def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldMerger({'bill_type_raw': ['bill_name']},
lambda x: re.sub(r'[^A-Z]*', '', x),
keep_fields=True),
FieldMerger({'bill_type': ['bill_type_raw']},
lambda x: self.bill_type_map.get(x, None),
keep_fields=True),
FieldMerger({'bill_no': ['bill_name']},
lambda x: self.digits.match(x).groups()[0]
if x and self.digits.match(x) else None,
keep_fields=True),
NoneFilter(),
IssueFilter(),
UnicodeFilter(),
CountEmitter(every=20000, log=self.log),
LoaderEmitter(BillLoader(
source=self.inpath,
description='load from denormalized CSVs',
imported_by="loadlobbying (%s)" %
os.getenv('LOGNAME', 'unknown'),
log=self.log,
),
commit_every=1),
)
def lobbying_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
UnicodeFilter(),
FieldRemover('Source'),
FieldMerger({'registrant_name': ('Registrant', 'RegistrantRaw')}, name_proc),
FieldMerger({'registrant_is_firm': ('IsFirm',)}, yn_proc),
FieldMerger({'client_name': ('Client', 'Client_raw')}, name_proc),
FieldMerger({'amount': ('Amount',)}, lambda x: float(x or 0)),
FieldMerger({'affiliate': ('Affiliate',)}, yn_proc),
FieldMerger({'filing_included_nsfs': ('IncludeNSFS',)}, yn_proc),
FieldMerger({'include_in_industry_totals': ('Ind',)}, yn_proc),
FieldMerger({'use': ('Use',)}, yn_proc),
FieldRenamer({
'transaction_id': 'Uniqid',
'transaction_type': 'Type',
'transaction_type_desc': 'TypeLong',
'year': 'Year',
'client_category': 'Catcode',
'client_parent_name': 'Ultorg',
'filing_type': 'Self',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def lobbying_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
UnicodeFilter(),
FieldRemover('Source'),
FieldMerger({'registrant_name': ('Registrant', 'RegistrantRaw')},
name_proc),
FieldMerger({'registrant_is_firm': ('IsFirm', )}, yn_proc),
FieldMerger({'client_name': ('Client', 'Client_raw')}, name_proc),
FieldMerger({'amount': ('Amount', )}, lambda x: float(x or 0)),
FieldMerger({'affiliate': ('Affiliate', )}, yn_proc),
FieldMerger({'filing_included_nsfs': ('IncludeNSFS', )}, yn_proc),
FieldMerger({'include_in_industry_totals': ('Ind', )}, yn_proc),
FieldMerger({'use': ('Use', )}, yn_proc),
FieldRenamer({
'transaction_id': 'Uniqid',
'transaction_type': 'Type',
'transaction_type_desc': 'TypeLong',
'year': 'Year',
'client_category': 'Catcode',
'client_parent_name': 'Ultorg',
'filing_type': 'Self',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def load_cpi_areas():
headers = ('area_code','area_name','display_level','selectable','sort_sequence')
url = "ftp://ftp.bls.gov/pub/time.series/cu/cu.area"
run_recipe(
CSVSource(utils.RemoteFile(url), delimiter='\t'),
CSVEmitter(open('bls_areas.csv', 'w'), headers)
)
def load_schools():
for index, filename in enumerate(("sc091aai.csv", "sc091akn.csv", "sc091aow.csv")):
run_recipe(
CSVSource(open(os.path.join(NCES_ROOT, filename))),
FieldRemover(("mzip409", "member09", "phone09", "ulocal09", "type09", "level09", "status09")),
FieldRenamer(
{
"school_id": "ncessch",
"name": "schnam09",
"street": "mstree09",
"city": "mcity09",
"state": "mstate09",
"zipcode": "mzip09",
"grade_low": "gslo09",
"grade_high": "gshi09",
}
),
FieldAdder("latitude", None),
FieldAdder("longitude", None),
FieldAdder("codes", None),
MSACoder(),
SqliteEmitter(DB, "nces_schools", fieldnames=HEADERS),
# DebugEmitter(),
CountEmitter(every=100),
)
def load_schools():
for filename in ('sc091aai.csv','sc091akn.csv','sc091aow.csv'):
run_recipe(
CSVSource(open(os.path.join(NCES_ROOT, filename))),
FieldRemover((
'mzip409',
'member09',
'phone09',
'ulocal09',
'type09',
'level09',
'status09',
)),
FieldRenamer({
'school_id': 'ncessch',
'name': 'schnam09',
'street': 'mstree09',
'city': 'mcity09',
'state': 'mstate09',
'zipcode': 'mzip09',
'grade_low': 'gslo09',
'grade_high': 'gshi09',
}),
FieldAdder('code', None),
GeoCoder(),
SqliteEmitter(DB, 'nces_schools', fieldnames=HEADERS),
#DebugEmitter(),
)
def load_occupations():
run_recipe(
K2LocationSource(),
MySQLOccupationEmitter('root', '', 'k2'),
#emitters.DebugEmitter(),
error_stream = emitters.DebugEmitter(),
)
def load_zipcodes():
headers = (
'country_code','postal_code','name',
'state_name','state_code',
'county_name','county_code',
'community_name','community_code',
'latitude','longitude','accuracy'
)
state_histogram = Histogram('state_code')
state_histogram.label_length = 2
csv_path = settings.dataset_path('default', filename='zipcodes.txt')
run_recipe(
CSVSource(open(csv_path), delimiter="\t", fieldnames=headers),
FieldKeeper(('postal_code','name','state_code','latitude','longitude')),
FieldModifier(('latitude','longitude'), float),
FieldMerger({'latlng': ('latitude', 'longitude')}, lambda lat, lng: (lat, lng)),
#MongoZipEmitter(),
#DebugEmitter(),
state_histogram,
)
return str(state_histogram)
def process_sopr_filing(sopr_xml_file):
from sunlightapi import live_settings as DJ_SETTINGS
DJ_APPLABEL = 'lobbyists'
saucebrush.run_recipe(lobbyists.parse_filings(sopr_xml_file),
# flatten non-list dictionaries & clean up some fields
DictFlattener(['filing', 'client', 'registrant']),
FieldRemover(['govt_entities', 'affiliated_orgs', 'foreign_entities',
'client_state_or_local_gov', 'client_status',
'filing_affiliated_orgs_url']),
FieldRenamer({'filing_date': 'filing_filing_date'}),
# process names & dates
FieldAdder('client_contact_name', ''),
NameCleaner('client_contact_name', prefix='client_contact_', nomatch_name='client_raw_contact_name'),
FieldModifier('filing_date', lambda x: x.split('.')[0]),
DateCleaner('filing_date', from_format='%Y-%m-%dT%H:%M:%S', to_format='%Y-%m-%d'),
# flatten lists
Flattener(['issues', 'lobbyists']),
FieldCopier({'issues.filing_id': 'filing_id',
'lobbyists.filing_id': 'filing_id'}),
# handle lists
saucebrush.filters.Splitter({
'issues':[DjangoModelEmitter(DJ_SETTINGS, DJ_APPLABEL, 'issue')],
'lobbyists':[FieldRemover(['indicator', 'status']),
NameCleaner('name', nomatch_name='raw_name'),
Unique(), # remove some duplicate lobbyists on a form
DjangoModelEmitter(DJ_SETTINGS, DJ_APPLABEL, 'lobbyist')
],
}),
FieldRemover(['issues', 'lobbyists']),
DjangoModelEmitter(DJ_SETTINGS, DJ_APPLABEL, 'filing')
)
def load_locations():
conn = MySQLdb.connect(
user=settings.MYSQL_USER,
passwd=settings.MYSQL_PASS,
db=settings.MYSQL_DATABASE,
host=settings.MYSQL_HOST,
port=settings.MYSQL_PORT,
)
cursor = conn.cursor()
cursor.execute("""DELETE FROM occupation_category""")
cursor.execute("""DELETE FROM occupation""")
cursor.close()
path = settings.dataset_path(None, filename='occupations.csv')
run_recipe(
sources.CSVSource(open(path)),
ValidOccupationFilter(),
CategoryIDFilter(),
CategoryEmitter(conn),
OccupationEmitter(conn),
#emitters.DebugEmitter(),
error_stream = emitters.DebugEmitter(),
)
conn.commit()
conn.close()
def code_a5():
run_recipe(
sources.CSVSource(open("/Users/Jeremy/Downloads/A5.csv")),
filters.FieldAdder("code", None),
MSAFilter(),
emitters.CSVEmitter(open("/Users/Jeremy/Downloads/A5.coded.csv", "w"), fieldnames=FIELD_NAMES),
)
def load_locations():
run_recipe(
K2LocationSource(),
DataFilter(),
filters.FieldRemover(('_id','ffiec','geo','oes','naccrra','nces','rpp_local','rpp_state')),
MySQLLocationEmitter('root', '', 'k2'),
#emitters.DebugEmitter(),
error_stream = emitters.DebugEmitter(),
)
def load_items():
path = os.path.join(CPI_ROOT, "cu.item")
headers = ('item_code','item_name','display_level','selectable','sort_sequence')
run_recipe(
csv.DictReader(local_file(path), delimiter='\t'),
#SqliteEmitter(DB, 'cpi_items', fieldnames=headers),
DebugEmitter(),
)
def load_ffiec():
run_recipe(
ffiec_iter(),
FieldModifier(('low','high'), float),
FieldMerger({'avg': ('low','high')}, lambda x, y: (x + y) / 2, keep_fields=True),
#SqliteEmitter(DB, 'ffiec_incomes', fieldnames=HEADERS),
GenericMSAFilter(),
MongoEmitter(),
#DebugEmitter(),
)
def load_areas():
path = os.path.join(CPI_ROOT, "cu.area")
headers = ('area_code','area_name','msa_code','display_level','selectable','sort_sequence')
run_recipe(
csv.DictReader(local_file(path), delimiter='\t'),
#FieldAdder('msa_code', None),
#MSAFilter(),
#SqliteEmitter(DB, 'cpi_areas', fieldnames=headers),
DebugEmitter(),
)
def agency_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldAdder('id', ''),
FieldRenamer({
'transaction': 'UniqID',
'agency_name': 'Agency',
'agency_ext_id': 'AgencyID',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def agency_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldAdder('id', ''),
FieldRenamer({
'transaction': 'UniqID',
'agency_name': 'Agency',
'agency_ext_id': 'AgencyID',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def bills_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldAdder('id', ''),
FieldRenamer({
'bill_id': 'B_ID',
'issue': 'SI_ID',
'congress_no': 'CongNo',
'bill_name': 'Bill_Name',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def bills_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldAdder('id', ''),
FieldRenamer({
'bill_id': 'B_ID',
'issue': 'SI_ID',
'congress_no': 'CongNo',
'bill_name': 'Bill_Name',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def load_locations():
csv_path = settings.datasource_path('locations', filename='locations.csv')
# load locations into mongodb
run_recipe(
CSVSource(open(csv_path)),
FieldRemover('points'),
MSAFilter(),
GeoJSONTestFilter(),
CensusTestFilter(),
GeoFilter(),
DebugEmitter(),
#MongoDBEmitter(settings.MONGO_DATABASE, 'locations'),
)
def issue_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldRenamer({
'id': 'SI_ID',
'transaction': 'UniqID',
'general_issue_code': 'IssueID',
'general_issue': 'Issue',
'specific_issue': 'SpecIssue',
'year': 'Year',
}),
FieldModifier(('general_issue', 'specific_issue'), lambda x: x.replace('\n', ' ')),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldRenamer(self.field_map),
FieldRemover('committee_fec_id committee_name report_year report_type is_amendment start_date end_date reporting_period_amount_all semi_annual_amount_all'.split()),
BundleFilter(),
#FieldModifier('file_num', lambda x: Bundle.objects.get(pk=x)),
# Convert any stray floats to integers
FieldModifier('amount semi_annual_amount'.split(), \
lambda x: int(round(float(x))) if x else None),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=500),
#DebugEmitter(),
DjangoModelEmitter('settings', LobbyistBundle)
)
def calculate_average(locs):
def locsource():
for loc in locs:
if 'ffiec' in loc and 'diff' in loc['ffiec']:
yield loc['ffiec']
sd = stats.StandardDeviation('diff')
run_recipe(
locsource(),
sd,
)
print "Average: %s" % sd.average()
print "stddev: %s" % sd.value()[0]
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldRenamer(self.field_map),
FieldRemover('committee_fec_id committee_name report_year report_type is_amendment start_date end_date reporting_period_amount_all semi_annual_amount_all'.split()),
BundleFilter(),
#FieldModifier('file_num', lambda x: Bundle.objects.get(pk=x)),
# Convert any stray floats to integers
FieldModifier('amount semi_annual_amount'.split(), \
lambda x: int(round(float(x))) if x else None),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=500),
#DebugEmitter(),
SimpleDjangoModelEmitter(LobbyistBundle)
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldModifier('year', lambda x: int(x) if x else None),
FieldRenamer({'transaction_id': 'transaction'}),
NoneFilter(),
TRANSACTION_FILTER,
UnicodeFilter(),
CountEmitter(every=10000, log=self.log),
LoaderEmitter(AgencyLoader(
source=self.inpath,
description='load from denormalized CSVs',
imported_by="loadlobbying (%s)" % os.getenv('LOGNAME', 'unknown'),
log=self.log,
), commit_every=100),
)
def lobbyist_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldAdder('id', ''),
FieldMerger({'lobbyist_name': ('Lobbyist', 'Lobbyist_raw')}, name_proc),
FieldMerger({'member_of_congress': ('FormerCongMem',)}, yn_proc),
FieldRenamer({
'transaction': 'Uniqid',
'year': 'Year',
'lobbyist_ext_id': 'LobbyistID',
'candidate_ext_id': 'CID',
'government_position': 'OfficalPos',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def issue_handler(inpath, outpath, infields, outfields):
run_recipe(
VerifiedCSVSource(open(inpath, 'r'), fieldnames=infields, quotechar='|'),
FieldCountValidator(len(FILE_TYPES['lob_issue'])),
CSVFieldVerifier(),
FieldRenamer({
'id': 'SI_ID',
'transaction': 'UniqID',
'general_issue_code': 'IssueID',
'general_issue': 'Issue',
'specific_issue': 'SpecIssue',
'year': 'Year',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldMerger({'bill_type_raw': ['bill_name']}, lambda x: re.sub(r'[^A-Z]*', '', x), keep_fields=True),
FieldMerger({'bill_type': ['bill_type_raw']}, lambda x: self.bill_type_map.get(x, None), keep_fields=True),
FieldMerger({'bill_no': ['bill_name']}, lambda x: self.digits.match(x).groups()[0] if x and self.digits.match(x) else None, keep_fields=True),
NoneFilter(),
IssueFilter(),
UnicodeFilter(),
CountEmitter(every=20000, log=self.log),
LoaderEmitter(BillLoader(
source=self.inpath,
description='load from denormalized CSVs',
imported_by="loadlobbying (%s)" % os.getenv('LOGNAME', 'unknown'),
log=self.log,
), commit_every=1),
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldModifier('year', lambda x: int(x) if x else None),
FieldRenamer({'transaction_id': 'transaction'}),
NoneFilter(),
TRANSACTION_FILTER,
UnicodeFilter(),
CountEmitter(every=10000, log=self.log),
LoaderEmitter(AgencyLoader(
source=self.inpath,
description='load from denormalized CSVs',
imported_by="loadlobbying (%s)" %
os.getenv('LOGNAME', 'unknown'),
log=self.log,
),
commit_every=100),
)
def lobbyist_handler(inpath, outpath, infields, outfields):
run_recipe(
CSVSource(open(inpath), fieldnames=infields, quotechar='|'),
FieldAdder('id', ''),
FieldMerger({'lobbyist_name': ('Lobbyist', 'Lobbyist_raw')},
name_proc),
FieldMerger({'member_of_congress': ('FormerCongMem', )}, yn_proc),
FieldRenamer({
'transaction': 'Uniqid',
'year': 'Year',
'lobbyist_ext_id': 'LobbyistID',
'candidate_ext_id': 'CID',
'government_position': 'OfficalPos',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldModifier('year', lambda x: int(x) if x else None),
FieldModifier('amount', lambda x: Decimal(x) if x else None),
FieldModifier((
'affiliate','filing_included_nsfs','include_in_industry_totals',
'registrant_is_firm','use'), lambda x: x == 'True'),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=20000, log=self.log),
LoaderEmitter(LobbyingLoader(
source=self.inpath,
description='load from denormalized CSVs',
imported_by="loadlobbying (%s)" % os.getenv('LOGNAME', 'unknown'),
log=self.log,
)),
)
def issue_handler(inpath, outpath, infields, outfields):
run_recipe(
VerifiedCSVSource(open(inpath, 'r'),
fieldnames=infields,
quotechar='|'),
FieldCountValidator(len(FILE_TYPES['lob_issue'])),
CSVFieldVerifier(),
FieldRenamer({
'id': 'SI_ID',
'transaction': 'UniqID',
'general_issue_code': 'IssueID',
'general_issue': 'Issue',
'specific_issue': 'SpecIssue',
'year': 'Year',
}),
#DebugEmitter(),
CSVEmitter(open(outpath, 'w'), fieldnames=outfields),
)
def load_naccrra():
csv_path = os.path.join(settings.dataset_path('default'), 'childcarecosts.csv')
run_recipe(
sources.CSVSource(open(csv_path)),
filters.FieldRenamer({
'state': 'State',
'family_infant': 'Family-Infant',
'family_4': 'Family-4-Year-Old',
'family_school': 'Family-School-Age',
'center_infant': 'Center-Infant',
'center_4': 'Center-4-Year-Old',
'center_school': 'Center-School-Age',
}),
MongoNACCRRAEmitter(),
emitters.CountEmitter(),
#emitters.DebugEmitter(),
error_stream = emitters.DebugEmitter(),
)
def process_record(self, record):
occ = record['occupation']
stats_filters = {}
for fieldname in STATS_FIELDS:
stats_filters[fieldname] = stats.StandardDeviation(fieldname)
run_recipe(
sources.SqliteSource(db_path, """SELECT * FROM locations WHERE occupation = ?""", (occ,)),
filters.FieldModifier(STATS_FIELDS, to_float),
Recipe(*stats_filters.values()),
error_stream = emitters.DebugEmitter(),
)
for fieldname, stats_filter in stats_filters.iteritems():
record['%s_stddev' % fieldname] = stats_filter.value()[0]
record['%s_mean' % fieldname] = stats_filter.average()
return record
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldModifier('year', lambda x: int(x) if x else None),
FieldModifier('amount', lambda x: Decimal(x) if x else None),
FieldModifier(
('affiliate', 'filing_included_nsfs',
'include_in_industry_totals', 'registrant_is_firm', 'use'),
lambda x: x == 'True'),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=20000, log=self.log),
LoaderEmitter(
LobbyingLoader(
source=self.inpath,
description='load from denormalized CSVs',
imported_by="loadlobbying (%s)" %
os.getenv('LOGNAME', 'unknown'),
log=self.log,
)),
)
def process_fec_year(year):
# committees
source = FixedWidthFileSource(open('%s/foiacm.dta' % year), CM_FIELDS)
#sqlite = SqliteOutput('fec%s.sqlite' % year, 'committee', [f[0] for f in CM_FIELDS if f[0] != 'filler'])
emit_mysql = SqlDumpEmitter(open('fec%s.sql' % year, 'a'), 'committee',
[f[0] for f in CM_FIELDS if f[0] != 'filler'])
run_recipe(source, emit_mysql)
# candidate
source = FixedWidthFileSource(open('%s/foiacn.dta' % year), CN_FIELDS)
fieldremover = FieldRemover(('fillerA', 'fillerB'))
#sqlite = SqliteOutput('fec%s.sqlite' % year, 'candidate', [f[0] for f in CN_FIELDS if f[0] != 'filler'])
emit_mysql = SqlDumpEmitter(
open('fec%s.sql' % year, 'a'), 'candidate',
[f[0] for f in CN_FIELDS if not f[0].startswith('filler')])
run_recipe(source, fieldremover, emit_mysql)
# contributions
source = FixedWidthFileSource(open('%s/itcont.dta' % year), INDIV_FIELDS)
decobolizer = FieldModifier(('amount', ), fix_cobol_number)
#sqlite = SqliteOutput('fec%s.sqlite' % year, 'contribution', [f[0] for f in INDIV_FIELDS if f[0] != 'filler'])
emit_mysql = SqlDumpEmitter(
open('fec%s.sql' % year, 'a'), 'contribution',
[f[0] for f in INDIV_FIELDS if f[0] != 'filler'])
run_recipe(source, decobolizer, emit_mysql)
def load_prices():
paths = [os.path.join(CPI_ROOT, fn) for fn in (
"cu.data.3.AsizeNorthEast",
"cu.data.4.AsizeNorthCentral",
"cu.data.5.AsizeSouth",
"cu.data.6.AsizeWest",
"cu.data.7.OtherNorthEast",
"cu.data.8.OtherNorthCentral",
"cu.data.9.OtherSouth",
"cu.data.10.OtherWest",
)]
headers = ('area_code','item_code','year','periodicity','period','value')
run_recipe(
csv.DictReader(local_files(paths), delimiter='\t'),
FieldModifier('value', lambda x: x.lstrip()),
SeriesIDFilter(),
#SqliteEmitter(DB, 'cpi_prices', fieldnames=headers),
DebugEmitter(),
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldRenamer(self.field_map),
# Values are [N|A]. Convert to boolean.
FieldModifier('is_amendment', \
lambda x: x == 'A'),
# Convert any stray floats to integers
FieldModifier('reporting_period_amount semi_annual_amount'.split(), \
lambda x: int(round(float(x))) if x else None),
# Convert date formats
FieldModifier('start_date end_date filing_date'.split(), \
lambda x: datetime.strptime(x, '%m/%d/%Y') if x else None),
# TODO: These following two lines (and the field value) need to be thoroughly tested on the next bundling load
FieldCopier({'pdf_url': 'first_image_num'}),
FieldModifier('pdf_url', \
lambda x: 'http://query.nictusa.com/pdf/{0}/{1}/{1}.pdf'.format(x[-3:], x)),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=200),
#DebugEmitter(),
DjangoModelEmitter('settings', Bundle)
)
def load_cpi():
urls = {
'north_east': 'ftp://ftp.bls.gov/pub/time.series/cu/cu.data.3.AsizeNorthEast',
'north_central': 'ftp://ftp.bls.gov/pub/time.series/cu/cu.data.4.AsizeNorthCentral',
'south': 'ftp://ftp.bls.gov/pub/time.series/cu/cu.data.5.AsizeSouth',
'west': 'ftp://ftp.bls.gov/pub/time.series/cu/cu.data.6.AsizeWest',
}
url = urls['west']
headers = ('series_id','survey_abbr','seasonal_code','periodicity_code',
'area_code','item_code','year','period','value','footnote_codes')
reader = remote_files(*urls.values(), headers=True)
run_recipe(
#CSVSource(utils.RemoteFile(url), delimiter='\t'),
CSVSource(reader, delimiter='\t'),
FieldModifier(('series_id','value'), str.strip),
SeriesIDFilter('series_id'),
ValueConditionalFilter('year', '2008'),
CSVEmitter(open('bls.csv', 'w'), headers)
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldRenamer(self.field_map),
# Values are [N|A]. Convert to boolean.
FieldModifier('is_amendment', \
lambda x: x == 'A'),
# Convert any stray floats to integers
FieldModifier('reporting_period_amount semi_annual_amount'.split(), \
lambda x: int(round(float(x.replace('$','').replace(',','')))) if x else None),
# Convert date formats
FieldModifier('start_date end_date filing_date'.split(), \
lambda x: datetime.strptime(x, '%m/%d/%Y') if x else None),
# TODO: These following two lines (and the field value) need to be thoroughly tested on the next bundling load
FieldCopier({'pdf_url': 'first_image_num'}),
FieldModifier('pdf_url', \
lambda x: 'http://query.nictusa.com/pdf/{0}/{1}/{1}.pdf'.format(x[-3:], x)),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=200),
#DebugEmitter(),
SimpleDjangoModelEmitter(Bundle)
)
def run(self):
run_recipe(
CSVSource(open(self.inpath)),
FieldRenamer(self.field_map),
# Values are [N|A]. Convert to boolean.
FieldModifier("is_amendment", lambda x: x == "A"),
# Convert any stray floats to integers
FieldModifier(
"reporting_period_amount semi_annual_amount".split(),
lambda x: int(round(float(x.replace("$", "").replace(",", "")))) if x else None,
),
# Convert date formats
FieldModifier(
"start_date end_date filing_date".split(), lambda x: datetime.strptime(x, "%m/%d/%Y") if x else None
),
# TODO: These following two lines (and the field value) need to be thoroughly tested on the next bundling load
FieldCopier({"pdf_url": "first_image_num"}),
FieldModifier("pdf_url", lambda x: "http://query.nictusa.com/pdf/{0}/{1}/{1}.pdf".format(x[-3:], x)),
NoneFilter(),
UnicodeFilter(),
CountEmitter(every=200),
# DebugEmitter(),
SimpleDjangoModelEmitter(Bundle),
)
from saucebrush.filters import Splitter, PhoneNumberCleaner, FieldMerger, FieldAdder
from saucebrush.emitters import DebugEmitter
import operator
from itertools import count
import saucebrush
data = [{
'person': {
'firstname': 'James',
'lastname': 'Turk'
},
'phones': [{
'phone': '222-222-2222'
}, {
'phone': '(202) 333-3321'
}]
}]
namemerger = FieldMerger({'name': ('firstname', 'lastname')},
lambda x, y: ' '.join((x, y)))
phonecleaner = PhoneNumberCleaner(('phone', ))
splitter = Splitter({'person': [namemerger], 'phones': [phonecleaner]})
ider = FieldAdder('id', count())
saucebrush.run_recipe(data, ider, splitter, DebugEmitter())
query = """
SELECT p.ID, p.post_author, u.user_login,
p.post_date, p.post_date_gmt, p.post_modified, p.post_modified_gmt,
p.post_content, p.post_title, p.post_category, p.post_excerpt, p.guid, p.post_type
FROM oh_posts p
INNER JOIN oh_users u ON p.post_author = u.ID
INNER JOIN oh_term_relationships tr ON p.ID = tr.object_id
INNER JOIN oh_term_taxonomy tt ON tr.term_taxonomy_id = tt.term_taxonomy_id
INNER JOIN oh_terms t ON tt.term_id = t.term_id
WHERE p.post_status = 'publish' AND p.post_type = 'post' AND (t.name = 'act' or t.name = 'The Day in Transparency')
ORDER BY p.post_date DESC
"""
mongo = Connection()
saucebrush.run_recipe(
MySQLSource(conn, query),
MetaFilter(conn),
TagFilter(conn),
ContentFilter(),
FieldModifier(('post_content', 'post_excerpt'), lambda x: Binary(x)),
MongoDBEmitter('openhouse', 'blog', drop_collection=True, conn=mongo),
#DebugEmitter(),
)
for d in mongo['openhouse']['blog'].find():
print "------", d['post_title']
conn.close()
def test_run_recipe(self):
saver = Saver()
run_recipe([1, 2], saver)
self.assertEqual(saver.saved, [1, 2])