Element43 margin script

station.haml

- extends "base.haml"
- load cache
- load humanize
- block title
  element43 - Station - Best Margins
- block content
  %h1
    {{station.name}}
  .row-fluid
    .span12
      %table.table.table-striped.table-condensed
        %thead
          %tr
            %th
            %th
              Daily Volume
            %th
              Best Ask
            %th
              Best Bid
            %th
              Margin
        %tbody
          - for item in top_margins
             %tr
               %td
                 %a{'href':'/market/region/{{station.region.id}}/{{item.type.id}}/'}
                   {{item.type.name}}
               %td
                 {{item.volume|intcomma}}
               %td
                 {{item.top_ask|intcomma}} ISK
               %td
                 {{item.top_bid|intcomma}} ISK
               %td
                 {{item.margin|floatformat:2}} %
          
    

station.py

def margins(request, station_id = 60003760):
	"""
	Generates a list like http://goonmetrics.com/importing/
	"""
	
	# Get station object - default to CNAP if something goes wrong
	try:
		station = StaStation.objects.get(id = station_id)
	except:
		station_id = 60003760
		station = StaStation.objects.get(id = station_id)
		
	# Get all orders grouped by type
	top_ask = Orders.objects.filter(stastation = station, is_bid = False).values('invtype').annotate(min_price=Min('price'))
	top_bid = Orders.objects.filter(stastation = station, is_bid = True).values('invtype').annotate(max_price=Max('price'))
	
	# Get re-structured dicts and sets
	ask = {}
	bid = {}
	
	for item in top_ask:
		ask[item['invtype']] = item['min_price']
		
	ask_set = set(ask)
	
	for item in top_bid:
		bid[item['invtype']] = item['max_price']
	bid_set = set(bid)
	
	# Get types with both buy and sell orders
	types = ask_set.intersection(bid_set)
	
	# Store margins
	top_margins = []
	
	for invtype in types:
		margin = ((ask[invtype] / bid[invtype]) * 100)
		
		try:
			volume = ast.literal_eval(History.objects.get(mapregion = station.region, invtype_id = invtype).history_data).itervalues().next()[4]
		except:
			volume = 0
			
		score = volume * margin
		
		# Filter Junk
		if (margin > 0) and (margin < 10000):
			top_margins.append({'type': invtype, 'margin': margin, 'score': score, 'volume':volume})
	
	# Sort margins
	top_margins = sorted(top_margins, key=lambda k: k['score'])
	top_margins.reverse()
	del top_margins[100:]
	
	# Take top 50 margins
	for margin in top_margins:
		margin['type'] = InvType.objects.get(id = margin['type'])
			
		margin['top_ask'] = Orders.objects.filter(stastation = station, is_bid = False, invtype_id = margin['type']).aggregate(Min('price'))['price__min']
		margin['top_bid'] = Orders.objects.filter(stastation = station, is_bid = True, invtype_id = margin['type']).aggregate(Max('price'))['price__max']
	
	rcontext = RequestContext(request, {'station':station, 'top_margins':top_margins})

	return render_to_response('trading/station/station.haml', rcontext)