def setup(self):
#configure loader
self.loader = tool.Persistence.instance()
self.graphHelper = tool.Graph.instance()
#CREATE BASE GRAPH
#When using the loading functionalities, it will be overwritten
self.vnq_model = VNQ()
self.setGraph(nx.MultiDiGraph()) # directed graph, which allows multiple 'similar' edges between two nodes
g = self.vnq_model.getGraph()
g.graph['name'] = 'DefaultGraph'
g.graph['version'] = 0.1
g.add_node(1, name='TransitNSP')
self.welcome()
#self.g.node[1]
#self.welcome()
#welcome
return
class VNQController:
def __init__(self):
self.loader = None
#self.g = None # Graph .. redundant step #never forget self when assigning ;)
self.pp = pprint.PrettyPrinter(indent=4) # advanced printing purposes
self.vnq_model = None
self.setup() # creating basic loader and basic graph
return
# Set the graph
# Pipes it through to the model
def setGraph(self, g):
#self.g = g
self.vnq_model.setGraph(g)
def getGraph(self):
return self.vnq_model.getGraph()
# Creates a base graph
# Creates required tools
def setup(self):
#configure loader
self.loader = tool.Persistence.instance()
self.graphHelper = tool.Graph.instance()
#CREATE BASE GRAPH
#When using the loading functionalities, it will be overwritten
self.vnq_model = VNQ()
self.setGraph(nx.MultiDiGraph()) # directed graph, which allows multiple 'similar' edges between two nodes
g = self.vnq_model.getGraph()
g.graph['name'] = 'DefaultGraph'
g.graph['version'] = 0.1
g.add_node(1, name='TransitNSP')
self.welcome()
#self.g.node[1]
#self.welcome()
#welcome
return
# Welcome message
# Points to view
def welcome(self):
vnq.VNQView.welcome()
def configure(self, file): #add parameters later
log.info('Configuring the graph ...')
self.setGraph(self.loader.fromJSON(file)) #
self.pp.pprint(self.vnq_model.getGraph())
node_list = self.graphHelper.list_nodes(self.vnq_model.getGraph())
log.info(str(len(node_list)) + ' nodes detected (#):')
for node in node_list:
log.info('Node: #'+ node)
#print params._VISUAL_INLINE_MARKER + params._VISUAL_INLINE_MARKER + '#' + node
return
# EXECUTION PHASE
# = ENTRY POINT
def execute(self, risk_params):
log.info('Executing the calculation ...')
results = self.run(risk_params)
log.info('Finished calculation ...')
return results
# Preparation and presentation of results
def results(self, result_data):
# Sort for human beings first
g = self.vnq_model.getGraph() # Load it from the model
ordered_results = self.graphHelper.order_results(g, result_data) # order the results to match human readability (nodes from 0 to n instead of arbitrary order)
#Presentation of results
vnq.VNQView.show_results(ordered_results) # console
#Presentation of reports (warnings etc.)
log.info("Analyzing reports ...")
key_reports = reporting.ReportManager.keyReports()
vnq.VNQView.displayStructuralFailures(key_reports)
report_results = [r for r in reporting.ReportManager.reports.values()]
vnq.VNQView.displayReports(report_results)
# Computation method
# TODO: Think about moving it to a model class
# Risk_params: Eclectic risks need external calculation or estimation; give a list of 1s if you want to skip this computation step
# entity_size: Pass market measurements only in the case the relationships are not on market scale (e.g. downscaled to a single service usage)
# Otherwise: Leave empty
# Manual input is always regarded to be symmetric (we do not distinguish whether an entity is a big supplier or customer, as the data typically does not exist. May be revised in the future.
# Order input by nodes
def run(self, risk_params, entity_sizes):
# Calculate basic bargaining powers
# Does not include
power_indicators = self.vnq_model.run(entity_sizes)
# Presentation of final bargaining powers
self.show_final_powers(power_indicators)
#Make a ready to use weighted risk indicator
rks_indicators = self.vnq_model.risks(risk_params) # [3_RK]per entity
#Bring both directions together and nosidepayments.py a single value
result = self.vnq_model.vn_level(power_indicators, rks_indicators) # [3_FINAL] calc
return result
# Display the bargaining powers
# -> Driving indicators
def show_final_powers(self,power_indicators):
# order to match human readability
print power_indicators[0]
g = self.vnq_model.getGraph() # load it from the model
power_in = self.graphHelper.order_results(g, power_indicators[0]) # Incoming relationships (suppliers)
power_out = self.graphHelper.order_results(g, power_indicators[1]) # Outgoing relationships (cusotmers)
# Console presentation
vnq.VNQView.displayFinalPowers(power_in, power_out)