def scrape_session(self, chamber, session):
billdirs_path = '/bills/%s/billhistory/%s_bills/' % (
session, chamber_name(chamber))
billdirs_url = urlparse.urljoin(self._ftp_root, billdirs_path)
with self.urlopen(billdirs_url) as bill_dirs:
for dir in parse_ftp_listing(bill_dirs):
bill_url = urlparse.urljoin(billdirs_url, dir) + '/'
with self.urlopen(bill_url) as bills:
for history in parse_ftp_listing(bills):
self.scrape_bill(chamber, session,
urlparse.urljoin(bill_url, history))
def scrape_bill(self, chamber, session, url):
with self.urlopen(url) as data:
bill = self.parse_bill_xml(chamber, session, data)
bill.add_source(url)
versions_url = url.replace("billhistory", "billtext/html")
versions_url = "/".join(versions_url.split("/")[0:-1])
bill_prefix = bill["bill_id"].split()[0]
bill_num = int(bill["bill_id"].split()[1])
long_bill_id = "%s%05d" % (bill_prefix, bill_num)
try:
with self.urlopen(versions_url) as versions_list:
bill.add_source(versions_url)
for version in parse_ftp_listing(versions_list):
if version.startswith(long_bill_id):
version_name = version.split(".")[0]
version_url = urlparse.urljoin(versions_url + "/", version)
bill.add_version(version_name, version_url)
except urllib2.URLError:
# Sometimes the text is missing
pass
self.save_bill(bill)
def scrape(self, chamber, session):
self.validate_session(session)
if len(session) == 2:
session = "%sR" % session
for btype in ["bills", "concurrent_resolutions", "joint_resolutions", "resolutions"]:
billdirs_path = "/bills/%s/billhistory/%s_%s/" % (session, chamber_name(chamber), btype)
billdirs_url = urlparse.urljoin(self._ftp_root, billdirs_path)
with self.urlopen(billdirs_url) as bill_dirs:
for dir in parse_ftp_listing(bill_dirs):
bill_url = urlparse.urljoin(billdirs_url, dir) + "/"
with self.urlopen(bill_url) as bills:
for history in parse_ftp_listing(bills):
self.scrape_bill(chamber, session, urlparse.urljoin(bill_url, history))
def scrape_bill(self, chamber, session, url):
with self.urlopen(url) as data:
bill = self.parse_bill_xml(chamber, session, data)
bill.add_source(url)
versions_url = url.replace('billhistory', 'billtext/html')
# URLs for versions inexplicably (H|S)(J|C) instead of (H|J)(CR|JR)
versions_url = versions_url.replace('JR', 'J').replace('CR', 'C')
versions_url = '/'.join(versions_url.split('/')[0:-1])
bill_prefix = bill['bill_id'].split()[0]
bill_num = int(bill['bill_id'].split()[1])
long_bill_id = "%s%05d" % (bill_prefix, bill_num)
try:
with self.urlopen(versions_url) as versions_list:
bill.add_source(versions_url)
for version in parse_ftp_listing(versions_list):
if version.startswith(long_bill_id):
version_name = version.split('.')[0]
version_url = urlparse.urljoin(versions_url + '/',
version)
bill.add_version(version_name, version_url)
except urllib2.URLError:
# Sometimes the text is missing
pass
self.save_bill(bill)
def scrape_session(self, chamber, session):
journal_root = urlparse.urljoin(self._ftp_root, ("/journals/" +
session +
"/html/"),
True)
if chamber == 'lower':
journal_root = urlparse.urljoin(journal_root, "house/", True)
else:
journal_root = urlparse.urljoin(journal_root, "senate/", True)
with self.urlopen(journal_root) as listing:
for name in parse_ftp_listing(listing):
if not name.startswith('81'):
continue
url = urlparse.urljoin(journal_root, name)
self.scrape_journal(url, chamber)
def scrape_bill(self, chamber, session, url):
with self.urlopen(url) as data:
bill = self.parse_bill_xml(chamber, session, data)
bill.add_source(url)
versions_url = url.replace('billhistory', 'billtext/html')
versions_url = '/'.join(versions_url.split('/')[0:-1])
bill_prefix = bill['bill_id'].split()[0]
bill_num = int(bill['bill_id'].split()[1])
long_bill_id = "%s%05d" % (bill_prefix, bill_num)
with self.urlopen(versions_url) as versions_list:
bill.add_source(versions_url)
for version in parse_ftp_listing(versions_list):
if version.startswith(long_bill_id):
version_name = version.split('.')[0]
version_url = urlparse.urljoin(versions_url + '/',
version)
bill.add_version(version_name, version_url)
self.save_bill(bill)
