def sponsor_for(sponsor_dict):
if sponsor_dict is None:
# TODO: This can hopefully be removed. In testing s414-113
# was missing sponsor data. But all bills have a sponsor?
return None
# TODO: Don't do regex matching here. Find another way.
m = re.match(r'(?P<title>(Rep|Sen))\. (?P<name>.*?) +\[(?P<party>[DRI])-(?P<state>[A-Z][A-Z])(-(?P<district>\d{1,2}|At Large))?\]$',
sponsor_dict['fullName'])
if m.group("district") is None:
district = None # a senator
elif m.group("district") == "At Large":
district = None # TODO: For backwards compatibility, we're returning None, but 0 would be better.
else:
# TODO: For backwards compatibility, we're returning a string, but an int would be better.
district = m.group('district')
return {
'title': m.group("title"),
'name': m.group("name"), # the firstName, middleName, lastName fields have inconsistent capitalization - some are all uppercase
'district': district,
'state': m.group('state'),
#'party': m.group('party'),
'thomas_id': utils.translate_legislator_id('bioguide', sponsor_dict['bioguideId'], 'thomas'), # TODO: Remove one day.
'bioguide_id': sponsor_dict['bioguideId'],
'type': 'person'
}
def create_govtrack_xml(amdt, options):
govtrack_type_codes = {'hr': 'h', 's': 's', 'hres': 'hr', 'sres': 'sr', 'hjres': 'hj', 'sjres': 'sj', 'hconres': 'hc', 'sconres': 'sc'}
root = etree.Element("amendment")
root.set("session", amdt['congress'])
root.set("chamber", amdt['amendment_type'][0])
root.set("number", str(amdt['number']))
root.set("updated", utils.format_datetime(amdt['updated_at']))
make_node = utils.make_node
if amdt.get("amends_bill", None):
make_node(root, "amends", None,
type=govtrack_type_codes[amdt["amends_bill"]["bill_type"]],
number=str(amdt["amends_bill"]["number"]),
sequence=str(amdt["house_number"]) if amdt.get("house_number", None) else "")
elif amdt.get("amends_treaty", None):
make_node(root, "amends", None,
type="treaty",
number=str(amdt["amends_treaty"]["number"]))
make_node(root, "status", amdt['status'], datetime=amdt['status_at'])
if amdt['sponsor'] and amdt['sponsor']['type'] == 'person':
v = amdt['sponsor']['bioguide_id']
if not options.get("govtrack", False):
make_node(root, "sponsor", None, bioguide_id=v)
else:
v = str(utils.translate_legislator_id('bioguide', v, 'govtrack'))
make_node(root, "sponsor", None, id=v)
elif amdt['sponsor'] and amdt['sponsor']['type'] == 'committee':
make_node(root, "sponsor", None, committee=amdt['sponsor']['name'])
else:
make_node(root, "sponsor", None)
make_node(root, "offered", None, datetime=amdt['introduced_at'])
make_node(root, "description", amdt["description"] if amdt["description"] else amdt["purpose"])
if amdt["description"]:
make_node(root, "purpose", amdt["purpose"])
actions = make_node(root, "actions", None)
for action in amdt['actions']:
a = make_node(actions,
action['type'] if action['type'] in ("vote",) else "action",
None,
datetime=action['acted_at'])
if action['type'] == 'vote':
a.set("how", action["how"])
a.set("result", action["result"])
if action.get("roll") != None:
a.set("roll", str(action["roll"]))
if action.get('text'):
make_node(a, "text", action['text'])
if action.get('in_committee'):
make_node(a, "committee", None, name=action['in_committee'])
for cr in action['references']:
make_node(a, "reference", None, ref=cr['reference'], label=cr['type'])
return etree.tostring(root, pretty_print=True)
def create_govtrack_xml(amdt, options):
govtrack_type_codes = {'hr': 'h', 's': 's', 'hres': 'hr', 'sres': 'sr', 'hjres': 'hj', 'sjres': 'sj', 'hconres': 'hc', 'sconres': 'sc'}
root = etree.Element("amendment")
root.set("session", amdt['congress'])
root.set("chamber", amdt['amendment_type'][0])
root.set("number", str(amdt['number']))
root.set("updated", utils.format_datetime(amdt['updated_at']))
make_node = utils.make_node
if amdt.get("amends_bill", None):
make_node(root, "amends", None,
type=govtrack_type_codes[amdt["amends_bill"]["bill_type"]],
number=str(amdt["amends_bill"]["number"]),
sequence=str(amdt["house_number"]) if amdt.get("house_number", None) else "")
elif amdt.get("amends_treaty", None):
make_node(root, "amends", None,
type="treaty",
number=str(amdt["amends_treaty"]["number"]))
make_node(root, "status", amdt['status'], datetime=amdt['status_at'])
if amdt['sponsor'] and amdt['sponsor']['type'] == 'person':
v = amdt['sponsor']['thomas_id']
if not options.get("govtrack", False):
make_node(root, "sponsor", None, thomas_id=v)
else:
v = str(utils.translate_legislator_id('thomas', v, 'govtrack'))
make_node(root, "sponsor", None, id=v)
elif amdt['sponsor'] and amdt['sponsor']['type'] == 'committee':
make_node(root, "sponsor", None, committee=amdt['sponsor']['name'])
else:
make_node(root, "sponsor", None)
make_node(root, "offered", None, datetime=amdt['introduced_at'])
make_node(root, "description", amdt["description"] if amdt["description"] else amdt["purpose"])
if amdt["description"]:
make_node(root, "purpose", amdt["purpose"])
actions = make_node(root, "actions", None)
for action in amdt['actions']:
a = make_node(actions,
action['type'] if action['type'] in ("vote",) else "action",
None,
datetime=action['acted_at'])
if action['type'] == 'vote':
a.set("how", action["how"])
a.set("result", action["result"])
if action.get("roll") != None:
a.set("roll", str(action["roll"]))
if action.get('text'):
make_node(a, "text", action['text'])
if action.get('in_committee'):
make_node(a, "committee", None, name=action['in_committee'])
for cr in action['references']:
make_node(a, "reference", None, ref=cr['reference'], label=cr['type'])
return etree.tostring(root, pretty_print=True)
def make_node(parent, tag, text, **attrs):
if options.get("govtrack", False):
# Rewrite bioguide_id attributes as just id with GovTrack person IDs.
attrs2 = {}
for k, v in attrs.items():
if v:
if k == "bioguide_id":
# remap "bioguide_id" attributes to govtrack "id"
k = "id"
v = str(utils.translate_legislator_id('bioguide', v, 'govtrack'))
if k == "thomas_id":
# remap "thomas_id" attributes to govtrack "id"
k = "id"
v = str(utils.translate_legislator_id('thomas', v, 'govtrack'))
attrs2[k] = v
attrs = attrs2
return utils.make_node(parent, tag, text, **attrs)
def make_node(parent, tag, text, **attrs):
if options.get("govtrack", False):
# Rewrite bioguide_id attributes as just id with GovTrack person IDs.
attrs2 = {}
for k, v in attrs.items():
if v:
if k == "bioguide_id":
# remap "bioguide_id" attributes to govtrack "id"
k = "id"
v = str(utils.translate_legislator_id('bioguide', v, 'govtrack'))
attrs2[k] = v
attrs = attrs2
return utils.make_node(parent, tag, text, **attrs)
def output_vote(vote, options, id_type=None):
logging.info("[%s] Writing to disk..." % vote['vote_id'])
# output JSON - so easy!
utils.write(
json.dumps(vote, sort_keys=True, indent=2, default=utils.format_datetime),
output_for_vote(vote["vote_id"], "json"),
options=options
)
# What kind of IDs are we passed for Members of Congress?
# For current data, we infer from the chamber. For historical data from voteview,
# we're passed the type in id_type, which is set to "bioguide".
if not id_type:
id_type = ("bioguide" if vote["chamber"] == "h" else "lis")
# output XML
root = etree.Element("roll")
root.set("where", "house" if vote['chamber'] == "h" else "senate")
root.set("session", str(vote["congress"]))
root.set("year", str(vote["date"].year))
root.set("roll", str(vote["number"]))
if "voteview" in vote["source_url"]:
root.set("source", "keithpoole")
else:
root.set("source", "house.gov" if vote["chamber"] == "h" else "senate.gov")
root.set("datetime", utils.format_datetime(vote['date']))
root.set("updated", utils.format_datetime(vote['updated_at']))
def get_votes(option):
return len(vote["votes"].get(option, []))
root.set("aye", str(get_votes("Yea") + get_votes("Aye")))
root.set("nay", str(get_votes("Nay") + get_votes("No")))
root.set("nv", str(get_votes("Not Voting")))
root.set("present", str(get_votes("Present")))
utils.make_node(root, "category", vote["category"])
utils.make_node(root, "type", vote["type"])
utils.make_node(root, "question", vote["question"])
utils.make_node(root, "required", vote["requires"])
utils.make_node(root, "result", vote["result"])
if vote.get("bill"):
govtrack_type_codes = {'hr': 'h', 's': 's', 'hres': 'hr', 'sres': 'sr', 'hjres': 'hj', 'sjres': 'sj', 'hconres': 'hc', 'sconres': 'sc'}
utils.make_node(root, "bill", None, session=str(vote["bill"]["congress"]), type=govtrack_type_codes[vote["bill"]["type"]], number=str(vote["bill"]["number"]))
if "amendment" in vote:
n = utils.make_node(root, "amendment", None)
if vote["amendment"]["type"] == "s":
n.set("ref", "regular")
n.set("session", str(vote["congress"]))
n.set("number", "s" + str(vote["amendment"]["number"]))
elif vote["amendment"]["type"] == "h-bill":
n.set("ref", "bill-serial")
n.set("session", str(vote["congress"]))
n.set("number", str(vote["amendment"]["number"]))
# well-known keys for certain vote types: +/-/P/0
option_keys = {"Aye": "+", "Yea": "+", "Nay": "-", "No": "-", "Present": "P", "Not Voting": "0", "Guilty": "+", "Not Guilty": "-" }
# preferred order of output: ayes, nays, present, then not voting, and similarly for guilty/not-guilty
# and handling other options like people's names for votes for the Speaker.
option_sort_order = ('Aye', 'Yea', 'Guilty', 'No', 'Nay', 'Not Guilty', 'OTHER', 'Present', 'Not Voting')
options_list = sorted(vote["votes"].keys(), key=lambda o: option_sort_order.index(o) if o in option_sort_order else option_sort_order.index("OTHER"))
for option in options_list:
if option not in option_keys:
option_keys[option] = option
utils.make_node(root, "option", option, key=option_keys[option])
for option in options_list:
for v in vote["votes"][option]:
n = utils.make_node(root, "voter", None)
if v == "VP":
n.set("id", "0")
n.set("VP", "1")
elif not options.get("govtrack", False):
n.set("id", str(v["id"]))
else:
n.set("id", str(utils.translate_legislator_id(id_type, v["id"], 'govtrack')))
n.set("vote", option_keys[option])
n.set("value", option)
if v != "VP":
n.set("state", v["state"])
if v.get("voteview_votecode_extra") is not None:
n.set("voteview_votecode_extra", v["voteview_votecode_extra"])
xmloutput = etree.tostring(root, pretty_print=True, encoding="utf8")
# mimick two hard line breaks in GovTrack's legacy output to ease running diffs
xmloutput = re.sub('(source=".*?") ', r"\1\n ", xmloutput)
xmloutput = re.sub('(updated=".*?") ', r"\1\n ", xmloutput)
utils.write(
xmloutput,
output_for_vote(vote['vote_id'], "xml"),
options=options
)
def output_vote(vote, options, id_type=None):
logging.info("[%s] Writing to disk..." % vote['vote_id'])
# output JSON - so easy!
utils.write(json.dumps(vote,
sort_keys=True,
indent=2,
default=utils.format_datetime),
output_for_vote(vote["vote_id"], "json"),
options=options)
# What kind of IDs are we passed for Members of Congress?
# For current data, we infer from the chamber. For historical data from voteview,
# we're passed the type in id_type, which is set to "bioguide".
if not id_type:
id_type = ("bioguide" if vote["chamber"] == "h" else "lis")
# output XML
root = etree.Element("roll")
root.set("where", "house" if vote['chamber'] == "h" else "senate")
root.set("session", str(vote["congress"]))
root.set("year", str(vote["date"].year))
root.set("roll", str(vote["number"]))
if "voteview" in vote["source_url"]:
root.set("source", "keithpoole")
else:
root.set("source",
"house.gov" if vote["chamber"] == "h" else "senate.gov")
root.set("datetime", utils.format_datetime(vote['date']))
root.set("updated", utils.format_datetime(vote['updated_at']))
def get_votes(option):
return len(vote["votes"].get(option, []))
root.set("aye", str(get_votes("Yea") + get_votes("Aye")))
root.set("nay", str(get_votes("Nay") + get_votes("No")))
root.set("nv", str(get_votes("Not Voting")))
root.set("present", str(get_votes("Present")))
utils.make_node(root, "category", vote["category"])
utils.make_node(root, "type", vote["type"])
utils.make_node(root, "question", vote["question"])
utils.make_node(root, "required", vote["requires"])
utils.make_node(root, "result", vote["result"])
if vote.get("bill"):
govtrack_type_codes = {
'hr': 'h',
's': 's',
'hres': 'hr',
'sres': 'sr',
'hjres': 'hj',
'sjres': 'sj',
'hconres': 'hc',
'sconres': 'sc'
}
utils.make_node(root,
"bill",
None,
session=str(vote["bill"]["congress"]),
type=govtrack_type_codes[vote["bill"]["type"]],
number=str(vote["bill"]["number"]))
if "amendment" in vote:
n = utils.make_node(root, "amendment", None)
if vote["amendment"]["type"] == "s":
n.set("ref", "regular")
n.set("session", str(vote["congress"]))
n.set("number", "s" + str(vote["amendment"]["number"]))
elif vote["amendment"]["type"] == "h-bill":
n.set("ref", "bill-serial")
n.set("session", str(vote["congress"]))
n.set("number", str(vote["amendment"]["number"]))
# well-known keys for certain vote types: +/-/P/0
option_keys = {
"Aye": "+",
"Yea": "+",
"Nay": "-",
"No": "-",
"Present": "P",
"Not Voting": "0",
"Guilty": "+",
"Not Guilty": "-"
}
# preferred order of output: ayes, nays, present, then not voting, and similarly for guilty/not-guilty
# and handling other options like people's names for votes for the Speaker.
option_sort_order = ('Aye', 'Yea', 'Guilty', 'No', 'Nay', 'Not Guilty',
'OTHER', 'Present', 'Not Voting')
options_list = sorted(
vote["votes"].keys(),
key=lambda o: option_sort_order.index(o)
if o in option_sort_order else option_sort_order.index("OTHER"))
for option in options_list:
if option not in option_keys:
option_keys[option] = option
utils.make_node(root, "option", option, key=option_keys[option])
for option in options_list:
for v in vote["votes"][option]:
# Rep-elect Letlow is included as not voting in the first House vote of the 117th Congress
# where the clerk calls a quorum roll call. But because Letlow had died prior to this date,
# he is not represented in congress-legislators and has no GovTrack-id, and so we cannot
# represent this record in the data.
if isinstance(v, dict) and v["id"] == "L000555" and options.get(
"govtrack", False):
continue
n = utils.make_node(root, "voter", None)
if v == "VP":
n.set("id", "0")
n.set("VP", "1")
elif not options.get("govtrack", False):
n.set("id", str(v["id"]))
else:
n.set(
"id",
str(
utils.translate_legislator_id(id_type, v["id"],
'govtrack')))
n.set("vote", option_keys[option])
n.set("value", option)
if v != "VP":
n.set("state", v["state"])
if v.get("voteview_votecode_extra") is not None:
n.set("voteview_votecode_extra",
v["voteview_votecode_extra"])
xmloutput = etree.tostring(root, pretty_print=True, encoding="unicode")
# mimick two hard line breaks in GovTrack's legacy output to ease running diffs
xmloutput = re.sub('(source=".*?") ', r"\1\n ", xmloutput)
xmloutput = re.sub('(updated=".*?") ', r"\1\n ", xmloutput)
utils.write(xmloutput,
output_for_vote(vote['vote_id'], "xml"),
options=options)
