class SourceArchivalObject(SourceComponentBase):
"""A component of a SourceResource."""
position = odin.IntegerField()
ref_id = odin.StringField()
component_id = odin.StringField(null=True)
display_string = odin.StringField()
restrictions_apply = odin.BooleanField()
ancestors = odin.ArrayOf(SourceAncestor)
resource = odin.DictAs(SourceRef)
has_unpublished_ancestor = odin.BooleanField()
instances = odin.ArrayOf(SourceInstance)
class SourceNameBase(odin.Resource):
"""Base class for structured representations of names.
Subclassed by more specific representations SourceNameCorporateEntity,
SourceNamePerson and SourceNameFamily.
"""
sort_name = odin.StringField()
authorized = odin.BooleanField()
is_display_name = odin.BooleanField()
# use_dates = odin.ArrayOf(SourceStructuredDate) # TODO: account for structured and nonstructured dates
rules = odin.StringField(choices=configs.NAME_RULES_CHOICES, null=True)
source = odin.StringField(choices=configs.NAME_SOURCE_CHOICES, null=True)
class ArchivesSpaceNameBase(odin.Resource):
"""Base class for names.
Subclassed by names specific to an agent type."""
rules = odin.StringField(default="dacs")
source = odin.StringField(default="local")
sort_name_auto_generate = odin.BooleanField(default=True)
class SourceComponentBase(odin.Resource):
"""Base class for archival components.
Subclassed by SourceArchivalObject and SourceResource.
Both language and lang_material need to exist in order to accomodate
ArchivesSpace API changes between v2.6 and v2.7.
"""
class Meta:
abstract = True
COMPONENT_TYPES = (('archival_object', 'Archival Object'), ('resource',
'Resource'))
dates = odin.ArrayOf(SourceDate)
extents = odin.ArrayOf(SourceExtent)
external_ids = odin.ArrayOf(SourceExternalId)
group = odin.DictAs(SourceGroup)
jsonmodel_type = odin.StringField(choices=COMPONENT_TYPES)
lang_materials = odin.ArrayOf(SourceLangMaterial, null=True)
language = odin.StringField(null=True)
level = odin.StringField()
linked_agents = odin.ArrayOf(SourceLinkedAgent)
notes = odin.ArrayOf(SourceNote)
publish = odin.BooleanField()
subjects = odin.ArrayOf(SourceRef)
suppressed = odin.StringField()
title = odin.StringField(null=True)
uri = odin.StringField()
class ArchivesSpaceDigitalObject(odin.Resource):
"""A digital object representing a group of archival records."""
jsonmodel_type = odin.StringField(default="digital_object")
publish = odin.BooleanField(default=False)
title = odin.StringField()
digital_object_id = odin.IntegerField()
file_versions = odin.ArrayOf(ArchivesSpaceFileVersion)
repository = odin.DictField(default={"ref": "/repositories/{}".format(settings.ARCHIVESSPACE['repo_id'])})
class SourceSubject(odin.Resource):
"""A topical term."""
external_ids = odin.ArrayOf(SourceExternalId)
group = odin.DictAs(SourceGroup)
publish = odin.BooleanField()
source = odin.StringField(choices=configs.SUBJECT_SOURCE_CHOICES)
terms = odin.ArrayOf(SourceTerm)
title = odin.StringField()
uri = odin.StringField()
class ArchivesSpaceNote(odin.Resource):
"""A note describing a group of archival records."""
publish = odin.BooleanField(default=False)
jsonmodel_type = odin.StringField()
type = odin.StringField()
label = odin.StringField(null=True)
subnotes = odin.ArrayOf(ArchivesSpaceSubnote, null=True)
content = odin.StringField(null=True)
items = odin.StringField(null=True)
class SourceResource(SourceComponentBase):
"""An aggregation of records."""
position = odin.IntegerField()
restrictions = odin.BooleanField()
ead_id = odin.StringField(null=True)
finding_aid_title = odin.StringField(null=True)
finding_aid_filing_title = odin.StringField(null=True)
id_0 = odin.StringField()
id_1 = odin.StringField(null=True)
id_2 = odin.StringField(null=True)
ancestors = odin.ArrayOf(SourceAncestor, null=True)
instances = odin.ArrayOf(SourceInstance)
class SourceNote(odin.Resource):
"""Human-readable note.
SourceNotes contain one or more SourceSubnotes.
"""
jsonmodel_type = odin.StringField()
type = odin.StringField(null=True)
label = odin.StringField(null=True)
subnotes = odin.ArrayOf(SourceSubnote, null=True)
content = odin.StringField(null=True)
items = odin.ArrayField(null=True)
publish = odin.BooleanField()
class Object(BaseResource):
"""An aggregation of archival records.
Objects may be contained within Collections, but do not contain other
Objects or Collections. Object is a first-class entity in the RAC data model.
"""
type = odin.StringField(default="object")
category = odin.StringField(default="collection")
dates = odin.ArrayOf(Date)
languages = odin.ArrayOf(Language)
extents = odin.ArrayOf(Extent)
notes = odin.ArrayOf(Note)
people = odin.ArrayOf(AgentReference)
organizations = odin.ArrayOf(AgentReference)
families = odin.ArrayOf(AgentReference)
terms = odin.ArrayOf(TermReference)
ancestors = odin.ArrayOf(RecordReference, null=True)
parent = odin.StringField()
position = odin.IntegerField()
formats = odin.ArrayField()
online = odin.BooleanField(default=False)
class SourceAgentBase(odin.Resource):
"""A base class for agents.
Subclassed by SourceAgentFamily, SourceAgentPerson and
SourceAgentCorporateEntity.
"""
class Meta:
abstract = True
AGENT_TYPES = (('agent_corporate_entity', 'Organization'),
('agent_family', 'Family'), ('agent_person', 'Person'))
agent_record_identifiers = odin.ArrayOf(SourceAgentRecordIdentifier,
null=True)
dates_of_existence = odin.ArrayField(null=True)
group = odin.DictAs(SourceGroup)
jsonmodel_type = odin.StringField(choices=AGENT_TYPES)
notes = odin.ArrayOf(SourceNote)
publish = odin.BooleanField()
title = odin.StringField()
uri = odin.StringField()
class ArchivesSpaceComponentBase(odin.Resource):
"""Base class for components, which describe groups of archival records."""
class Meta:
abstract = True
COMPONENT_TYPES = (
('archival_object', 'Archival Object'),
('accession', 'Accession'),
)
dates = odin.ArrayOf(ArchivesSpaceDate)
extents = odin.ArrayOf(ArchivesSpaceExtent)
external_ids = odin.ArrayOf(ArchivesSpaceExternalId)
instances = odin.ArrayField(null=True)
jsonmodel_type = odin.StringField(choices=COMPONENT_TYPES)
lang_materials = odin.ArrayOf(ArchivesSpaceLangMaterial, null=True)
linked_agents = odin.ArrayOf(ArchivesSpaceLinkedAgent)
notes = odin.ArrayOf(ArchivesSpaceNote)
publish = odin.BooleanField(default=False)
repository = odin.DictField(default={"ref": "/repositories/{}".format(settings.ARCHIVESSPACE['repo_id'])})
rights_statements = odin.ArrayOf(ArchivesSpaceRightsStatement)
title = odin.StringField(null=True)
uri = odin.StringField()
class Book(LibraryBook):
class Meta:
key_field_name = 'isbn'
title = odin.StringField()
isbn = odin.StringField()
num_pages = odin.IntegerField()
rrp = odin.FloatField(default=20.4, use_default_if_not_provided=True)
fiction = odin.BooleanField(is_attribute=True)
genre = odin.StringField(choices=(
('sci-fi', 'Science Fiction'),
('fantasy', 'Fantasy'),
('biography', 'Biography'),
('others', 'Others'),
('computers-and-tech', 'Computers & technology'),
))
published = odin.TypedArrayField(odin.DateTimeField())
authors = odin.ArrayOf(Author, use_container=True)
publisher = odin.DictAs(Publisher, null=True)
def __eq__(self, other):
if other:
return vars(self) == vars(other)
return False
class Collection(BaseResource):
"""An aggregation of archival records.
Collections contain other aggregations of records (either Objects or
Collections), and may themselves be contained within another Collection.
Collection is a first-class entity in the RAC data model.
"""
type = odin.StringField(default="collection")
category = odin.StringField(default="collection")
level = odin.StringField()
dates = odin.ArrayOf(Date)
creators = odin.ArrayOf(AgentReference)
languages = odin.ArrayOf(Language)
extents = odin.ArrayOf(Extent)
notes = odin.ArrayOf(Note)
people = odin.ArrayOf(AgentReference)
organizations = odin.ArrayOf(AgentReference)
families = odin.ArrayOf(AgentReference)
terms = odin.ArrayOf(TermReference)
ancestors = odin.ArrayOf(RecordReference, null=True)
parent = odin.StringField(null=True)
position = odin.IntegerField()
formats = odin.ArrayField()
online = odin.BooleanField(default=False)
class SourceInstance(odin.Resource):
"""The physical or digital instantiation of a group of records."""
instance_type = odin.StringField(choices=configs.INSTANCE_TYPE_CHOICES)
is_representative = odin.BooleanField()
sub_container = odin.DictAs(SourceSubcontainer, null=True)
digital_object = odin.DictAs(SourceRef, null=True)
class Tariff(odin.Resource):
"""A collection of charges associated with a specific utility service"""
name = odin.StringField(null=True)
code = odin.StringField(null=True)
utility_name = odin.StringField(null=True)
utility_code = odin.StringField(null=True)
service = odin.StringField(choices=SERVICE_CHOICES, null=True)
sector = odin.StringField(choices=SECTOR_CHOICES, null=True)
description = odin.StringField(null=True)
currency = odin.StringField(null=True)
timezone = odin.StringField(null=True)
min_consumption = odin.FloatField(null=True)
max_consumption = odin.FloatField(null=True)
min_demand = odin.FloatField(null=True)
max_demand = odin.FloatField(null=True)
charges = odin.ArrayOf(Charge, null=True)
monthly_charge = odin.FloatField(null=True)
minimum_charge = odin.FloatField(null=True)
times = odin.DictAs(Times, null=True)
seasons = odin.DictAs(Seasons, null=True)
net_metering = odin.BooleanField(null=True)
billing_period = odin.StringField(choices=PERIOD_CHOICES,
null=True,
default='monthly',
use_default_if_not_provided=True)
demand_window = odin.StringField(choices=DEMAND_WINDOW_CHOICES,
null=True,
default='30min',
use_default_if_not_provided=True)
consumption_unit = odin.StringField(choices=CONSUMPTION_UNIT_CHOICES,
null=True)
demand_unit = odin.StringField(choices=DEMAND_UNIT_CHOICES, null=True)
@odin.calculated_field
def charge_types(self):
charge_types = set()
for charge in self.charges:
if charge.season:
charge_types.add('seasonal')
if charge.time:
charge_types.add('tou')
if charge.rate_bands:
charge_types.add('block')
if charge.rate_schedule:
charge_types.add('scheduled')
if charge.type == 'demand':
charge_types.add('demand')
if charge.type == 'consumption':
charge_types.add('consumption')
if charge.type == 'generation':
charge_types.add('generation')
if charge.type == 'fixed':
charge_types.add('fixed')
return list(charge_types)
def calc_charge(self, row, charge, cost_items, block_accum_dict):
if charge.rate:
try:
cost_items[charge.name]['cost'] += charge.rate * float(
row[charge.meter or charge.type])
except KeyError:
raise Exception(
"The specified meter data file has no field named %s" %
str(charge.meter or charge.type))
if charge.rate_bands:
charge_time_step = float()
prev_block_usage = 0.0
for rate_band_index, rate_band in enumerate(charge.rate_bands):
if block_accum_dict[charge.name] > rate_band.limit:
continue
try:
block_usage = max((min(
(rate_band.limit - block_accum_dict[charge.name],
row[charge.meter or charge.type] - prev_block_usage)),
0.0))
except KeyError:
raise Exception(
"The specified meter data file has no field named %s" %
str(charge.meter or charge.type))
prev_block_usage += block_usage
charge_time_step += rate_band.rate * block_usage
block_accum_dict[charge.name] += block_usage
cost_items[charge.name]['cost'] += charge_time_step
return cost_items, block_accum_dict
def apply_by_charge_type(self,
meter_data,
cost_items,
charge_type='consumption'):
"""
Calculates the cost of energy given a tariff and load.
:param meter_data: a three-column pandas array with datetime, imported energy (kwh), exported energy (kwh)
:param cost_items: a dictionary containing the charge components (e.g. off_peak, shoulder, peak, total)
:param charge_type: a string specifying the charge type, options include 'consumption', 'demand' etc
:return: a dictionary containing the charge components (e.g. off_peak, shoulder, peak, total)
"""
block_accum_dict = defaultdict(float)
for dt, row in meter_data.iterrows():
time = datetime.time(hour=dt.hour, minute=dt.minute)
# If the billing cycle changes over, reset block charge accumulations
if (charge_type == 'consumption'
and self.billing_period == 'monthly' and dt.day == 1
and dt.hour == 0 and dt.minute == 0
or self.billing_period == 'quarterly' and dt.month % 3
and dt.day == 1 and dt.hour == 0 and dt.minute == 0
or self.billing_period == 'annually' and dt.month == 1
and dt.day == 1 and dt.hour == 0 and dt.minute == 0):
block_accum_dict = defaultdict(float)
if charge_type == 'demand':
block_accum_dict = defaultdict(float)
if self.charges:
for charge in self.charges:
if charge.type == charge_type or (
charge_type == 'consumption'
and charge.type == 'generation'):
if charge.time and charge.season:
if datetime.date(year=dt.year, month=charge.season.from_month, day=charge.season.from_day) \
<= dt.date() <= datetime.date(year=dt.year, month=charge.season.to_month,
day=charge.season.to_day):
for period in charge.time.periods:
if period.from_weekday <= dt.dayofweek <= period.to_weekday and datetime.time(
hour=period.from_hour,
minute=period.from_minute
) <= time <= datetime.time(
hour=period.to_hour,
minute=period.to_minute):
cost_items, block_accum_dict = self.calc_charge(
row, charge, cost_items,
block_accum_dict)
break
elif charge.season:
if datetime.date(year=dt.year, month=charge.season.from_month, day=charge.season.from_day)\
<= dt.date() <= datetime.date(year=dt.year, month=charge.season.to_month,
day=charge.season.to_day):
cost_items, block_accum_dict = self.calc_charge(
row, charge, cost_items, block_accum_dict)
elif charge.time:
for period in charge.time.periods:
if period.from_weekday <= dt.dayofweek <= period.to_weekday and datetime.time(
hour=period.from_hour,
minute=period.from_minute
) <= time <= datetime.time(
hour=period.to_hour,
minute=period.to_minute):
cost_items, block_accum_dict = self.calc_charge(
row, charge, cost_items,
block_accum_dict)
elif charge.rate_schedule:
for schedule_item in charge.rate_schedule:
if dt.to_pydatetime() < schedule_item.datetime:
cost_items[charge.type + 'scheduled'][
'cost'] += schedule_item.rate * float(
row[charge.type])
break
else:
cost_items, block_accum_dict = self.calc_charge(
row, charge, cost_items, block_accum_dict)
return cost_items
def apply(self, meter_data, start=None, end=None):
"""
Calculates the cost of energy given a tariff and load.
:param meter_data: a three-column pandas array with datetime, imported energy (kwh), exported energy (kwh)
:param start: an optional datetime to select the commencement of the bill calculation
:param end: an optional datetime to select the termination of the bill calculation
:return: a dictionary containing the charge components (e.g. off_peak, shoulder, peak, total)
"""
meter_data.truncate(before=start, after=end)
# Create empty dict to hold calculated cost items
cost_items = dict()
for charge in self.charges:
season = charge.season.name if charge.season else None
time = charge.time.name if charge.time else None
cost_items[charge.name] = {
'name': charge.name,
'code': charge.code,
'type': charge.type,
'season': season,
'time': time,
'cost': 0.0,
}
if 'consumption' in self.charge_types or 'generation' in self.charge_types:
consumption_data = meter_data
# Resample meter data if finer resolution data not required by the specified tariff types
if 'tou' not in self.charge_types:
if 'seasonal' in self.charge_types:
consumption_data = meter_data.resample('D').sum()
else:
consumption_data = meter_data.resample(
PERIOD_TO_TIMESTEP[self.billing_period]).sum()
cost_items.update(
self.apply_by_charge_type(consumption_data, cost_items,
'consumption'))
if 'demand' in self.charge_types:
# Resample meter data to the demand window and then take the maximum for each billing period
demand_data = meter_data.resample(
PERIOD_TO_TIMESTEP[self.demand_window]).mean()
peak_monthly = demand_data.resample(
PERIOD_TO_TIMESTEP[self.billing_period]).max()
cost_items.update(
self.apply_by_charge_type(peak_monthly, cost_items, 'demand'))
if 'fixed' in self.charge_types:
billing_periods = len(
meter_data.resample(
PERIOD_TO_TIMESTEP[self.billing_period]).mean())
for charge in self.charges:
if charge.type == 'fixed':
cost_items[
charge.name]['cost'] = billing_periods * charge.rate
# Transform the output data into the specified output format
cost_dict = {
'name': self.name,
'code': self.code,
'items': cost_items.values()
}
return dict_codec.load(cost_dict, Cost)
class ArchivesSpaceSubnote(odin.Resource):
"""A repeatable object containing note content."""
jsonmodel_type = odin.StringField()
publish = odin.BooleanField(default=False)
content = odin.StringField(null=True)
items = odin.StringField(null=True)
class Tariff(odin.Resource):
"""A collection of charges associated with a specific utility service"""
name = odin.StringField(null=True)
code = odin.StringField(null=True)
utility_name = odin.StringField(null=True)
utility_code = odin.StringField(null=True)
service = odin.StringField(choices=SERVICE_CHOICES, null=True)
sector = odin.StringField(choices=SECTOR_CHOICES, null=True)
description = odin.StringField(null=True)
currency = odin.StringField(null=True)
timezone = odin.StringField(null=True)
min_consumption = odin.FloatField(null=True)
max_consumption = odin.FloatField(null=True)
min_demand = odin.FloatField(null=True)
max_demand = odin.FloatField(null=True)
charges = odin.ArrayOf(Charge, null=True)
monthly_charge = odin.FloatField(null=True)
minimum_charge = odin.FloatField(null=True)
times = odin.DictAs(Times, null=True)
seasons = odin.DictAs(Seasons, null=True)
net_metering = odin.BooleanField(null=True)
billing_period = odin.StringField(choices=PERIOD_CHOICES,
null=True,
default='monthly',
use_default_if_not_provided=True)
demand_window = odin.StringField(choices=DEMAND_WINDOW_CHOICES,
null=True,
default='30min',
use_default_if_not_provided=True)
consumption_unit = odin.StringField(choices=CONSUMPTION_UNIT_CHOICES,
null=True)
demand_unit = odin.StringField(choices=DEMAND_UNIT_CHOICES, null=True)
@odin.calculated_field
def charge_types(self):
charge_types = set()
for charge in self.charges:
if charge.season:
charge_types.add('seasonal')
if charge.time:
charge_types.add('tou')
if charge.rate_bands:
charge_types.add('block')
if charge.rate_schedule:
charge_types.add('scheduled')
if charge.type == 'demand':
charge_types.add('demand')
if charge.type == 'consumption':
charge_types.add('consumption')
return list(charge_types)
def calc_charge(self, name, row, charge, charge_array, block_accum_dict):
if charge.rate:
charge_array[name].append(charge.rate * float(row[charge.meter]))
if charge.rate_bands:
charge_time_step = float()
for rate_band_index, rate_band in enumerate(charge.rate_bands):
if block_accum_dict[name] > rate_band.limit:
continue
block_usage = max((min(
(rate_band.limit - block_accum_dict[name],
row[charge.meter] - block_accum_dict[name])), 0.0))
charge_time_step += rate_band.rate * block_usage
block_accum_dict[name] += block_usage
charge_array[name].append(charge_time_step)
return charge_array, block_accum_dict
def apply_by_charge_type(self, meter_data, charge_type='consumption'):
"""
Calculates the cost of energy given a tariff and load.
:param meter_data: a three-column pandas array with datetime, imported energy (kwh), exported energy (kwh)
:param step: the time step in minutes of the meter data
:param start: an optional datetime to select the commencement of the bill calculation
:param end: an optional datetime to select the termination of the bill calculation
:return: a dictionary containing the charge components (e.g. off_peak, shoulder, peak, total)
"""
charge_array = defaultdict(list)
block_accum_dict = defaultdict(float)
for dt, row in meter_data.iterrows():
time = datetime.time(hour=dt.hour, minute=dt.minute)
# If the billing cycle changes over, reset block charge accumulations
if (charge_type == 'consumption'
and self.billing_period == 'monthly' and dt.day == 1
and dt.hour == 0 and dt.minute == 0
or self.billing_period == 'quarterly' and dt.month % 3
and dt.day == 1 and dt.hour == 0 and dt.minute == 0
or self.billing_period == 'annually' and dt.month == 1
and dt.day == 1 and dt.hour == 0 and dt.minute == 0):
block_accum_dict = defaultdict(float)
if charge_type == 'demand':
block_accum_dict = defaultdict(float)
if self.charges:
for charge in self.charges:
if charge.type == charge_type:
if charge.time and charge.season:
found = False
if datetime.date(year=dt.year, month=charge.season.from_month, day=charge.season.from_day) \
< dt.date() <= datetime.date(year=dt.year, month=charge.season.to_month,
day=charge.season.to_day):
for period in charge.time.periods:
if datetime.time(
hour=period.from_hour,
minute=period.from_minute
) < time <= datetime.time(
hour=period.to_hour,
minute=period.to_minute):
charge_array, block_accum_dict = self.calc_charge(
self.service + charge_type +
charge.season.name +
charge.time.name, row, charge,
charge_array, block_accum_dict)
found = True
break
if not found:
charge_array[self.service + charge_type +
charge.season.name +
charge.time.name].append(0.0)
elif charge.season and not charge.time:
if datetime.date(year=dt.year, month=charge.season.from_month, day=charge.season.from_day)\
< dt.date() <= datetime.date(year=dt.year, month=charge.season.to_month,
day=charge.season.to_day):
charge_array, block_accum_dict = self.calc_charge(
self.service + charge_type +
charge.season.name, row, charge,
charge_array, block_accum_dict)
else:
charge_array[self.service + charge_type +
charge.season.name].append(0.0)
elif charge.time and not charge.season:
found = False
for period in charge.time.periods:
if datetime.time(hour=period.from_hour
) < time <= datetime.time(
hour=period.to_hour):
charge_array, block_accum_dict = self.calc_charge(
self.service + charge_type +
charge.time.name, row, charge,
charge_array, block_accum_dict)
found = True
if not found:
charge_array[self.service + charge_type +
charge.time.name].append(0.0)
elif charge.rate_schedule:
for schedule_item in charge.rate_schedule:
if dt <= schedule_item.datetime:
charge_array[self.service + charge_type +
'scheduled'].append(
schedule_item.rate *
float(row[charge.meter]))
break
else:
charge_array, block_accum_dict = self.calc_charge(
self.service + charge_type, row, charge,
charge_array, block_accum_dict)
return charge_array
def apply(self, meter_data, start=None, end=None, output_format='total'):
"""
Calculates the cost of energy given a tariff and load.
:param meter_data: a three-column pandas array with datetime, imported energy (kwh), exported energy (kwh)
:param step: the time step in minutes of the meter data
:param start: an optional datetime to select the commencement of the bill calculation
:param end: an optional datetime to select the termination of the bill calculation
:return: a dictionary containing the charge components (e.g. off_peak, shoulder, peak, total)
"""
meter_data.truncate(before=start, after=end)
charge_array = defaultdict(list)
if 'consumption' in self.charge_types:
consumption_data = meter_data
# Resample meter data if finer resolution data not required by the specified tariff types
if 'tou' not in self.charge_types and output_format != 'input-timestep' and output_format != 'input-timestep-components':
if 'seasonal' in self.charge_types:
consumption_data = meter_data.resample('D').sum()
else:
consumption_data = meter_data.resample(
PERIOD_TO_TIMESTEP[self.billing_period]).sum()
consumption_charges = self.apply_by_charge_type(
consumption_data, 'consumption')
charge_array.update(consumption_charges)
if 'demand' in self.charge_types:
if output_format == 'input-timestep' or output_format == 'input-timestep-components':
raise UserWarning(
"The output_format cannot be specified as 'input-timestep' if demand charges have "
"been assigned.")
# Resample meter data to the demand window and then take the maximum for each billing period
demand_data = meter_data.resample(
PERIOD_TO_TIMESTEP[self.demand_window]).mean()
peak_monthly = demand_data.resample(
PERIOD_TO_TIMESTEP[self.billing_period]).max()
demand_charges = self.apply_by_charge_type(peak_monthly, 'demand')
charge_array.update(demand_charges)
# Transform the output data into the specified output format
if output_format == 'total':
output = 0.0
for v in charge_array.values():
output += sum(v)
elif output_format == 'total-components':
output = dict()
for k, v in charge_array.items():
output[k] = sum(v)
else:
df = pandas.DataFrame.from_dict(data=charge_array)
df.index = meter_data.index
if output_format == 'billing-period':
output = df.resample(
PERIOD_TO_TIMESTEP[self.billing_period].sum()).sum(1)
elif output_format == 'billing-period-components':
output = df.resample(
PERIOD_TO_TIMESTEP[self.billing_period].sum())
elif output_format == 'input-timestep':
output = df.sum(1)
elif output_format == 'input-timestep-components':
output = df
else:
raise UserWarning('Unsupported output format: %s' %
output_format)
return output