def get_graph(**options):
"""
This function builds the graph that needs to be executed.
:return: bonobo.Graph
"""
graph = bonobo.Graph()
graph.add_chain(extract_accounts,
transform,
bonobo.JsonWriter('aws_accounts_ex.json'),
valid_aws_account,
_name="main")
graph.add_chain(
bonobo.JsonWriter('aws_accounts.json'),
_input="main",
)
graph.add_chain(
bonobo.UnpackItems(0),
bonobo.CsvWriter('aws_accounts.csv'),
_input=valid_aws_account,
)
graph.add_chain(
bonobo.UnpackItems(0),
bonobo_sqlalchemy.InsertOrUpdate(table_name='aws_accounts' +
options['table_suffix'],
discriminant=('account_id', ),
engine='db'),
_input=valid_aws_account,
)
return graph
def get_graph(graph=None, *, _limit=(), _print=()):
"""
Extracts a list of cafes with on euro in Paris, renames the name, address and zipcode fields,
reorders the fields and formats to json and csv files.
"""
graph = graph or bonobo.Graph()
producer = (
graph.get_cursor() >> ODSReader(dataset="liste-des-cafes-a-un-euro",
netloc="opendata.paris.fr") >>
PartialGraph(*_limit) >> bonobo.UnpackItems(0) >> bonobo.Rename(
name="nom_du_cafe", address="adresse", zipcode="arrondissement") >>
bonobo.Format(city="Paris", country="France") >> bonobo.OrderFields([
"name", "address", "zipcode", "city", "country", "geometry",
"geoloc"
]) >> PartialGraph(*_print))
# Comma separated values.
graph.get_cursor(producer.output) >> bonobo.CsvWriter(
"coffeeshops.csv",
fields=["name", "address", "zipcode", "city"],
delimiter=",")
# Standard JSON
graph.get_cursor(
producer.output) >> bonobo.JsonWriter(path="coffeeshops.json")
# Line-delimited JSON
graph.get_cursor(
producer.output) >> bonobo.LdjsonWriter(path="coffeeshops.ldjson")
return graph
def get_graph(graph=None, *, _limit=(), _print=()):
graph = graph or bonobo.Graph()
graph.add_chain(
OpenDataSoftAPI(dataset=API_DATASET),
*_limit,
normalize,
bonobo.UnpackItems(0),
*_print,
bonobo.JsonWriter(path='fablabs.json'),
)
return graph
def get_graph(*, _limit=None, _print=False):
graph = bonobo.Graph()
trunk = graph.add_chain(bonobo.JsonReader('datasets/theaters.json'), *((bonobo.Limit(_limit),) if _limit else ()))
if _print:
graph.add_chain(bonobo.PrettyPrinter(), _input=trunk.output)
graph.add_chain(bonobo.JsonWriter('theaters.json', fs='fs.output'), _input=trunk.output)
graph.add_chain(bonobo.LdjsonWriter('theaters.ldjson', fs='fs.output'), _input=trunk.output)
return graph
def get_graph(*, _limit=None, _print=False):
graph = bonobo.Graph()
trunk = graph.add_chain(bonobo.JsonReader("theaters.json", fs="fs.static"),
*((bonobo.Limit(_limit), ) if _limit else ()))
if _print:
graph.add_chain(bonobo.PrettyPrinter(), _input=trunk.output)
graph.add_chain(bonobo.JsonWriter("theaters.output.json", fs="fs.output"),
_input=trunk.output)
graph.add_chain(bonobo.LdjsonWriter("theaters.output.ldjson",
fs="fs.output"),
_input=trunk.output)
return graph
def get_graph(graph=None, *, _limit=(), _print=()):
"""
Extracts a list of cafes with on euro in Paris, renames the name, address and zipcode fields,
reorders the fields and formats to json and csv files.
"""
graph = graph or bonobo.Graph()
producer = graph.add_chain(
ODSReader(dataset='liste-des-cafes-a-un-euro',
netloc='opendata.paris.fr'),
*_limit,
bonobo.UnpackItems(0),
bonobo.Rename(name='nom_du_cafe',
address='adresse',
zipcode='arrondissement'),
bonobo.Format(city='Paris', country='France'),
bonobo.OrderFields([
'name', 'address', 'zipcode', 'city', 'country', 'geometry',
'geoloc'
]),
*_print,
)
# Comma separated values.
graph.add_chain(
bonobo.CsvWriter('coffeeshops.csv',
fields=['name', 'address', 'zipcode', 'city'],
delimiter=','),
_input=producer.output,
)
# Standard JSON
graph.add_chain(
bonobo.JsonWriter(path='coffeeshops.json'),
_input=producer.output,
)
# Line-delimited JSON
graph.add_chain(
bonobo.LdjsonWriter(path='coffeeshops.ldjson'),
_input=producer.output,
)
return graph
import bonobo
def split_one(line):
return line.split(', ', 1)
graph = bonobo.Graph(
bonobo.FileReader('coffeeshops.txt'),
split_one,
bonobo.JsonWriter('coffeeshops.json'),
)
if __name__ == '__main__':
bonobo.run(graph, services={'fs': bonobo.open_examples_fs('datasets')})
def get_graph(**options):
"""
This function builds the graph that needs to be executed.
:return: bonobo.Graph
"""
graph = bonobo.Graph()
graph.add_chain(
bonobo.CsvWriter('billing.csv'),
bonobo.JsonWriter('billing.json'),
invalid_entries,
fix_numbers,
parse_dates,
#bonobo.PrettyPrinter(),
filter_summary,
#bonobo.PrettyPrinter(),
lookup_account_sk,
lookup_date_sk,
summarize_costs,
bonobo.UnpackItems(0),
bonobo_sqlalchemy.InsertOrUpdate(
table_name='fact_itsm_aws_historical_cost' +
options['table_suffix'],
discriminant=(
'productname',
'date_sk',
'account_name_sk',
),
engine='database'),
_name="main",
_input=None,
)
now = options['now']
# Go to beginning of month
now += relativedelta(day=1, hour=0, minute=0, second=0, microsecond=0)
when = now
for log in range(0, options['months']):
when = when + relativedelta(months=-1)
tstamp = when.strftime("%Y-%m")
print("# %d Processing %s" % (log, tstamp))
if options['limit']:
_limit = (bonobo.Limit(options['limit']), )
else:
_limit = ()
graph.add_chain(
AwsBillingReader('%s-aws-cost-allocation-%s.csv' %
(options['aws_account_id'], tstamp),
fs='s3',
skip=1),
*_limit,
_output="main",
)
graph.add_chain(
bonobo_sqlalchemy.InsertOrUpdate(
table_name=options['table'] + options['table_suffix'],
discriminant=('invoiceid', 'linkedaccountid', 'payeraccountid',
'recordid'),
engine='database'),
_input=parse_dates,
)
return graph
(row.get('postal_code', None), row.get('city', None)))),
row.get('county', None),
row.get('country'),
)))
print(' - {}address{}: {address}'.format(Fore.BLUE,
Style.RESET_ALL,
address=', '.join(address)))
print(' - {}links{}: {links}'.format(Fore.BLUE,
Style.RESET_ALL,
links=', '.join(row['links'])))
print(' - {}geometry{}: {geometry}'.format(Fore.BLUE, Style.RESET_ALL,
**row))
print(' - {}source{}: {source}'.format(Fore.BLUE,
Style.RESET_ALL,
source='datanova/' + API_DATASET))
graph = bonobo.Graph(
OpenDataSoftAPI(dataset=API_DATASET,
netloc=API_NETLOC,
timezone='Europe/Paris'),
normalize,
bonobo.Filter(filter=lambda row: row.get('country') == 'France'),
bonobo.JsonWriter(path='fablabs.txt', ioformat='arg0'),
bonobo.Tee(display),
)
if __name__ == '__main__':
bonobo.run(graph, services=get_default_services(__file__))
import bonobo
def split_one(line):
return dict(zip(("name", "address"), line.split(', ', 1)))
graph = bonobo.Graph(
bonobo.FileReader('coffeeshops.txt'),
split_one,
bonobo.JsonWriter('coffeeshops.json', fs='fs.output'),
)
def get_services():
return {
'fs': bonobo.open_examples_fs('datasets'),
'fs.output': bonobo.open_fs(),
}
if __name__ == '__main__':
bonobo.run(graph, services=get_services())
import bonobo
graph = bonobo.Graph(
bonobo.CsvReader('Google_facebook_statuses.csv'),
bonobo.JsonWriter('output.json'),
)
if __name__ == '__main__':
bonobo.run(graph)
import bonobo
def split_one(line):
return line.split(', ', 1)
graph = bonobo.Graph(
bonobo.FileReader('coffeeshops.txt'),
split_one,
bonobo.JsonWriter('coffeeshops.json', fs='fs.output', ioformat='arg0'),
)
def get_services():
return {
'fs': bonobo.open_examples_fs('datasets'),
'fs.output': bonobo.open_fs(),
}
if __name__ == '__main__':
bonobo.run(graph, services=get_services())
(row.get('postal_code', None), row.get('city', None)))),
row.get('county', None),
row.get('country'),
)))
print(' - {}address{}: {address}'.format(Fore.BLUE,
Style.RESET_ALL,
address=', '.join(address)))
print(' - {}links{}: {links}'.format(Fore.BLUE,
Style.RESET_ALL,
links=', '.join(row['links'])))
print(' - {}geometry{}: {geometry}'.format(Fore.BLUE, Style.RESET_ALL,
**row))
print(' - {}source{}: {source}'.format(Fore.BLUE,
Style.RESET_ALL,
source='datanova/' + API_DATASET))
graph = bonobo.Graph(
OpenDataSoftAPI(dataset=API_DATASET,
netloc=API_NETLOC,
timezone='Europe/Paris'),
normalize,
filter_france,
bonobo.Tee(display),
bonobo.JsonWriter(path='fablabs.txt'),
)
if __name__ == '__main__':
bonobo.run(graph, services=get_default_services(__file__))
