def scan_committees(abbr):
metadata = db.metadata.find_one({'_id': abbr})
level = metadata['level']
duplicate_sources = defaultdict(int)
report = {'upper_count': 0,
'lower_count': 0,
'joint_count': 0,
'empty_count': 0,
'_updated_today_count': 0,
'_updated_this_month_count': 0,
'_updated_this_year_count': 0,
'_member_count': 0,
'_members_with_leg_id_count': 0,
'sourceless_count': 0,
'unmatched_leg_ids': set(),
}
for com in db.committees.find({'level': level, level: abbr}):
update_common(com, report)
if com['chamber'] == 'upper':
report['upper_count'] += 1
elif com['chamber'] == 'lower':
report['lower_count'] += 1
elif com['chamber'] == 'joint':
report['joint_count'] += 1
# members
if not com['members']:
report['empty_count'] += 1
for member in com['members']:
report['_member_count'] += 1
if member.get('leg_id'):
report['_members_with_leg_id_count'] += 1
else:
report['unmatched_leg_ids'].add((com.get('term', ''),
com['chamber'],
member['name']))
# sources
for source in com['sources']:
duplicate_sources[source['url']] += 1
report['duplicate_sources'] = []
for url, n in duplicate_sources.iteritems():
if n > 1:
report['duplicate_sources'].append(url)
return report
def scan_legislators(abbr):
metadata = db.metadata.find_one({'_id': abbr})
level = metadata['level']
duplicate_sources = defaultdict(int)
report = {'upper_active_count': 0,
'lower_active_count': 0,
'inactive_count': 0,
'_updated_today_count': 0,
'_updated_this_month_count': 0,
'_updated_this_year_count': 0,
'sourceless_count': 0,
}
seats_filled = {'upper': defaultdict(int), 'lower': defaultdict(int)}
for key in checked_keys:
report[key] = 0
for leg in db.legislators.find({'level': level, level: abbr}):
# do common details
update_common(leg, report)
# most checks only apply to active set
if leg.get('active'):
chamber = leg.get('chamber')
if chamber == 'upper':
report['upper_active_count'] += 1
elif chamber == 'lower':
report['lower_active_count'] += 1
else:
# TODO: track these? (executives)
continue
seats_filled[chamber][leg['district']] += 1
# TODO: check seats_filled against districts
for key in checked_keys:
if leg.get(key):
report[key] += 1
else:
report['inactive_count'] += 1
for source in leg['sources']:
duplicate_sources[source['url']] += 1
report['duplicate_sources'] = []
for url, n in duplicate_sources.iteritems():
if n > 1:
report['duplicate_sources'].append(url)
return report
def scan_legislators(abbr):
duplicate_sources = defaultdict(int)
report = {'upper_active_count': 0,
'lower_active_count': 0,
'inactive_count': 0,
'_updated_today_count': 0,
'_updated_this_month_count': 0,
'_updated_this_year_count': 0,
}
for key in checked_keys:
report[key] = 0
# initialize seat counts
district_seats = {'upper': defaultdict(int), 'lower': defaultdict(int)}
for district in db.districts.find({'abbr': abbr}):
district_seats[district['chamber']][district['name']] = \
district['num_seats']
for leg in db.legislators.find({settings.LEVEL_FIELD: abbr}):
# do common details
update_common(leg, report)
# most checks only apply to active set
if leg.get('active'):
chamber = leg.get('chamber')
if chamber == 'upper':
report['upper_active_count'] += 1
elif chamber == 'lower':
report['lower_active_count'] += 1
else:
# TODO: track these? (executives)
continue
# decrement empty seats (if it goes negative, we have too many)
district_seats[chamber][leg['district']] -= 1
for key in checked_keys:
if leg.get(key):
report[key] += 1
else:
report['inactive_count'] += 1
for source in leg['sources']:
duplicate_sources[source['url']] += 1
report['duplicate_sources'] = []
for url, n in duplicate_sources.iteritems():
if n > 1:
report['duplicate_sources'].append(url)
# copy over seat issues into report
report['overfilled_seats'] = []
report['vacant_seats'] = []
for chamber, chamber_seats in district_seats.iteritems():
for seat, count in chamber_seats.iteritems():
if count < 0:
report['overfilled_seats'].append((chamber, seat, -count))
elif count > 0:
report['vacant_seats'].append((chamber, seat, count))
return report
def scan_bills(abbr):
metadata = db.metadata.find_one({'_id': abbr})
level = metadata['level']
duplicate_sources = defaultdict(int)
duplicate_versions = defaultdict(int)
other_actions = defaultdict(int)
uncategorized_subjects = defaultdict(int)
sessions = defaultdict(_bill_report_dict)
for bill in db.bills.find({'level': level, level: abbr}):
session_d = sessions[bill['session']]
# chamber count & bill_types
if bill['chamber'] == 'lower':
session_d['lower_count'] += 1
elif bill['chamber'] == 'upper':
session_d['upper_count'] += 1
for type in bill['type']:
session_d['bill_types'][type] += 1
update_common(bill, session_d)
# actions
last_date = datetime.datetime(1900, 1, 1)
for action in bill['actions']:
date = action['date']
if date < last_date:
session_d['actions_unsorted'].add(bill['_id'])
session_d['action_count'] += 1
for type in action['type']:
session_d['actions_per_type'][type] += 1
if 'other' in action['type']:
other_actions[action['action']] += 1
session_d['actions_per_actor'][action['actor']] += 1
session_d['actions_per_month'][date.strftime('%Y-%m')] += 1
if not bill['actions']:
session_d['actionless_count'] += 1
# sponsors
for sponsor in bill['sponsors']:
session_d['_sponsor_count'] += 1
if sponsor.get('leg_id'):
session_d['_sponsors_with_leg_id_count'] += 1
else:
# keep missing leg_ids
session_d['unmatched_leg_ids'].add(
(term_for_session(abbr, bill['session']), bill['chamber'],
sponsor['name']))
session_d['sponsors_per_type'][sponsor['type']] += 1
if not bill['sponsors']:
session_d['sponsorless_count'] += 1
# votes
for vote in bill['votes']:
session_d['vote_count'] += 1
if vote['passed']:
session_d['_passed_vote_count'] += 1
session_d['votes_per_chamber'][vote['chamber']] += 1
if not vote.get('type'):
logger.warning('vote is missing type on %s' % bill['_id'])
continue
session_d['votes_per_type'][vote.get('type')] += 1
if not vote.get('date'):
logger.warning('vote is missing date on %s' % bill['_id'])
continue
session_d['votes_per_month'][vote['date'].strftime('%Y-%m')] += 1
# roll calls
has_rollcalls = False
for rc in (vote['yes_votes'] + vote['no_votes'] +
vote['other_votes']):
has_rollcalls = True
session_d['_rollcall_count'] += 1
if rc.get('leg_id'):
session_d['_rollcalls_with_leg_id_count'] += 1
else:
# keep missing leg_ids
session_d['unmatched_leg_ids'].add(
(term_for_session(abbr, bill['session']),
vote['chamber'], rc['name']))
# check counts if any rollcalls are present
if (has_rollcalls
and (len(vote['yes_votes']) != vote['yes_count']
or len(vote['no_votes']) != vote['no_count']
or len(vote['other_votes']) != vote['other_count'])):
session_d['bad_vote_counts'].add(bill['_id'])
# subjects
for subj in bill.get('scraped_subjects', []):
uncategorized_subjects[subj] += 1
if bill.get('subjects'):
session_d['_subjects_count'] += 1
for subject in bill['subjects']:
session_d['bills_per_subject'][subject] += 1
# sources
for source in bill['sources']:
duplicate_sources[source['url']] += 1
# versions
if not bill['versions']:
# total num of bills w/o versions
session_d['versionless_count'] += 1
else:
# total num of versions
session_d['version_count'] += len(bill['versions'])
for doc in bill['versions']:
duplicate_versions[doc['url']] += 1
# TODO: add a duplicate documents back in?
dup_version_urls = []
dup_source_urls = []
for url, n in duplicate_versions.iteritems():
if n > 1:
dup_version_urls.append(url)
for url, n in duplicate_sources.iteritems():
if n > 1:
dup_source_urls.append(url)
return {
'duplicate_versions': dup_version_urls,
'duplicate_sources': dup_source_urls,
'other_actions': other_actions.items(),
'uncategorized_subjects': uncategorized_subjects.items(),
'sessions': sessions,
}
def scan_legislators(abbr):
duplicate_sources = defaultdict(int)
report = {
'upper_active_count': 0,
'lower_active_count': 0,
'inactive_count': 0,
'_updated_today_count': 0,
'_updated_this_month_count': 0,
'_updated_this_year_count': 0,
}
for key in checked_keys:
report[key] = 0
# initialize seat counts
district_seats = {'upper': defaultdict(int), 'lower': defaultdict(int)}
for district in db.districts.find({'abbr': abbr}):
district_seats[district['chamber']][district['name']] = \
district['num_seats']
for leg in db.legislators.find({settings.LEVEL_FIELD: abbr}):
# do common details
update_common(leg, report)
# most checks only apply to active set
if leg.get('active'):
chamber = leg.get('chamber')
if chamber == 'upper':
report['upper_active_count'] += 1
elif chamber == 'lower':
report['lower_active_count'] += 1
else:
# TODO: track these? (executives)
continue
# decrement empty seats (if it goes negative, we have too many)
district_seats[chamber][leg['district']] -= 1
for key in checked_keys:
if leg.get(key):
report[key] += 1
else:
report['inactive_count'] += 1
for source in leg['sources']:
duplicate_sources[source['url']] += 1
report['duplicate_sources'] = []
for url, n in duplicate_sources.items():
if n > 1:
report['duplicate_sources'].append(url)
# copy over seat issues into report
report['overfilled_seats'] = []
report['vacant_seats'] = []
for chamber, chamber_seats in district_seats.items():
for seat, count in chamber_seats.items():
if count < 0:
report['overfilled_seats'].append((chamber, seat, -count))
elif count > 0:
report['vacant_seats'].append((chamber, seat, count))
return report
def scan_bills(abbr):
duplicate_sources = defaultdict(int)
duplicate_versions = defaultdict(int)
other_actions = defaultdict(int)
uncategorized_subjects = defaultdict(int)
sessions = defaultdict(_bill_report_dict)
# load exception data into sets of ids indexed by exception type
quality_exceptions = get_quality_exceptions(abbr)
for bill in db.bills.find({settings.LEVEL_FIELD: abbr}):
session_d = sessions[bill['session']]
# chamber count & bill_types
if bill['chamber'] == 'lower':
session_d['lower_count'] += 1
elif bill['chamber'] == 'upper':
session_d['upper_count'] += 1
for type in bill['type']:
session_d['bill_types'][type] += 1
update_common(bill, session_d)
# actions
last_date = datetime.datetime(1900, 1, 1)
for action in bill['actions']:
date = action['date']
if date < last_date:
session_d['actions_unsorted'].add(bill['_id'])
session_d['action_count'] += 1
for type in action['type']:
session_d['actions_per_type'][type] += 1
if 'other' in action['type']:
other_actions[action['action']] += 1
session_d['actions_per_actor'][action['actor']] += 1
session_d['actions_per_month'][date.strftime('%Y-%m')] += 1
# handle no_actions bills
if not bill['actions']:
if bill['_id'] not in quality_exceptions['bills:no_actions']:
session_d['actionless_count'] += 1
else:
quality_exceptions['bills:no_actions'].remove(bill['_id'])
# sponsors
for sponsor in bill['sponsors']:
session_d['_sponsor_count'] += 1
if sponsor.get('leg_id') or sponsor.get('committee_id'):
session_d['_sponsors_with_id_count'] += 1
else:
# keep list of unmatched sponsors
session_d['unmatched_sponsors'].add(
(term_for_session(abbr, bill['session']), bill['chamber'],
sponsor['name'])
)
session_d['sponsors_per_type'][sponsor['type']] += 1
# handle no sponsors bills
if not bill['sponsors']:
if bill['_id'] not in quality_exceptions['bills:no_sponsors']:
session_d['sponsorless_count'] += 1
else:
quality_exceptions['bills:no_sponsors'].remove(bill['_id'])
# subjects
for subj in bill.get('scraped_subjects', []):
uncategorized_subjects[subj] += 1
if bill.get('subjects'):
session_d['_subjects_count'] += 1
for subject in bill['subjects']:
session_d['bills_per_subject'][subject] += 1
# sources
for source in bill['sources']:
duplicate_sources[source['url']] += 1
# versions
if not bill['versions']:
# total num of bills w/o versions
if bill['_id'] not in quality_exceptions['bills:no_versions']:
session_d['versionless_count'] += 1
else:
quality_exceptions['bills:no_versions'].remove(bill['_id'])
else:
# total num of versions
session_d['version_count'] += len(bill['versions'])
for doc in bill['versions']:
duplicate_versions[doc['url']] += 1
# TODO: add duplicate document detection back in?
# Check for progress meter gaps.
progress_meter_gaps = session_d['progress_meter_gaps']
action_dates = bill['action_dates']
bill_chamber = bill['chamber']
other_chamber = dict(lower='upper', upper='lower')[bill_chamber]
# Check for bills that were signed but didn't pass both chambers.
if bill['type'] == 'bill':
if action_dates['signed']:
if not action_dates['passed_upper']:
progress_meter_gaps.add(bill['_id'])
elif not action_dates['passed_lower']:
progress_meter_gaps.add(bill['_id'])
else:
# Check for nonbills that were signed but didn't pass their
# house of origin.
if action_dates['signed']:
if not action_dates['passed_' + bill_chamber]:
progress_meter_gaps.add(bill['_id'])
if action_dates['passed_' + other_chamber]:
if not action_dates['passed_' + bill_chamber]:
progress_meter_gaps.add(bill['_id'])
dup_version_urls = []
dup_source_urls = []
for url, n in duplicate_versions.items():
if n > 1:
dup_version_urls.append(url)
for url, n in duplicate_sources.items():
if n > 1:
dup_source_urls.append(url)
# do logging of unnecessary exceptions
for qe_type, qes in quality_exceptions.items():
if qes:
logger.warning('unnecessary {0} exceptions for {1} bills: \n {2}'
.format(qe_type, len(qes), '\n '.join(qes)))
return {'duplicate_versions': dup_version_urls,
'duplicate_sources': dup_source_urls,
'other_actions': other_actions.items(),
'uncategorized_subjects': uncategorized_subjects.items(),
'sessions': sessions,
'progress_meter_gaps': []
}
def scan_bills(abbr):
duplicate_sources = defaultdict(int)
duplicate_versions = defaultdict(int)
other_actions = defaultdict(int)
uncategorized_subjects = defaultdict(int)
sessions = defaultdict(_bill_report_dict)
# load exception data into sets of ids indexed by exception type
quality_exceptions = get_quality_exceptions(abbr)
for bill in db.bills.find({settings.LEVEL_FIELD: abbr}):
session_d = sessions[bill['session']]
# chamber count & bill_types
if bill['chamber'] == 'lower':
session_d['lower_count'] += 1
elif bill['chamber'] == 'upper':
session_d['upper_count'] += 1
for type in bill['type']:
session_d['bill_types'][type] += 1
update_common(bill, session_d)
# actions
last_date = datetime.datetime(1900, 1, 1)
for action in bill['actions']:
date = action['date']
if not date:
continue
if date < last_date:
session_d['actions_unsorted'].add(bill['_id'])
session_d['action_count'] += 1
for type in action['type']:
session_d['actions_per_type'][type] += 1
if 'other' in action['type']:
other_actions[action['action']] += 1
session_d['actions_per_actor'][action['actor']] += 1
session_d['actions_per_month'][date.strftime('%Y-%m')] += 1
# handle no_actions bills
if not bill['actions']:
if bill['_id'] not in quality_exceptions['bills:no_actions']:
session_d['actionless_count'] += 1
else:
quality_exceptions['bills:no_actions'].remove(bill['_id'])
# sponsors
for sponsor in bill['sponsors']:
session_d['_sponsor_count'] += 1
if sponsor.get('leg_id') or sponsor.get('committee_id'):
session_d['_sponsors_with_id_count'] += 1
else:
# keep list of unmatched sponsors
session_d['unmatched_sponsors'].add(
(term_for_session(abbr, bill['session']), bill['chamber'],
sponsor['name'])
)
session_d['sponsors_per_type'][sponsor['type']] += 1
# handle no sponsors bills
if not bill['sponsors']:
if bill['_id'] not in quality_exceptions['bills:no_sponsors']:
session_d['sponsorless_count'] += 1
else:
quality_exceptions['bills:no_sponsors'].remove(bill['_id'])
# subjects
for subj in bill.get('scraped_subjects', []):
uncategorized_subjects[subj] += 1
if bill.get('subjects'):
session_d['_subjects_count'] += 1
for subject in bill['subjects']:
session_d['bills_per_subject'][subject] += 1
# sources
for source in bill['sources']:
duplicate_sources[source['url']] += 1
# versions
if not bill['versions']:
# total num of bills w/o versions
if bill['_id'] not in quality_exceptions['bills:no_versions']:
session_d['versionless_count'] += 1
else:
quality_exceptions['bills:no_versions'].remove(bill['_id'])
else:
# total num of versions
session_d['version_count'] += len(bill['versions'])
for doc in bill['versions']:
duplicate_versions[doc['url']] += 1
# TODO: add duplicate document detection back in?
# Check for progress meter gaps.
progress_meter_gaps = session_d['progress_meter_gaps']
action_dates = bill['action_dates']
bill_chamber = bill['chamber']
other_chamber = dict(lower='upper', upper='lower')[bill_chamber]
# Check for bills that were signed but didn't pass both chambers.
if bill['type'] == 'bill':
if action_dates['signed']:
if not action_dates['passed_upper']:
progress_meter_gaps.add(bill['_id'])
elif not action_dates['passed_lower']:
progress_meter_gaps.add(bill['_id'])
else:
# Check for nonbills that were signed but didn't pass their
# house of origin.
if action_dates['signed']:
if not action_dates['passed_' + bill_chamber]:
progress_meter_gaps.add(bill['_id'])
if action_dates['passed_' + other_chamber]:
if not action_dates['passed_' + bill_chamber]:
progress_meter_gaps.add(bill['_id'])
dup_version_urls = []
dup_source_urls = []
for url, n in duplicate_versions.items():
if n > 1:
dup_version_urls.append(url)
for url, n in duplicate_sources.items():
if n > 1:
dup_source_urls.append(url)
# do logging of unnecessary exceptions
for qe_type, qes in quality_exceptions.items():
if qes:
logger.warning('unnecessary {0} exceptions for {1} bills: \n {2}'
.format(qe_type, len(qes), '\n '.join(qes)))
return {'duplicate_versions': dup_version_urls,
'duplicate_sources': dup_source_urls,
'other_actions': other_actions.items(),
'uncategorized_subjects': uncategorized_subjects.items(),
'sessions': sessions,
'progress_meter_gaps': []
}
def scan_bills(abbr):
metadata = db.metadata.find_one({'_id': abbr})
level = metadata['level']
duplicate_sources = defaultdict(int)
duplicate_versions = defaultdict(int)
other_actions = defaultdict(int)
uncategorized_subjects = defaultdict(int)
sessions = defaultdict(_bill_report_dict)
for bill in db.bills.find({'level': level, level: abbr}):
session_d = sessions[bill['session']]
# chamber count & bill_types
if bill['chamber'] == 'lower':
session_d['lower_count'] += 1
elif bill['chamber'] == 'upper':
session_d['upper_count'] += 1
for type in bill['type']:
session_d['bill_types'][type] += 1
update_common(bill, session_d)
# actions
last_date = datetime.datetime(1900, 1, 1)
for action in bill['actions']:
date = action['date']
if date < last_date:
session_d['actions_unsorted'].add(bill['_id'])
session_d['action_count'] += 1
for type in action['type']:
session_d['actions_per_type'][type] += 1
if 'other' in action['type']:
other_actions[action['action']] += 1
session_d['actions_per_actor'][action['actor']] += 1
session_d['actions_per_month'][date.strftime('%Y-%m')] += 1
if not bill['actions']:
session_d['actionless_count'] += 1
# sponsors
for sponsor in bill['sponsors']:
session_d['_sponsor_count'] += 1
if sponsor.get('leg_id'):
session_d['_sponsors_with_leg_id_count'] += 1
else:
# keep missing leg_ids
session_d['unmatched_leg_ids'].add(
(term_for_session(abbr, bill['session']), bill['chamber'],
sponsor['name'])
)
session_d['sponsors_per_type'][sponsor['type']] += 1
if not bill['sponsors']:
session_d['sponsorless_count'] += 1
# votes
for vote in bill['votes']:
session_d['vote_count'] += 1
if vote['passed']:
session_d['_passed_vote_count'] += 1
session_d['votes_per_chamber'][vote['chamber']] += 1
if not vote.get('type'):
logger.warning('vote is missing type on %s' % bill['_id'])
continue
session_d['votes_per_type'][vote.get('type')] += 1
if not vote.get('date'):
logger.warning('vote is missing date on %s' % bill['_id'])
continue
session_d['votes_per_month'][vote['date'].strftime('%Y-%m')] += 1
# roll calls
has_rollcalls = False
for rc in (vote['yes_votes'] + vote['no_votes'] +
vote['other_votes']):
has_rollcalls = True
session_d['_rollcall_count'] += 1
if rc.get('leg_id'):
session_d['_rollcalls_with_leg_id_count'] += 1
else:
# keep missing leg_ids
session_d['unmatched_leg_ids'].add(
(term_for_session(abbr, bill['session']),
vote['chamber'],
rc['name'])
)
# check counts if any rollcalls are present
if (has_rollcalls and
(len(vote['yes_votes']) != vote['yes_count'] or
len(vote['no_votes']) != vote['no_count'] or
len(vote['other_votes']) != vote['other_count'])):
session_d['bad_vote_counts'].add(bill['_id'])
# subjects
for subj in bill.get('scraped_subjects', []):
uncategorized_subjects[subj] += 1
if bill.get('subjects'):
session_d['_subjects_count'] += 1
for subject in bill['subjects']:
session_d['bills_per_subject'][subject] += 1
# sources
for source in bill['sources']:
duplicate_sources[source['url']] += 1
# versions
if not bill['versions']:
# total num of bills w/o versions
session_d['versionless_count'] += 1
else:
# total num of versions
session_d['version_count'] += len(bill['versions'])
for doc in bill['versions']:
duplicate_versions[doc['url']] += 1
# TODO: add a duplicate documents back in?
dup_version_urls = []
dup_source_urls = []
for url, n in duplicate_versions.iteritems():
if n > 1:
dup_version_urls.append(url)
for url, n in duplicate_sources.iteritems():
if n > 1:
dup_source_urls.append(url)
return {'duplicate_versions': dup_version_urls,
'duplicate_sources': dup_source_urls,
'other_actions': other_actions.items(),
'uncategorized_subjects': uncategorized_subjects.items(),
'sessions': sessions,
}
def scan_bills(abbr):
duplicate_sources = defaultdict(int)
duplicate_versions = defaultdict(int)
other_actions = defaultdict(int)
uncategorized_subjects = defaultdict(int)
sessions = defaultdict(_bill_report_dict)
# load exception data into sets of ids indexed by exception type
quality_exceptions = get_quality_exceptions(abbr)
for bill in db.bills.find({settings.LEVEL_FIELD: abbr}):
session_d = sessions[bill["session"]]
# chamber count & bill_types
if bill["chamber"] == "lower":
session_d["lower_count"] += 1
elif bill["chamber"] == "upper":
session_d["upper_count"] += 1
for type in bill["type"]:
session_d["bill_types"][type] += 1
update_common(bill, session_d)
# actions
last_date = datetime.datetime(1900, 1, 1)
for action in bill["actions"]:
date = action["date"]
if date < last_date:
session_d["actions_unsorted"].add(bill["_id"])
session_d["action_count"] += 1
for type in action["type"]:
session_d["actions_per_type"][type] += 1
if "other" in action["type"]:
other_actions[action["action"]] += 1
session_d["actions_per_actor"][action["actor"]] += 1
session_d["actions_per_month"][date.strftime("%Y-%m")] += 1
# handle no_actions bills
if not bill["actions"]:
if bill["_id"] not in quality_exceptions["bills:no_actions"]:
session_d["actionless_count"] += 1
else:
quality_exceptions["bills:no_actions"].remove(bill["_id"])
# sponsors
for sponsor in bill["sponsors"]:
session_d["_sponsor_count"] += 1
if sponsor.get("leg_id") or sponsor.get("committee_id"):
session_d["_sponsors_with_id_count"] += 1
else:
# keep list of unmatched sponsors
session_d["unmatched_sponsors"].add(
(term_for_session(abbr, bill["session"]), bill["chamber"], sponsor["name"])
)
session_d["sponsors_per_type"][sponsor["type"]] += 1
# handle no sponsors bills
if not bill["sponsors"]:
if bill["_id"] not in quality_exceptions["bills:no_sponsors"]:
session_d["sponsorless_count"] += 1
else:
quality_exceptions["bills:no_sponsors"].remove(bill["_id"])
# subjects
for subj in bill.get("scraped_subjects", []):
uncategorized_subjects[subj] += 1
if bill.get("subjects"):
session_d["_subjects_count"] += 1
for subject in bill["subjects"]:
session_d["bills_per_subject"][subject] += 1
# sources
for source in bill["sources"]:
duplicate_sources[source["url"]] += 1
# versions
if not bill["versions"]:
# total num of bills w/o versions
if bill["_id"] not in quality_exceptions["bills:no_versions"]:
session_d["versionless_count"] += 1
else:
quality_exceptions["bills:no_versions"].remove(bill["_id"])
else:
# total num of versions
session_d["version_count"] += len(bill["versions"])
for doc in bill["versions"]:
duplicate_versions[doc["url"]] += 1
# TODO: add duplicate document detection back in?
dup_version_urls = []
dup_source_urls = []
for url, n in duplicate_versions.iteritems():
if n > 1:
dup_version_urls.append(url)
for url, n in duplicate_sources.iteritems():
if n > 1:
dup_source_urls.append(url)
# do logging of unnecessary exceptions
for qe_type, qes in quality_exceptions.iteritems():
if qes:
logger.warning(
"unnecessary {0} exceptions for {1} bills: \n {2}".format(qe_type, len(qes), "\n ".join(qes))
)
return {
"duplicate_versions": dup_version_urls,
"duplicate_sources": dup_source_urls,
"other_actions": other_actions.items(),
"uncategorized_subjects": uncategorized_subjects.items(),
"sessions": sessions,
}
