frame_link, bill)
votes_page_element = row_elements[7]
element, attribute, link, pos = votes_page_element.iterlinks().next()
frame_link = base_url() + link.split('?Open&target=')[1]
self.scrape_votes(frame_link, chamber, bill)
def scrape_bill(self, chamber, session, bill_type, number):
""" Creates a bill object
"""
if len(session) == 4:
session_url = session+'rs'
else:
session_url = session
url = BILL_URL % (session_url, bill_type, number)
with self.urlopen(url) as html:
doc = lxml.html.fromstring(html)
# title
# find <a name="Title">, get parent dt, get parent dl, then get dd within dl
title = doc.cssselect('a[name=Title]')[0] \
.getparent().getparent().cssselect('dd')[0].text.strip()
# create the bill object now that we have the title
print "%s %d %s" % (bill_type, number, title)
bill = Bill(session, chamber, "%s %d" % (bill_type, number), title)
bill.add_source(url)
self.parse_bill_sponsors(doc, bill) # sponsors
self.parse_bill_actions(doc, bill) # actions
self.parse_bill_documents(doc, bill) # documents and versions
self.parse_bill_votes(doc, bill) # votes
# add bill to collection
self.save_bill(bill)
def parse_senate_billpage(self, bill_url, year):
with self.urlopen(bill_url) as bill_page:
bill_page = BeautifulSoup(bill_page)
# get all the info needed to record the bill
bill_id = bill_page.find(id="lblBillNum").b.font.contents[0]
bill_title = bill_page.find(id="lblBillTitle").font.string
bill_desc = bill_page.find(id="lblBriefDesc").font.contents[0]
bill_lr = bill_page.find(id="lblLRNum").font.string
bill = Bill(year, 'upper', bill_id, bill_desc, bill_url=bill_url,
bill_lr=bill_lr, official_title=bill_title)
bill.add_source(bill_url)
# Get the primary sponsor
bill_sponsor = bill_page.find(id="hlSponsor").i.font.contents[0]
bill_sponsor_link = bill_page.find(id="hlSponsor").href
bill.add_sponsor('primary', bill_sponsor,
sponsor_link=bill_sponsor_link)
# cosponsors show up on their own page, if they exist
cosponsor_tag = bill_page.find(id="hlCoSponsors")
if cosponsor_tag and 'href' in cosponsor_tag:
self.parse_senate_cosponsors(bill, cosponsor_tag['href'])
# get the actions
action_url = bill_page.find(id="hlAllActions")['href']
self.parse_senate_actions(bill, action_url)
# stored on a separate page
versions_url = bill_page.find(id="hlFullBillText")
if versions_url:
self.parse_senate_bill_versions(bill, versions_url['href'])
self.save_bill(bill)
def scrape_bill(self, bill_url, chamber, session):
with self.urlopen(bill_url) as text:
if "Specified Bill could not be found" in text:
return False
page = lxml.html.fromstring(text)
page.make_links_absolute(bill_url)
bill_id = page.xpath("string(//h2)").split()[0]
summary = page.xpath(
"string(//*[starts-with(text(), 'Summary: ')])")
summary = summary.replace('Summary: ', '')
match = re.match(r"^([^:]+): ([^(]+)", summary)
if match:
subjects = [match.group(1).strip()]
title = match.group(2).strip()
else:
raise ScrapeError("Bad title")
bill = Bill(session, chamber, bill_id, title,
subjects=subjects)
bill.add_source(bill_url)
history_link = page.xpath("//a[text() = 'History']")[0]
history_url = history_link.attrib['href']
self.scrape_history(bill, history_url)
authors_link = page.xpath("//a[text() = 'Authors']")[0]
authors_url = authors_link.attrib['href']
self.scrape_authors(bill, authors_url)
try:
versions_link = page.xpath(
"//a[text() = 'Text - All Versions']")[0]
versions_url = versions_link.attrib['href']
self.scrape_versions(bill, versions_url)
except IndexError:
# Only current version
try:
version_link = page.xpath(
"//a[text() = 'Text - Current']")[0]
version_url = version_link.attrib['href']
bill.add_version("%s Current" % bill_id, version_url)
except IndexError:
# Some bills don't have any versions :(
pass
try:
votes_link = page.xpath("//a[text() = 'Votes']")[0]
self.scrape_votes(bill, votes_link.attrib['href'])
except IndexError:
# Some bills don't have any votes
pass
self.save_bill(bill)
return True
def parse_special_session_bill_status_page(self, bill_id, status_page, bill_table, session, chamber, sources):
title = bill_table.xpath('//tr[3]/td[2]')[0].text_content()
bill = Bill(session, chamber, bill_id, title)
for source in sources:
bill.add_source(source)
self.add_sponsors(bill, self.get_sponsor_table(status_page))
self.add_actions(bill, self.get_action_table(status_page))
return bill
def scrape(self, chamber, year):
# Data prior to 1997 is contained in pdfs
if year < "1997":
raise NoDataForYear(year)
bills_url = "http://www.leg.state.co.us/CLICS/CLICS" + year + "A/csl.nsf/%28bf-1%29?OpenView&Count=2000"
with self.lxml_context(bills_url) as bills_page:
table_rows = bills_page.cssselect("tr")
# Eliminate empty rows
table_rows = table_rows[0 : len(table_rows) : 2]
for row in table_rows:
print "row"
row_elements = row.cssselect("td")
bill_document = row_elements[0]
bill_document.make_links_absolute("http://www.leg.state.co.us")
element, attribute, link, pos = bill_document.iterlinks().next()
bill_id = element.text_content().rstrip(".pdf")
bill_document_link = link
title_and_sponsors = row_elements[1]
title_match = re.search("([A-Z][a-z]+.+[a-z])[A-Z]", title_and_sponsors.text_content())
sponsors_match = re.search("[a-z]([A-Z]+.+)", title_and_sponsors.text_content())
title = title_match.group(1)
sponsors = sponsors_match.group(1)
separated_sponsors = sponsors.split("--")
bill = Bill(year, chamber, bill_id, title)
bill.add_version("current", bill_document_link)
if separated_sponsors[1] == "(NONE)":
bill.add_sponsor("primary", separated_sponsors[0])
else:
bill.add_sponsor("cosponsor", separated_sponsors[0])
bill.add_sponsor("cosponsor", separated_sponsors[1])
versions_page_element = row_elements[2]
versions_page_element.make_links_absolute("http://www.leg.state.co.us")
element, attribute, link, pos = versions_page_element.iterlinks().next()
bill.add_source(link)
self.scrape_versions(link, bill)
actions_page_element = row_elements[3]
element, attribute, link, pos = actions_page_element.iterlinks().next()
frame_link = "http://www.leg.state.co.us" + link.split("?Open&target=")[1]
self.scrape_actions(frame_link, bill)
votes_page_element = row_elements[7]
element, attribute, link, pos = votes_page_element.iterlinks().next()
frame_link = "http://www.leg.state.co.us" + link.split("?Open&target=")[1]
self.scrape_votes(link, chamber, bill)
def scrape_year(self, year, chamber):
sep = '<h1>House</h1>'
if chamber == 'upper':
after = False
reg = '[5-9]'
else:
after = True
reg = '[1-4]'
with self.lxml_context("http://apps.leg.wa.gov/billinfo/dailystatus.aspx?year=" + str(year), sep, after) as page:
for element, attribute, link, pos in page.iterlinks():
if re.search("bill=" + reg + "[0-9]{3}", link) != None:
bill_page_url = "http://apps.leg.wa.gov/billinfo/" + link
with self.lxml_context(bill_page_url) as bill_page:
raw_title = bill_page.cssselect('title')
split_title = string.split(raw_title[0].text_content(), ' ')
bill_id = split_title[0] + ' ' + split_title[1]
bill_id = bill_id.strip()
session = split_title[3].strip()
title_element = bill_page.get_element_by_id("ctl00_ContentPlaceHolder1_lblSubTitle")
title = title_element.text_content()
bill = Bill(session, chamber, bill_id, title)
bill.add_source(bill_page_url)
self.scrape_actions(bill_page, bill)
for element, attribute, link, pos in bill_page.iterlinks():
if re.search("billdocs", link) != None:
if re.search("Amendments", link) != None:
bill.add_document("Amendment: " + element.text_content(), link)
elif re.search("Bills", link) != None:
bill.add_version(element.text_content(), link)
else:
bill.add_document(element.text_content(), link)
elif re.search("senators|representatives", link) != None:
with self.lxml_context(link) as senator_page:
try:
name_tuple = self.scrape_legislator_name(senator_page)
bill.add_sponsor('primary', name_tuple[0])
except:
pass
elif re.search("ShowRollCall", link) != None:
match = re.search("([0-9]+,[0-9]+)", link)
match = match.group(0)
match = match.split(',')
id1 = match[0]
id2 = match[1]
url = "http://flooractivityext.leg.wa.gov/rollcall.aspx?id=" + id1 + "&bienId=" +id2
with self.lxml_context(url) as vote_page:
self.scrape_votes(vote_page, bill, url)
self.save_bill(bill)
def scrape(self, chamber, session):
self.validate_session(session)
if chamber == "lower":
bill_abbr = "HB"
else:
bill_abbr = "SB"
bill_list_url = "http://www.le.state.ut.us/~%s/bills.htm" % (
session.replace(' ', ''))
self.log("Getting bill list for %s, %s" % (session, chamber))
try:
base_bill_list = self.soup_parser(self.urlopen(bill_list_url))
except:
# this session doesn't exist for this year
return
bill_list_link_re = re.compile('.*%s\d+ht.htm$' % bill_abbr)
for link in base_bill_list.findAll('a', href=bill_list_link_re):
bill_list = self.soup_parser(self.urlopen(link['href']))
bill_link_re = re.compile('.*billhtm/%s.*.htm' % bill_abbr)
for bill_link in bill_list.findAll('a', href=bill_link_re):
bill_id = bill_link.find(text=True).strip()
bill_info_url = bill_link['href']
bill_info = self.soup_parser(self.urlopen(bill_info_url))
bill_title, primary_sponsor = bill_info.h3.contents[2].replace(
' ', ' ').strip().split(' -- ')
bill = Bill(session, chamber, bill_id, bill_title)
bill.add_source(bill_info_url)
bill.add_sponsor('primary', primary_sponsor)
status_re = re.compile('.*billsta/%s.*.htm' %
bill_abbr.lower())
status_link = bill_info.find('a', href=status_re)
if status_link:
self.parse_status(bill, status_link['href'])
text_find = bill_info.find(
text="Bill Text (If you are having trouble viewing")
if text_find:
text_link_re = re.compile('.*\.htm')
for text_link in text_find.parent.parent.findAll(
'a', href=text_link_re)[1:]:
version_name = text_link.previous.strip()
bill.add_version(version_name, text_link['href'])
self.save_bill(bill)
def parse_bill(self, chamber, session, bill_id, bill_info_url):
with self.urlopen(bill_info_url) as bill_info_data:
bill_info = self.soup_parser(bill_info_data)
version_url = '%s/bill.doc' % bill_id
version_link = bill_info.find(href=version_url)
if not version_link:
# This bill was withdrawn
return
bill_title = version_link.findNext('p').contents[0].strip()
bill = Bill(session, chamber, bill_id, bill_title)
bill.add_version("Most Recent Version",
session_url(session) + version_url)
bill.add_source(bill_info_url)
sponsor_links = bill_info.findAll(href=re.compile(
'legislator/[SH]\d+\.htm'))
for sponsor_link in sponsor_links:
bill.add_sponsor('primary', sponsor_link.contents[0].strip())
action_p = version_link.findAllNext('p')[-1]
for action in action_p.findAll(text=True):
action = action.strip()
if (not action or action == 'last action' or
'Prefiled' in action):
continue
action_date = action.split('-')[0]
action_date = dt.datetime.strptime(action_date, '%b %d')
# Fix:
action_date = action_date.replace(
year=int('20' + session[2:4]))
action = '-'.join(action.split('-')[1:])
if action.endswith('House') or action.endswith('(H)'):
actor = 'lower'
elif action.endswith('Senate') or action.endswith('(S)'):
actor = 'upper'
else:
actor = chamber
bill.add_action(actor, action, action_date)
vote_link = bill_info.find(href=re.compile('.*/vote_history.pdf'))
if vote_link:
bill.add_document(
'vote_history.pdf',
bill_info_url.replace('.htm', '') + "/vote_history.pdf")
self.save_bill(bill)
def parse_standard_bill_status_page(self, bill_id, status_page, session, chamber, sources):
try:
title = status_page.xpath("/div/form[1]/table[2]/tr[3]/td[2]")[0].text_content()
except IndexError:
if len(status_page.xpath("/html/html")) == 2:
title = status_page.xpath('/html/html[2]/tr[1]/td[2]')[0].text_content()
else:
title = status_page.xpath('/html/html[3]/tr[1]/td[2]')[0].text_content()
bill = Bill(session, chamber, bill_id, title)
for source in sources:
bill.add_source(source)
self.add_sponsors(bill, self.get_sponsor_table(status_page))
self.add_actions(bill, self.get_action_table(status_page))
return bill
def get_bill_info(self, chamber, session, bill_detail_url, version_list_url):
"""Extracts all the requested info for a given bill.
Calls the parent's methods to enter the results into JSON files.
"""
if chamber == "House":
chamber = 'lower'
else:
chamber = 'upper'
with self.urlopen(bill_detail_url) as bill_html:
doc = lxml.html.fromstring(bill_html)
bill_id = doc.xpath('//title/text()')[0].split()[0]
bill_title = doc.xpath('//font[@size=-1]/text()')[0]
bill_type = {'F': 'bill', 'R':'resolution',
'C': 'concurrent resolution'}[bill_id[1]]
bill = Bill(session, chamber, bill_id, bill_title, type=bill_type)
bill.add_source(bill_detail_url)
# grab sponsors
sponsors = doc.xpath('//table[@summary="Show Authors"]/descendant::a/text()')
if sponsors:
primary_sponsor = sponsors[0].strip()
bill.add_sponsor('primary', primary_sponsor)
cosponsors = sponsors[1:]
for leg in cosponsors:
bill.add_sponsor('cosponsor', leg.strip())
# Add Actions performed on the bill.
bill_actions = self.extract_bill_actions(doc, chamber)
for action in bill_actions:
bill.add_action(action['action_chamber'],
action['action_text'],
action['action_date'],
type=action['action_type'])
# Get all versions of the bill.
# Versions of a bill are on a separate page, linked to from the column
# labeled, "Bill Text", on the search results page.
with self.urlopen(version_list_url) as version_html:
version_doc = lxml.html.fromstring(version_html)
for v in version_doc.xpath('//a[starts-with(@href, "/bin/getbill.php")]'):
version_url = urlparse.urljoin(VERSION_URL_BASE, v.get('href'))
bill.add_version(v.text.strip(), version_url)
self.save_bill(bill)
def scrape(self, chamber, session):
# internal id for the session, store on self so all methods have access
self.site_id = self.metadata['session_details'][session]['site_id']
self.build_subject_map()
# used for skipping bills from opposite chamber
start_letter = 'H' if chamber == 'lower' else 'S'
url = 'http://leg6.state.va.us/cgi-bin/legp604.exe?%s+lst+ALL' % self.site_id
while url:
with self.urlopen(url) as html:
doc = lxml.html.fromstring(html)
url = None # no more unless we encounter 'More...'
bills = doc.xpath('//ul[@class="linkSect"]/li')
for bill in bills:
link = bill.getchildren()[0]
bill_id = link.text_content()
# check if this is the 'More...' link
if bill_id == 'More...':
url = BASE_URL + link.get('href')
# skip bills from the other chamber
elif not bill_id.startswith(start_letter):
continue
else:
# create a bill
desc = bill.xpath('text()')[0].strip()
bill_type = {'B': 'bill',
'J': 'joint resolution',
'R': 'resolution'}[bill_id[1]]
bill = Bill(session, chamber, bill_id, desc,
type=bill_type)
bill_url = BASE_URL + link.get('href')
self.fetch_sponsors(bill)
self.scrape_bill_details(bill_url, bill)
bill['subjects'] = self.subject_map[bill_id]
bill.add_source(bill_url)
self.save_bill(bill)
def parse_bill_status_page(self, status_url, bill_url, session, chamber):
status_page = ElementTree(lxml.html.fromstring(self.urlopen(status_url)))
# see 2007 HB 2... weird.
try:
bill_id = status_page.xpath("/div/form[1]/table[2]/tr[2]/td[2]")[0].text_content()
except IndexError:
bill_id = status_page.xpath("/html/html[2]/tr[1]/td[2]")[0].text_content()
try:
title = status_page.xpath("/div/form[1]/table[2]/tr[3]/td[2]")[0].text_content()
except IndexError:
title = status_page.xpath("/html/html[3]/tr[1]/td[2]")[0].text_content()
bill = Bill(session, chamber, bill_id, title)
bill.add_source(bill_url)
self.add_sponsors(bill, status_page)
self.add_actions(bill, status_page)
return bill
def scrape_bill(self, chamber, session, bill_type, number):
""" Creates a bill object
"""
if len(session) == 4:
session_url = session+'rs'
else:
session_url = session
url = BILL_URL % (session_url, bill_type, number)
with self.urlopen(url) as html:
doc = lxml.html.fromstring(html)
# find <a name="Title">, get parent dt, get parent dl, then dd n dl
title = doc.xpath('//a[@name="Title"][1]/../../dd[1]/text()')[0].strip()
synopsis = doc.xpath('//font[@size="3"]/p/text()')[0].strip()
#print "%s %d %s" % (bill_type, number, title)
if 'B' in bill_type:
_type = ['bill']
elif 'J' in bill_type:
_type = ['joint resolution']
bill = Bill(session, chamber, "%s %d" % (bill_type, number), title,
type=_type, synopsis=synopsis)
bill.add_source(url)
self.parse_bill_sponsors(doc, bill) # sponsors
self.parse_bill_actions(doc, bill) # actions
self.parse_bill_documents(doc, bill) # documents and versions
self.parse_bill_votes(doc, bill) # votes
# subjects
subjects = []
for subj in doc.xpath('//a[contains(@href, "/subjects/")]'):
subjects.append(subj.text.split('-see also-')[0])
bill['subjects'] = subjects
# add bill to collection
self.save_bill(bill)
def scrape_assem_bills(self, chamber, insert, session):
doc_type = [1, 3, 5, 6]
for doc in doc_type:
parentpage_url = 'http://www.leg.state.nv.us/Session/%s/Reports/HistListBills.cfm?DoctypeID=%s' % (insert, doc)
links = self.scrape_links(parentpage_url)
count = 0
for link in links:
count = count + 1
page_path = 'http://www.leg.state.nv.us/Session/%s/Reports/%s' % (insert, link)
with self.urlopen(page_path) as page:
root = lxml.etree.fromstring(page, lxml.etree.HTMLParser())
bill_id = root.xpath('string(/html/body/div[@id="content"]/table[1]/tr[1]/td[1]/font)')
title = root.xpath('string(/html/body/div[@id="content"]/table[1]/tr[5]/td)')
bill = Bill(session, chamber, bill_id, title)
primary, secondary = self.scrape_sponsors(page_path)
if primary[0] == 'By:':
primary.pop(0)
if primary[0] == 'ElectionsProceduresEthicsand':
primary[0] = 'Elections Procedures Ethics and'
full_name = ''
for part_name in primary:
full_name = full_name + part_name + " "
bill.add_sponsor('primary', full_name)
else:
for leg in primary:
bill.add_sponsor('primary', leg)
for leg in secondary:
bill.add_sponsor('cosponsor', leg)
self.scrape_actions(page_path, bill, "Assembly")
self.scrape_votes(page_path, bill, "Assembly", insert, title)
bill.add_source(page_path)
self.save_bill(bill)
def scrape_bill(self, chamber, session, bill_type, number):
""" Creates a bill object
"""
if len(session) == 4:
session_url = session + "rs"
else:
session_url = session
url = BILL_URL % (session_url, bill_type, number)
with self.urlopen(url) as html:
doc = lxml.html.fromstring(html)
# find <a name="Title">, get parent dt, get parent dl, then dd n dl
title = doc.xpath('//a[@name="Title"][1]/../../dd[1]/text()')[0].strip()
# create the bill object now that we have the title
print "%s %d %s" % (bill_type, number, title)
if "B" in bill_type:
_type = ["bill"]
elif "J" in bill_type:
_type = ["joint resolution"]
bill = Bill(session, chamber, "%s %d" % (bill_type, number), title, type=_type)
bill.add_source(url)
self.parse_bill_sponsors(doc, bill) # sponsors
self.parse_bill_actions(doc, bill) # actions
self.parse_bill_documents(doc, bill) # documents and versions
self.parse_bill_votes(doc, bill) # votes
# subjects
subjects = []
for subj in doc.xpath('//a[contains(@href, "/subjects/")]'):
subjects.append(subj.text.split("-see also-")[0])
bill["subjects"] = subjects
# add bill to collection
self.save_bill(bill)
def parse_bill(self, chamber, session, special, link):
bill_number = link.contents[0]
type = re.search('type=(B|R|)', link['href']).group(1)
bill_id = "%s%s %s" % (bill_abbr(chamber), type, bill_number)
bill_info_url = info_url(chamber, session, special, type, bill_number)
with self.urlopen(bill_info_url) as info_page:
info_page = BeautifulSoup(info_page)
title_label = info_page.find(text='Short Title:')
title = title_label.findNext().contents[0]
bill = Bill(session, chamber, bill_id, title)
bill.add_source(bill_info_url)
self.parse_bill_versions(bill, info_page)
self.parse_history(bill, history_url(chamber, session, special,
type, bill_number))
self.parse_votes(bill, vote_url(chamber, session, special,
type, bill_number))
self.save_bill(bill)
def scrape_bill(self, chamber, session, bill_number, ga_num):
bill_url = self.urls['info'] % (bill_number, ga_num)
with self.urlopen(bill_url) as page:
page = lxml.html.fromstring(page)
title = page.xpath("//span[@id='lblAbstract']")[0].text
bill = Bill(session, chamber, bill_number, title)
bill.add_source(bill_url)
# Primary Sponsor
sponsor = page.xpath("//span[@id='lblBillSponsor']")[0].text_content().split("by")[-1]
sponsor = sponsor.replace('*','').strip()
bill.add_sponsor('primary',sponsor)
# Co-sponsors unavailable for scraping (loaded into page via AJAX)
# Full summary doc
summary = page.xpath("//span[@id='lblBillSponsor']/a")[0]
bill.add_document('Full summary', summary.get('href'))
# Actions
tables = page.xpath("//table[@id='tabHistoryAmendments_tabHistory_gvBillActionHistory']")
actions_table = tables[0]
action_rows = actions_table.xpath("tr[position()>1]")
for ar in action_rows:
action_taken = ar.xpath("td")[0].text
action_date = datetime.datetime.strptime(ar.xpath("td")[1].text.strip(), '%m/%d/%Y')
bill.add_action(chamber, action_taken, action_date)
votes_link = page.xpath("//span[@id='lblBillVotes']/a")
if(len(votes_link) > 0):
votes_link = votes_link[0].get('href')
bill = self.scrape_votes(bill, sponsor, 'http://wapp.capitol.tn.gov/apps/Billinfo/%s' % (votes_link,))
self.save_bill(bill)
def scrape_session_new(self, chamber, session):
if chamber == "lower":
bill_abbr = "H."
else:
bill_abbr = "S."
bill_list_path = "docs/bills.cfm?Session=%s&Body=%s" % (
session.split('-')[1], bill_abbr[0])
bill_list_url = "http://www.leg.state.vt.us/" + bill_list_path
bill_list = BeautifulSoup(self.urlopen(bill_list_url))
bill_link_re = re.compile('.*?Bill=%s\.\d+.*' % bill_abbr[0])
for bill_link in bill_list.findAll('a', href=bill_link_re):
bill_id = bill_link.string
bill_title = bill_link.parent.findNext('b').string
bill_info_url = "http://www.leg.state.vt.us" + bill_link['href']
bill = Bill(session, chamber, bill_id, bill_title)
bill.add_source(bill_info_url)
info_page = BeautifulSoup(self.urlopen(bill_info_url))
text_links = info_page.findAll('blockquote')[1].findAll('a')
for text_link in text_links:
bill.add_version(text_link.string,
"http://www.leg.state.vt.us" +
text_link['href'])
act_table = info_page.findAll('blockquote')[2].table
for row in act_table.findAll('tr')[1:]:
action = ""
for s in row.findAll('td')[1].findAll(text=True):
action += s + " "
action = clean_action(action)
match = re.search('Governor on (.*)$', action)
if match:
act_date = parse_exec_date(match.group(1).strip())
actor = 'Governor'
else:
if row['bgcolor'] == 'Salmon':
actor = 'lower'
else:
actor = 'upper'
if row.td.a:
act_date = row.td.a.string
else:
act_date = row.td.string
try:
act_date = re.search(
'\d{1,2}/\d{1,2}/\d{4,4}', act_date).group(0)
except AttributeError:
# No date, skip
continue
act_date = dt.datetime.strptime(act_date, '%m/%d/%Y')
bill.add_action(actor, action, act_date,
type=action_type(action))
vote_link = row.find('a', text='Details')
if vote_link:
vote_url = vote_link.parent['href']
self.parse_vote_new(bill, actor, vote_url)
sponsors = info_page.find(
text='Sponsor(s):').parent.parent.findAll('b')
bill.add_sponsor('primary', sponsors[0].string)
for sponsor in sponsors[1:]:
bill.add_sponsor('cosponsor', sponsor.string)
self.save_bill(bill)
def scrape_session_old(self, chamber, session):
if chamber == "lower":
bill_abbr = "H."
chamber_name = "House"
other_chamber = "Senate"
else:
bill_abbr = "S."
chamber_name = "Senate"
other_chamber = "House"
start_date = '1/1/%s' % session.split('-')[0]
data = urllib.urlencode({'Date': start_date,
'Body': bill_abbr[0],
'Session': session.split('-')[1]})
bill_list_url = "http://www.leg.state.vt.us/database/"\
"rintro/results.cfm"
bill_list = BeautifulSoup(urllib2.urlopen(bill_list_url, data))
bill_link_re = re.compile('.*?Bill=%s.\d+.*' % bill_abbr[0])
for bill_link in bill_list.findAll('a', href=bill_link_re):
bill_id = bill_link.string
bill_title = bill_link.parent.parent.findAll('td')[1].string
bill_info_url = "http://www.leg.state.vt.us" + bill_link['href']
bill = Bill(session, chamber, bill_id, bill_title)
bill.add_source(bill_info_url)
info_page = BeautifulSoup(self.urlopen(bill_info_url))
text_links = info_page.findAll('blockquote')[-1].findAll('a')
for text_link in text_links:
bill.add_version(text_link.string,
"http://www.leg.state.vt.us" +
text_link['href'])
sponsors = info_page.find(
text='Sponsor(s):').parent.findNext('td').findAll('b')
bill.add_sponsor('primary', sponsors[0].string)
for sponsor in sponsors[1:]:
bill.add_sponsor('cosponsor', sponsor.string)
# Grab actions from the originating chamber
act_table = info_page.find(
text='%s Status:' % chamber_name).findNext('table')
for row in act_table.findAll('tr')[3:]:
action = clean_action(row.td.string.replace(
' ', '').strip(':'))
act_date = row.findAll('td')[1].b.string.replace(' ', '')
if act_date != "":
detail = row.findAll('td')[2].b
if detail and detail.string != "":
action += ": %s" % detail.string.replace(' ', '')
bill.add_action(chamber, action, act_date,
type=action_type(action))
# Grab actions from the other chamber
act_table = info_page.find(
text='%s Status:' % other_chamber).findNext('table')
if act_table:
if chamber == 'upper':
act_chamber = 'lower'
else:
act_chamber = 'upper'
for row in act_table.findAll('tr')[3:]:
action = clean_action(row.td.string.replace(
' ', '').strip(':'))
act_date = row.findAll('td')[1].b.string.replace(
' ', '')
if act_date != "":
detail = row.findAll('td')[2].b
if detail and detail.string != "":
action += ": %s" % detail.string.replace(
' ', '')
date = dt.datetime.strptime(act_date, '%m/%d/%Y')
bill.add_action(act_chamber, action, act_date,
type=action_type(action))
self.save_bill(bill)
def scrape_session(self, chamber, year):
if chamber == 'upper':
bill_abbr = 'SB|SCR|SJR'
elif chamber == 'lower':
bill_abbr = 'HB|HCR|HJR'
# Sessions last 2 years, 1993-1994 was the 18th
session = str(18 + ((int(year) - 1993) / 2))
year2 = str(int(year) + 1)
# Full calendar year
date1 = '0101' + year[2:]
date2 = '1231' + year2[2:]
# Get bill list
bill_list_url = 'http://www.legis.state.ak.us/'\
'basis/range_multi.asp?session=%s&date1=%s&date2=%s' % (
session, date1, date2)
self.log("Getting bill list for %s %s (this may take a long time)." %
(chamber, session))
bill_list = self.soup_parser(self.urlopen(bill_list_url))
# Find bill links
re_str = "bill=%s\d+" % bill_abbr
links = bill_list.findAll(href=re.compile(re_str))
for link in links:
bill_id = link.contents[0].replace(' ', '')
bill_name = link.parent.parent.findNext('td').find(
'font').contents[0].strip()
bill = Bill(session, chamber, bill_id, bill_name.strip())
# Get the bill info page and strip malformed t
info_url = "http://www.legis.state.ak.us/basis/%s" % link['href']
info_page = self.soup_parser(self.urlopen(info_url))
bill.add_source(info_url)
# Get sponsors
spons_str = info_page.find(
text="SPONSOR(s):").parent.parent.contents[1]
sponsors_match = re.match(
' (SENATOR|REPRESENTATIVE)\([Ss]\) ([^,]+(,[^,]+){0,})',
spons_str)
if sponsors_match:
sponsors = sponsors_match.group(2).split(',')
bill.add_sponsor('primary', sponsors[0].strip())
for sponsor in sponsors[1:]:
bill.add_sponsor('cosponsor', sponsor.strip())
else:
# Committee sponsorship
bill.add_sponsor('committee', spons_str.strip())
# Get actions
act_rows = info_page.findAll('table', 'myth')[1].findAll('tr')[1:]
for row in act_rows:
cols = row.findAll('td')
act_date = cols[0].font.contents[0]
act_date = dt.datetime.strptime(act_date, '%m/%d/%y')
if cols[2].font.string == "(H)":
act_chamber = "lower"
elif cols[2].font.string == "(S)":
act_chamber = "upper"
else:
act_chamber = chamber
action = cols[3].font.contents[0].strip()
if re.match("\w+ Y(\d+) N(\d+)", action):
try:
vote = self.parse_vote(bill, action,
act_chamber, act_date,
cols[1].a['href'])
bill.add_vote(vote)
except:
self.log("Failed parsing vote at %s" %
cols[1].a['href'])
bill.add_action(act_chamber, action, act_date)
# Get subjects
bill['subjects'] = []
subject_link_re = re.compile('.*subject=\w+$')
for subject_link in info_page.findAll('a', href=subject_link_re):
subject = subject_link.contents[0].strip()
bill['subjects'].append(subject)
# Get versions
text_list_url = "http://www.legis.state.ak.us/"\
"basis/get_fulltext.asp?session=%s&bill=%s" % (
session, bill_id)
text_list = self.soup_parser(self.urlopen(text_list_url))
bill.add_source(text_list_url)
text_link_re = re.compile('^get_bill_text?')
for text_link in text_list.findAll('a', href=text_link_re):
text_name = text_link.parent.previousSibling.contents[0]
text_name = text_name.strip()
text_url = "http://www.legis.state.ak.us/basis/%s" % (
text_link['href'])
bill.add_version(text_name, text_url)
self.save_bill(bill)
def scrape_bill_pages(self, session, year_abr):
""" assemble information on a bill from a number of DBF files
"""
#Main Bill information
main_bill_url, main_bill_db = self.get_dbf(year_abr, 'MAINBILL')
# keep a dictionary of bills (mapping bill_id to Bill obj)
bill_dict = {}
for rec in main_bill_db:
bill_type = rec["billtype"]
bill_number = int(rec["billnumber"])
bill_id = bill_type + str(bill_number)
title = rec["synopsis"]
if bill_type[0] == 'A':
chamber = "lower"
else:
chamber = "upper"
bill = Bill(str(session), chamber, bill_id, title,
type=self._bill_types[bill_type[1:]])
bill.add_source(main_bill_url)
bill_dict[bill_id] = bill
#Sponsors
bill_sponsors_url, bill_sponsors_db = self.get_dbf(year_abr, 'BILLSPON')
for rec in bill_sponsors_db:
bill_type = rec["billtype"]
bill_number = int(rec["billnumber"])
bill_id = bill_type + str(bill_number)
bill = bill_dict[bill_id]
name = rec["sponsor"]
sponsor_type = rec["type"]
if sponsor_type == 'P':
sponsor_type = "Primary"
else:
sponsor_type = "Co-sponsor"
bill.add_sponsor(sponsor_type, name)
#Documents
bill_document_url, bill_document_db = self.get_dbf(year_abr, 'BILLWP')
#print bill_document_db[2]
for rec in bill_document_db:
bill_type = rec["billtype"]
bill_number = int(rec["billnumber"])
bill_id = bill_type + str(bill_number)
bill = bill_dict[bill_id]
document = rec["document"]
document = document.split('\\')
document = document[-2] + "/" + document[-1]
year = str(year_abr) + str((year_abr + 1))
#doc_url = "ftp://www.njleg.state.nj.us/%s/%s" % (year, document)
htm_url = 'http://www.njleg.state.nj.us/%s/Bills/%s' % (year_abr,
document.replace('.DOC', '.HTM'))
# name document based _doctype
doc_name = self._doctypes[rec['doctype']]
if rec['comment']:
doc_name += ' ' + rec['comment']
if rec['doctype'] in self._version_types:
bill.add_version(doc_name, htm_url)
else:
bill.add_document(doc_name, htm_url)
#Senate Votes
file1 = 'A' + str(year_abr)
file2 = 'A' + str(year_abr + 1)
file3 = 'S' + str(year_abr)
file4 = 'S' + str(year_abr + 1)
if str(year_abr) != '2010':
vote_info_list = [file1, file2, file3, file4]
else:
vote_info_list = [file1, file3]
for bill_vote_file in vote_info_list:
s_vote_url = 'ftp://www.njleg.state.nj.us/votes/%s.zip' % bill_vote_file
s_vote_zip, resp = self.urlretrieve(s_vote_url)
zipedfile = zipfile.ZipFile(s_vote_zip)
vfile = "%s.txt" % bill_vote_file
vote_file = zipedfile.open(vfile, 'U')
vdict_file = csv.DictReader(vote_file)
votes = {}
if bill_vote_file[0] == "A":
chamber = "lower"
else:
chamber = "upper"
for rec in vdict_file:
bill_id = rec["Bill"]
bill_id = bill_id.strip()
leg = rec["Full_Name"]
date = rec["Session_Date"]
date = datetime.strptime(date, "%m/%d/%Y")
action = rec["Action"]
leg_vote = rec["Legislator_Vote"]
vote_id = bill_id + "_" + action
vote_id = vote_id.replace(" ", "_")
passed = None
if vote_id not in votes:
votes[vote_id] = Vote(chamber, date, action, passed, None,
None, None, bill_id=bill_id)
if leg_vote == "Y":
votes[vote_id].yes(leg)
elif leg_vote == "N":
votes[vote_id].no(leg)
else:
votes[vote_id].other(leg)
#Counts yes/no/other votes and saves overall vote
for vote in votes.itervalues():
vote_yes_count = len(vote["yes_votes"])
vote_no_count = len(vote["no_votes"])
vote_other_count = len(vote["other_votes"])
vote["yes_count"] = vote_yes_count
vote["no_count"] = vote_no_count
vote["other_count"] = vote_other_count
if vote_yes_count > vote_no_count:
vote["passed"] = True
else:
vote["passed"] = False
vote_bill_id = vote["bill_id"]
bill = bill_dict[vote_bill_id]
bill.add_vote(vote)
#Actions
bill_action_url, bill_action_db = self.get_dbf(year_abr, 'BILLHIST')
for rec in bill_action_db:
bill_type = rec["billtype"]
bill_number = int(rec["billnumber"])
bill_id = bill_type + str(bill_number)
bill = bill_dict[bill_id]
action = rec["action"]
date = rec["dateaction"]
actor = rec["house"]
comment = rec["comment"]
action, atype = self.categorize_action(action)
if comment:
action += (' ' + comment)
bill.add_action(actor, action, date, type=atype)
# Subjects
subject_url, subject_db = self.get_dbf(year_abr, 'BILLSUBJ')
for rec in subject_db:
bill_id = rec['billtype'] + str(int(rec['billnumber']))
bill = bill_dict.get(bill_id)
if bill:
bill.setdefault('subjects', []).append(rec['subjectkey'])
else:
self.warning('invalid bill id in BILLSUBJ.DBF: %s' % bill_id)
# save all bills at the end
for bill in bill_dict.itervalues():
# add sources
bill.add_source(bill_sponsors_url)
bill.add_source(bill_document_url)
bill.add_source(bill_action_url)
bill.add_source(subject_url)
self.save_bill(bill)
def scrape_year(self, chamber, year, session):
if chamber == 'upper':
chamber_abbr = 'H'
elif chamber == 'lower':
chamber_abbr = 'S'
#set up POST data
values = [('txtsessionyear', year),
('txttitle', ''),
('txtlsrnumber', ''),
('Submit1', 'Submit')]
params = urllib.urlencode(values)
search_url = 'http://www.gencourt.state.nh.us/bill_status/Results.aspx'
#request page with list of all bills in year
with self.urlopen(search_url + '?' + params) as doc:
soup = BeautifulSoup(doc)
#parse results
bills = soup.find("table", {"class": "ptable"})
trs = soup.findAll("tr")
#go through all of the table rows with relevant data
tr_start = 8
tr_hop = 11
i = 0
while (tr_start + (tr_hop * i)) < len(trs):
tr = trs[tr_start + (tr_hop * i)]
i = i + 1
# strip off extra white space from name
id = tr.find("big").string.strip()
bill_id = tr.find("big").string.strip()
exp = re.compile("^(\w*)")
bill_id = exp.search(id).group(1)
# check to see if its in the proper chamber
exp = re.compile("^" + chamber_abbr)
if exp.search(bill_id) == None:
continue # in wrong house
# check to see it is a bill and not a resolution
exp = re.compile("B")
if exp.search(bill_id) == None:
continue # not a bill
# get bill_id suffix if exists
exp = re.compile("(-\w*)$")
res = exp.search(id)
if res != None:
bill_id = bill_id + res.group(1)
# get bill title
title = tr.findAll("b")[0]
bill_title = title.nextSibling.string
bill_title = bill_title.strip()
bill_title = bill_title.encode('ascii', 'xmlcharrefreplace')
# grab url of bill text
urls = tr.findAll("a")
textexp = re.compile("Bill Text")
textdoc = re.compile("Bill Docket")
textstat = re.compile("Bill Status")
textcall = re.compile("Roll Calls")
textaudio = re.compile("Audio Files")
for url in urls:
if textexp.search(str(url.string)) != None:
bill_url = self.get_bill_text(url)
if textdoc.search(str(url.string)) != None:
pass
if textstat.search(str(url.string)) != None:
add_bill_sponsors()
if textcall.search(str(url.string)) != None:
pass
if textaudio.search(str(url.string)) != None:
pass
bill = Bill(session, chamber, bill_id, bill_title)
bill.add_version("Bill text", bill_url)
bill.add_source(search_url)
self.save_bill(bill)
def scrape_bill(self, chamber, session, url):
url = url + "&Year=%s" % session
with self.urlopen(url) as page:
page = page.replace(' ', ' ').replace('<br>', '\n')
page = lxml.html.fromstring(page)
page.make_links_absolute(url)
title = page.xpath('//h3')[0].text.strip()
title = re.match(r"^\w+\s+\d+:\s+(.*)$", title).group(1)
bill_id = page.xpath("string(//pre[@class='billhistory']/b)")
bill_id = bill_id.split()[0].strip()
bill = Bill(session, chamber, bill_id, title)
bill.add_source(url)
hist = page.xpath("string(//pre[@class='billhistory'])").strip()
act_re = re.compile(r'^ (\d\d/\d\d/\d\d) (SENATE|HOUSE)'
r'(.*\n(\s{16,16}.*\n){0,})',
re.MULTILINE)
# Actions
for match in act_re.finditer(hist):
action = match.group(3).replace('\n', ' ')
action = re.sub(r'\s+', ' ', action).strip()
if match.group(2) == 'SENATE':
actor = 'upper'
else:
actor = 'lower'
date = match.group(1)
date = datetime.datetime.strptime(date, "%m/%d/%y")
for act_text in re.split(' -[HS]J \d+;? ?', action):
act_text = act_text.strip()
if not act_text:
continue
types = []
act_lower = act_text.lower()
if act_lower.startswith('introduced'):
types.append('bill:introduced')
if 'referred to' in act_lower:
types.append('committee:referred')
if 'died in committee' in act_lower:
types.append('committee:failed')
if 'favorable by' in act_lower:
types.append('committee:passed:favorable')
if 'amendment(s) adopted' in act_lower:
types.append('amendment:passed')
bill.add_action(actor, act_text, date, type=types)
# Sponsors
primary_sponsor = re.search(r'by ([^;(\n]+;?|\w+)',
hist).group(1).strip('; ')
bill.add_sponsor('primary', primary_sponsor)
cospon_re = re.compile(r'\((CO-SPONSORS|CO-AUTHORS)\) '
'([\w .]+(;[\w .\n]+){0,})',
re.MULTILINE)
match = cospon_re.search(hist)
if match:
for cosponsor in match.group(2).split(';'):
cosponsor = cosponsor.replace('\n', '').strip()
bill.add_sponsor('cosponsor', cosponsor)
# Versions
for link in page.xpath("//a[contains(@href, 'billtext/html')]"):
version = link.xpath('string(../../td[1])').strip()
bill.add_version(version, link.attrib['href'])
# House Votes
for link in page.xpath("//a[contains(@href, 'votes/html/h')]"):
bill.add_vote(self.scrape_lower_vote(link.attrib['href']))
# Senate Votes
for link in page.xpath("//a[contains(@href, 'votes/html/S')]"):
bill.add_vote(self.scrape_upper_vote(link.attrib['href']))
self.save_bill(bill)