- split +ve and -ve contributions

- update code to ensure -ve voting is taken into account

app/grants/clr.py

@@ -153,7 +153,12 @@ def calculate_new_clr(aggregated_contributions, pair_totals, threshold=0.0, tota
 '''
 def calculate_new_clr_final(totals_pos, totals_neg, total_pot=0.0):
     # calculate final totals
-    totals = [{'id': x['id'], 'clr_amount': (math.sqrt(x['clr_amount']) - math.sqrt(y['clr_amount']))**2} for x in totals_pos for y in totals_neg if x['id'] == y['id']]
+
+    if len(totals_neg) == 0:
+        totals = totals_pos
+    else:
+        totals = [{'id': x['id'], 'clr_amount': (math.sqrt(x['clr_amount']) - math.sqrt(y['clr_amount']))**2} for x in totals_pos for y in totals_neg if x['id'] == y['id']]
+
     for x in totals:
         if x['clr_amount'] < 0:
             x['clr_amount'] = 0
@@ -171,7 +176,6 @@ def calculate_new_clr_final(totals_pos, totals_neg, total_pot=0.0):
     return bigtot, totals
 
 
-
 '''
     Clubbed function that intakes grant data, calculates necessary intermediate calculations, and spits out clr calculations. This function is re-used for positive and negative contributions
 
@@ -224,20 +228,18 @@ def grants_clr_calculate_pos_neg(pos_totals, neg_totals, total_pot=0.0):
     return final_bigtot, final_totals
 
 
-def calculate_clr_for_donation(grant, amount, grant_contributions, total_pot, threshold):
-
-    ### ADD NEW GRANT CONTRIBUTIONS MODEL IS UPDATED AND PULL SEPARATE POSITIVE AND NEGATIVE CONTRIBUTIONS, INPUT VARIABLES IN THE METHOD grants_clr_calculate LINES 292 AND ON WILL NEED TO BE CHANGED
+def calculate_clr_for_donation(grant, amount, positive_grant_contributions, negative_grant_contributions, total_pot, threshold):
 
-    _grant_contributions = copy.deepcopy(grant_contributions)
+    _positive_grant_contributions = copy.deepcopy(positive_grant_contributions)
     # find grant in contributions list and add donation
     if amount != 0:
-        for grant_contribution in _grant_contributions:
+        for grant_contribution in _positive_grant_contributions:
             if grant_contribution['id'] == grant.id:
                 # add this donation with a new profile (id 99999999999) to get impact
                 grant_contribution['contributions'].append({'999999999999': amount})
 
-    pos_totals = grants_clr_calculate(_grant_contributions, total_pot=total_pot, threshold=threshold, positive=True)
-    neg_totals = grants_clr_calculate(_grant_contributions, total_pot=total_pot, threshold=threshold, positive=False)
+    pos_totals = grants_clr_calculate(_positive_grant_contributions, total_pot=total_pot, threshold=threshold, positive=True)
+    neg_totals = grants_clr_calculate(negative_grant_contributions, total_pot=total_pot, threshold=threshold, positive=False)
     _, grants_clr = grants_clr_calculate_pos_neg(pos_totals, neg_totals, total_pot=total_pot)
 
     # find grant we added the contribution to and get the new clr amount
@@ -258,7 +260,8 @@ def calculate_clr_for_donation(grant, amount, grant_contributions, total_pot, th
 
     Returns:
         grants: list of grants based on clr_type
-        contrib_data: [{'id': <int>, 'contributions': <Object>}]
+        positive_contrib_data: [{'id': <int>, 'contributions': <Object>}]
+        negative_contrib_data: [{'id': <int>, 'contributions': <Object>}]
         total_pot: float
         threshold : int
 '''
@@ -285,44 +288,70 @@ def populate_data_for_clr(clr_type=None, network='mainnet'):
         return None, None, None, None
 
     # set up data to load contributions for each grant
-    contrib_data = []
+    positive_contrib_data = []
+    negative_contrib_data = []
 
     for grant in grants:
-        # go through all the individual contributions for each grant
-        g_contributions = copy.deepcopy(contributions).filter(subscription__grant_id=grant.id)
-
-        # put in correct format
-        phantom_funding_profiles = PhantomFunding.objects.filter(grant_id=grant.id, created_on__gte=CLR_START_DATE, created_on__lte=from_date)
-        all_contributing_profile_ids = list(set([c.subscription.contributor_profile.id for c in g_contributions] + [p.profile_id for p in phantom_funding_profiles]))
-        all_summed_contributions = []
+        grant_id = grant.defer_clr_to.pk if grant.defer_clr_to else grant.id
 
-        for profile_id in all_contributing_profile_ids:
-            # get sum of contributions per grant for each profile
-            profile_g_contributions = g_contributions.filter(subscription__contributor_profile_id=profile_id)
-            sum_of_each_profiles_contributions = float(sum([c.subscription.get_converted_monthly_amount() for c in profile_g_contributions]))
+        # Get the +ve and -ve contributions
+        positive_contributions = copy.deepcopy(contributions).filter(subscription__grant_id=grant.id, subscription__match_direction='+')
+        negative_contributions = copy.deepcopy(contributions).filter(subscription__grant_id=grant.id, subscription__match_direction='-')
 
-            phantom_funding = PhantomFunding.objects.filter(created_on__gte=CLR_START_DATE, grant_id=grant.id, profile_id=profile_id, created_on__lte=from_date)
-            if phantom_funding.exists():
-                sum_of_each_profiles_contributions = sum_of_each_profiles_contributions + phantom_funding.first().value
+        # Generate list of profiles who've made +ve and -ve contributions to the grant
+        phantom_funding_profiles = PhantomFunding.objects.filter(grant_id=grant.id, created_on__gte=CLR_START_DATE, created_on__lte=from_date)
+        positive_contributing_profile_ids = list(set([c.subscription.contributor_profile.id for c in positive_contributions] + [p.profile_id for p in phantom_funding_profiles]))
+        negative_contributing_profile_ids = list(set([c.subscription.contributor_profile.id for c in negative_contributions]))
+
+        # print(f'positive contrib profiles : {positive_contributing_profile_ids}')
+        # print(f'negative contrib profiles : {negative_contributing_profile_ids}')
+        # print(f'positive contributions : {positive_contributions}')
+        # print(f'negative contributions : {negative_contributions}')
+
+        positive_summed_contributions = []
+        negative_summed_contributions = []
+
+        # POSITIVE CONTRIBUTIONS
+        if len(positive_contributing_profile_ids) > 0:
+            for profile_id in positive_contributing_profile_ids:
+                # get sum of contributions per grant for each profile
+                profile_positive_contributions = positive_contributions.filter(subscription__contributor_profile_id=profile_id)
+                sum_of_each_profiles_contributions = float(sum([c.subscription.get_converted_monthly_amount() for c in profile_positive_contributions]))
+
+                phantom_funding = PhantomFunding.objects.filter(created_on__gte=CLR_START_DATE, grant_id=grant.id, profile_id=profile_id, created_on__lte=from_date)
+                if phantom_funding.exists():
+                    sum_of_each_profiles_contributions = sum_of_each_profiles_contributions + phantom_funding.first().value
+
+                positive_summed_contributions.append({str(profile_id): sum_of_each_profiles_contributions})
+
+            # for each grant, list the contributions in key value pairs like {'profile id': sum of contributions}
+            positive_contrib_data.append({
+                'id': grant_id,
+                'contributions': positive_summed_contributions
+            })
 
-            all_summed_contributions.append({str(profile_id): sum_of_each_profiles_contributions})
+        # NEGATIVE CONTRIBUTIONS
+        if len(negative_contributing_profile_ids) > 0:
+            for profile_id in negative_contributing_profile_ids:
+                profile_negative_contributions = negative_contributions.filter(subscription__contributor_profile_id=profile_id)
+                sum_of_each_negative_contributions = float(sum([c.subscription.get_converted_monthly_amount() for c in profile_negative_contributions]))
+                negative_summed_contributions.append({str(profile_id): sum_of_each_negative_contributions})
 
-        # for each grant, list the contributions in key value pairs like {'profile id': sum of contributions}
-        grant_id = grant.defer_clr_to.pk if grant.defer_clr_to else grant.id
-        contrib_data.append({
-            'id': grant_id,
-            'contributions': all_summed_contributions
-        })
+            negative_contrib_data.append({
+                'id': grant_id,
+                'contributions': negative_summed_contributions
+            })
 
-    #print(f'\n contributions data: {contrib_data} \n')
-    return (grants, contrib_data, total_pot, threshold)
+    # print(f'\n positive contributions data: {positive_contrib_data} \n')
+    # print(f'\n negative contributions data: {negative_contrib_data} \n')
+    return (grants, positive_contrib_data, negative_contrib_data, total_pot, threshold)
 
 
 def predict_clr(save_to_db=False, from_date=None, clr_type=None, network='mainnet'):
     # setup
     clr_calc_start_time = timezone.now()
     debug_output = []
-    grants, contrib_data, total_pot, threshold = populate_data_for_clr(clr_type, network)
+    grants, positive_contrib_data, negative_contrib_data, total_pot, threshold = populate_data_for_clr(clr_type, network)
 
     # calculate clr given additional donations
     for grant in grants:
@@ -332,7 +361,14 @@ def predict_clr(save_to_db=False, from_date=None, clr_type=None, network='mainne
 
         for amount in potential_donations:
             # calculate clr with each additional donation and save to grants model
-            predicted_clr, grants_clr = calculate_clr_for_donation(grant, amount, contrib_data, total_pot, threshold)
+            predicted_clr, grants_clr = calculate_clr_for_donation(
+                grant,
+                amount,
+                positive_contrib_data,
+                negative_contrib_data,
+                total_pot,
+                threshold
+            )
             potential_clr.append(predicted_clr)
 
         if save_to_db:
@@ -372,29 +408,40 @@ def predict_clr(save_to_db=False, from_date=None, clr_type=None, network='mainne
         grant: <Grant> - grant to be funded
         contributor: <Profile> - contributor making a contribution
         amount: <float> - amount with which is grant is to be funded
-        grant_contributions: [Object]
+        match_direction: <char> + , -
+        positive_grant_contributions: [Object]
+        negative_grant_contributions: [Object]
         total_pot: <float>
         threshold: <float>
 
     Returns:
         clr_amount
 '''
-def calculate_clr_for_donation_live(grant, contributor, amount, grant_contributions, total_pot, threshold):
+def calculate_clr_for_donation_live(grant, contributor, amount, match_direction, positive_grant_contributions, negative_grant_contributions, total_pot, threshold):
 
     if amount == 0:
         return 0
 
-    _grant_contributions = copy.deepcopy(grant_contributions)
+    _positive_grant_contributions = copy.deepcopy(positive_grant_contributions)
+    _negative_grant_contributions = copy.deepcopy(negative_grant_contributions)
+
     profile_id = str(contributor.pk)
 
-    for grant_contribution in _grant_contributions:
-        if grant_contribution['id'] == grant.id:
-            grant_contribution['contributions'].append({
-                profile_id: amount
-            })
+    if match_direction == '+':
+        for grant_contribution in _positive_grant_contributions:
+            if grant_contribution['id'] == grant.id:
+                grant_contribution['contributions'].append({
+                    profile_id: amount
+                })
+    else:
+        for grant_contribution in _negative_grant_contributions:
+            if grant_contribution['id'] == grant.id:
+                grant_contribution['contributions'].append({
+                    profile_id: amount
+                })
 
-    pos_totals = grants_clr_calculate(_grant_contributions, total_pot=total_pot, threshold=threshold, positive=True)
-    neg_totals = grants_clr_calculate(_grant_contributions, total_pot=total_pot, threshold=threshold, positive=False)
+    pos_totals = grants_clr_calculate(_positive_grant_contributions, total_pot=total_pot, threshold=threshold, positive=True)
+    neg_totals = grants_clr_calculate(_negative_grant_contributions, total_pot=total_pot, threshold=threshold, positive=False)
     _, grants_clr = grants_clr_calculate_pos_neg(pos_totals, neg_totals, total_pot=total_pot)
 
     # find grant we added the contribution to and get the new clr amount
@@ -414,11 +461,12 @@ def calculate_clr_for_donation_live(grant, contributor, amount, grant_contributi
         grant: <Grant>
         contributor: <Profile>
         amount: <float>
+        match_direction: <char> + , -
 
     Returns:
         predicted_clr_match
 '''
-def predict_clr_live(grant, contributor, amount):
+def predict_clr_live(grant, contributor, amount, match_direction='+'):
 
     if not grant or not contributor:
         print('error: predict_clr_live - missing parameters')
@@ -429,10 +477,10 @@ def predict_clr_live(grant, contributor, amount):
 
     clr_type = grant.grant_type
     network = grant.network
-    _, contrib_data, total_pot, threshold = populate_data_for_clr(clr_type, network)
+    _, positive_contrib_data, negative_contrib_data , total_pot, threshold = populate_data_for_clr(clr_type, network)
 
     predicted_clr_match = calculate_clr_for_donation_live(
-        grant, contributor, amount, contrib_data, total_pot, threshold
+        grant, contributor, amount, match_direction, positive_contrib_data, negative_contrib_data, total_pot, threshold
     )
 
     return predicted_clr_match

app/grants/views.py

@@ -1061,7 +1061,8 @@ def grant_categories(request):
 def predict_clr_v1(request, grant_id):
     '''
         {
-            amount: <float>
+            amount: <float>,
+            match_direction: <char> +, -
         }
     '''
     response = {
@@ -1092,7 +1093,15 @@ def predict_clr_v1(request, grant_id):
         response['message'] = 'error: missing parameter amount'
         return JsonResponse(response)
 
-    predicted_clr_match = predict_clr_live(grant, profile, float(amount))
+
+    match_direction = request.GET.get('match_direction', '+')
+
+    predicted_clr_match = predict_clr_live(
+        grant,
+        profile,
+        float(amount),
+        match_direction
+    )
 
     response = {
         'status': 200,