# Imports
from otree.api import *
from otree.api import *
from statistics import NormalDist
import math
import random


# Define Constants
class C(BaseConstants):
    NAME_IN_URL = 'Trading_Experiment'
    PLAYERS_PER_GROUP = None
    NUM_ROUNDS = 1
    # Asset Characteristics
    CURRENT_ASSETS = 10000.00
    NEW_INVEST = 3000.00
    TOTAL_INVEST = CURRENT_ASSETS + NEW_INVEST
    DRIFT_CURR = 0.03
    STDDEV_CURR = 0.015
    DRIFT_NEW = 0.07
    STDDEV_NEW = 0.04
    CORR_A = 0.8
    CORR_B_LOWDIFF = 0.2
    CORR_B_HIGHDIFF = -0.8
    # Cholesky factorization
    CF_A = [1, 0, CORR_A, math.sqrt(1 - CORR_A ** 2)]
    CF_B_LOWDIFF = [1, 0, CORR_B_LOWDIFF, math.sqrt(1 - CORR_B_LOWDIFF ** 2)]
    CF_B_HIGHDIFF = [1, 0, CORR_B_HIGHDIFF, math.sqrt(1 - CORR_B_HIGHDIFF ** 2)]
    # Payoff (Lotterie Tickets)
    LTICKETS_PER_RETURN = 1

    # Non Profit Organisations
    Purpose = ['Gender discrimination', 'Human trafficking', 'Refugees', 'Poverty', 'Animal welfare', 'Environment']
    NonProfit_Name = ['UN Women', 'International Justice Mission', 'UN Refugee Agency', 'The Global Red Cross Network',
                      'PETA', 'WWF']
    NonProfit_Description = [
        'The United Nations Entity for Gender Equality and the Empowerment of Women (UN Women) is the UN agency dedicated to gender equality.',
        'The International Justice Mission is one of the largest organizations fighting human trafficking through policy change and training.',
        'The UNHCR, the UN Refugee Agency, is a global organization dedicated to saving lives, protecting rights and building a better future for refugees, forcibly displaced communities and stateless people.',
        'As poverty causes a lot of hunger, starvation, and unfair suffering, the Red Cross and Red Crescent organizations work together to fight these injustices all around the world.',
        'People for the Ethical Treatment of Animals (PETA) is the largest animal rights organization in the world, and PETA entities have more than 9 million members and supporters globally.',
        'WWF is the world’s largest conservation organization.'
    ]


class Subsession(BaseSubsession):
    pass


class Group(BaseGroup):
    pass


class Player(BasePlayer):
    # Questionnaire
    risk = models.IntegerField(
        choices=[['5', '5 - Willing to accept substantial risk to potentially earn a greater return'],
                 ['4', '4'], ['3', '3'], ['2', '2'], ['1', '1 - Not willing to accept any risk']],
        label='1) Please estimate your willingness to take financial risk',
        widget=widgets.RadioSelect,
    )
    understanding = models.IntegerField(
        choices=[['1', 'Increases'], ['2', 'Decreases'], ['3', 'Stays the same']],
        label='2) When investors spread their money across different (not perfectly correlated) assets,'
              ' the risk of losing a lot of money…',
        widget=widgets.RadioSelect,
    )
    # Demographics
    age = models.IntegerField(label='1) What is your age?', min=13, max=125)
    gender = models.IntegerField(
        choices=[['1', 'Male'], ['2', 'Female'], ['2', 'Diverse']],
        label='2) What is your gender?',
        widget=widgets.RadioSelect,
    )
    education = models.IntegerField(
        choices=[['1', 'Bachelor'], ['2', 'Master'], ['3', 'Doctor'], ['4', 'Different educational program'],
                 ['5', 'In no educational program']],
        label='3) Are you currently in one of the following educational programmes?',
        widget=widgets.RadioSelect,
    )
    semester = models.IntegerField(
        label='4) In which semester (1/2 year) of your current degree programme are you? (Enter 0 if in no educational program)',
        min=0, max=16)
    # Social Preferences
    DonationDecision = models.IntegerField(
        label='3) Imagine the following situation: Today you unexpectedly received 1.000€. How much of this amount would you donate to a good cause? (Values between 0 and 1000 are allowed.)',
        min=0, max=1000)
    WillingnessToGive = models.IntegerField(
        choices=[['7', '7 - Very much'], ['6', '6'], ['5', '5'], ['4', '4'], ['3', '3'], ['2', '2'], ['1', '1 - Not at all']],
        label='4) How willing are you to give to good causes without expecting anything in return?',
        widget=widgets.RadioSelect,
    )
    social_preference = models.IntegerField(
        choices=[['1', 'Gender discrimination'],
                 ['2', 'Human trafficking'],
                 ['3', 'Refugees'],
                 ['4', 'Poverty'],
                 ['5', 'Animal welfare'],
                 ['6', 'Environment']],
        label='Please select only one of the following issues that you care most about.',
        widget=widgets.RadioSelect,
    )
    # Non Profit Organisations Choice
    NP_Purpose = models.StringField
    NP_Choice_Name = models.StringField
    NP_Choice_Description = models.StringField
    # Neutral Treatment
    NT_low_asset_chosen = models.IntegerField(
        choices=[['1', 'Asset A'], ['2', 'Asset B']],
        label='Choose the asset you want to invest in.',
        widget=widgets.RadioSelect,
    )
    NT_high_asset_chosen = models.IntegerField(
        choices=[['1', 'Asset A'], ['2', 'Asset B']],
        label='Choose the asset you want to invest in.',
        widget=widgets.RadioSelect,
    )
    # Treatment
    T_low_asset_chosen = models.IntegerField(
        choices=[['1', 'Asset A'], ['2', 'Asset B']],
        label='Choose the asset you want to invest in.',
        widget=widgets.RadioSelect,
    )
    T_high_asset_chosen = models.IntegerField(
        choices=[['1', 'Asset A'], ['2', 'Asset B']],
        label='Choose the asset you want to invest in.',
        widget=widgets.RadioSelect,
    )
    # Results
    email = models.StringField(
        label='To participate in the lottery of 5 x 20 € Amazon gift cards, please enter your email address below.',
        blank=True)
    donation_info = models.IntegerField(
        blank=True,
        choices=[['1', 'Yes'], ['2', 'No']],
        label='Would you also like to be informed about the final donations via e-mail?',
        widget=widgets.RadioSelect,
    )
    # Performance
    # Round 1
    Round1_Decision = models.StringField()
    Round1_Val = models.FloatField()
    Round1_Ret = models.FloatField()
    Round1_LTickets = models.IntegerField()
    # Round 2
    Round2_Decision = models.StringField()
    Round2_Val = models.FloatField()
    Round2_Ret = models.FloatField()
    Round2_LTickets = models.IntegerField()
    # Round 3
    Round3_Decision = models.StringField()
    Round3_Val = models.FloatField()
    Round3_Ret = models.FloatField()
    Round3_LTickets = models.IntegerField()
    # Round 4
    Round4_Decision = models.StringField()
    Round4_Val = models.FloatField()
    Round4_Ret = models.FloatField()
    Round4_LTickets = models.IntegerField()
    # Financial and Non Financial Outcome
    total_LTickets = models.IntegerField()
    Donation = models.IntegerField()
    Impact = models.StringField()
    # Round Number
    RoundNumber = models.IntegerField(initial=1)
    # Randomisation
    TreatmentGroup = models.IntegerField()
    Pages_Neutral = models.IntegerField()
    Pages_Treatment = models.IntegerField()


# PAGES
class Welcome(Page):
    def before_next_page(player, timeout_happened):  # Randomisation Treatment and High Low Correlation difference order
        player.TreatmentGroup = random.randint(0, 1)
        player.Pages_Neutral = random.randint(1, 2)
        player.Pages_Treatment = random.randint(1, 2)


class Introduction(Page):
    pass

class Introduction2(Page):
    pass


class Questionnaire(Page):
    form_model = 'player'
    form_fields = ['risk', 'understanding', 'DonationDecision', 'WillingnessToGive']


class SocialPreferences(Page):
    form_model = 'player'
    form_fields = ['social_preference']


class SocialPreferencesResult_Pos(Page):
    form_model = 'player'

    @staticmethod
    def vars_for_template(player: Player): # Converting the index of the social preference choice into name and descriptions
        player.NP_Purpose = C.Purpose[player.social_preference - 1]
        player.NP_Choice_Name = C.NonProfit_Name[player.social_preference - 1]
        player.NP_Choice_Description = C.NonProfit_Description[player.social_preference - 1]

    def is_displayed(player): # Show this page only in the positive treatment group
        return player.TreatmentGroup == 1


class SocialPreferencesResult_Neg(Page):
    form_model = 'player'

    @staticmethod
    def vars_for_template(player: Player): # Converting the index of the social preference choice into name and descriptions
        player.NP_Purpose = C.Purpose[player.social_preference - 1]
        player.NP_Choice_Name = C.NonProfit_Name[player.social_preference - 1]
        player.NP_Choice_Description = C.NonProfit_Description[player.social_preference - 1]

    def is_displayed(player): # Show this page only in the negative treatment group
        return player.TreatmentGroup == 0


class Demographics(Page):
    form_model = 'player'
    form_fields = ['age', 'gender', 'education', 'semester']

    # Calculate results already in this function to avoid generating new return values every time a participant
    # refreshes the results page when the results were calculated in that class
    def before_next_page(player, timeout_happened):
        # Following Arrays: Index for each Trading Round
        Asset_Choice = [player.NT_low_asset_chosen, player.NT_high_asset_chosen, player.T_high_asset_chosen,
                        player.T_low_asset_chosen]
        Asset_Choice_String = ['None'] * 4
        Portfolio_Value = [0] * 4
        Portfolio_Return = [0] * 4
        Lottery_Tickets = [0] * 4
        player.Donation = 0
        HL_Seq = []
        # Create an array with the sequence of high and low correlation differences
        if player.Pages_Neutral == 5:
            HL_Seq.append(0)
            HL_Seq.append(1)
        else:
            HL_Seq.append(1)
            HL_Seq.append(0)
        if player.Pages_Treatment == 5:
            HL_Seq.append(0)
            HL_Seq.append(1)
        else:
            HL_Seq.append(1)
            HL_Seq.append(0)
        # Calculate Results for each Trading Round
        for x in range(4):
            # Generate random, normal distributed values for high and low correlation difference
            e_curr = NormalDist(mu=0, sigma=1).inv_cdf(random.random())
            e_esg = NormalDist(mu=0, sigma=1).inv_cdf(random.random())
            e_non_esg = NormalDist(mu=0, sigma=1).inv_cdf(random.random())
            # Create the target correlation
            if HL_Seq[x] == 1:
                ESG = [e_curr * C.CF_A[0] + e_esg * C.CF_A[1], e_curr * C.CF_A[2] + e_esg * C.CF_A[3]]
                NON_ESG = [e_curr * C.CF_B_HIGHDIFF[0] + e_non_esg * C.CF_B_HIGHDIFF[1],
                           e_curr * C.CF_B_HIGHDIFF[2] + e_non_esg * C.CF_B_HIGHDIFF[3]]
            else:
                ESG = [e_curr * C.CF_A[0] + e_esg * C.CF_A[1], e_curr * C.CF_A[2] + e_esg * C.CF_A[3]]
                NON_ESG = [e_curr * C.CF_B_LOWDIFF[0] + e_non_esg * C.CF_B_LOWDIFF[1],
                           e_curr * C.CF_B_LOWDIFF[2] + e_non_esg * C.CF_B_LOWDIFF[3]]
            # Create the target drift and standard deviation
            Current_return = C.DRIFT_CURR + ESG[0] * C.STDDEV_CURR
            ESG_return = C.DRIFT_NEW + ESG[1] * C.STDDEV_NEW
            NON_ESG_return = C.DRIFT_NEW + NON_ESG[1] * C.STDDEV_NEW
            # Compute resulting asset value
            Current_abs = C.CURRENT_ASSETS * (1 + Current_return)
            ESG_abs = C.NEW_INVEST * (1 + ESG_return)
            NON_ESG_abs = C.NEW_INVEST * (1 + NON_ESG_return)
            # Compute resulting portfolio values
            if Asset_Choice[x] == 1:
                Portfolio_Value[x] = Current_abs + ESG_abs
                Portfolio_Return[x] = ((Current_abs + ESG_abs) / (C.TOTAL_INVEST) - 1) * 100
                Asset_Choice_String[x] = 'Asset A'
                if x > 1:
                    player.Donation = player.Donation + 1
            else:
                Portfolio_Value[x] = Current_abs + NON_ESG_abs
                Portfolio_Return[x] = ((Current_abs + NON_ESG_abs) / (C.TOTAL_INVEST) - 1) * 100
                Asset_Choice_String[x] = 'Asset B'
            # Compute resulting number of lottery ticket and direction of impact
            Lottery_Tickets[x] = Portfolio_Return[x] * 10
        if player.TreatmentGroup == 1:
            player.Impact = 'additional'
        else:
            player.Impact = 'deducted'
            player.Donation = 2 - player.Donation # Negative Treatment case: convert to number of deducted votes
        # Store Performance data
        # Round 1
        player.Round1_Decision = Asset_Choice_String[0]
        player.Round1_Val = round(Portfolio_Value[0], 2)
        player.Round1_Ret = round(Portfolio_Return[0], 2)
        player.Round1_LTickets = int(round(Lottery_Tickets[0], 0))
        # Round 2
        player.Round2_Decision = Asset_Choice_String[1]
        player.Round2_Val = round(Portfolio_Value[1], 2)
        player.Round2_Ret = round(Portfolio_Return[1], 2)
        player.Round2_LTickets = int(round(Lottery_Tickets[1], 0))
        # Round3
        player.Round3_Decision = Asset_Choice_String[2]
        player.Round3_Val = round(Portfolio_Value[2], 2)
        player.Round3_Ret = round(Portfolio_Return[2], 2)
        player.Round3_LTickets = int(round(Lottery_Tickets[2], 0))
        # Round4
        player.Round4_Decision = Asset_Choice_String[3]
        player.Round4_Val = round(Portfolio_Value[3], 2)
        player.Round4_Ret = round(Portfolio_Return[3], 2)
        player.Round4_LTickets = int(round(Lottery_Tickets[3], 0))
        # Calculate total number of lottery tickets
        player.total_LTickets = player.Round1_LTickets + player.Round2_LTickets + player.Round3_LTickets + player.Round4_LTickets


class EndPage(Page):
    pass


class NeutralTreatment_lowdiff(Page):
    form_model = "player"
    form_fields = ["NT_low_asset_chosen"]

    @staticmethod
    def before_next_page(player, timeout_happened):
        player.RoundNumber = player.RoundNumber + 1
        player.Pages_Neutral = player.Pages_Neutral + 2

    def is_displayed(player): # Show this page as either first or second trading round in the neutral treatment case
        return player.Pages_Neutral == 1 or player.Pages_Neutral == 4


class NeutralTreatment_highdiff(Page):
    form_model = "player"
    form_fields = ["NT_high_asset_chosen"]

    @staticmethod
    def before_next_page(player, timeout_happened):
        player.RoundNumber = player.RoundNumber + 1
        player.Pages_Neutral = player.Pages_Neutral + 2

    def is_displayed(player): # Show this page as either first or second trading round in the neutral treatment case
        return player.Pages_Neutral == 2 or player.Pages_Neutral == 3


class PositiveTreatment_lowdiff(Page):
    form_model = "player"
    form_fields = ["T_low_asset_chosen"]

    @staticmethod
    def before_next_page(player, timeout_happened):
        player.RoundNumber = player.RoundNumber + 1
        player.Pages_Treatment = player.Pages_Treatment + 2

    def is_displayed(player): # Show this page as either first or second trading round only in the positive treatment case
        return player.TreatmentGroup == 1 and (player.Pages_Treatment == 1 or player.Pages_Treatment == 4)


class PositiveTreatment_highdiff(Page):
    form_model = "player"
    form_fields = ["T_high_asset_chosen"]

    @staticmethod
    def before_next_page(player, timeout_happened):
        player.RoundNumber = player.RoundNumber + 1
        player.Pages_Treatment = player.Pages_Treatment + 2

    def is_displayed(player): # Show this page as either first or second trading round only in the positive treatment case
        return player.TreatmentGroup == 1 and (player.Pages_Treatment == 2 or player.Pages_Treatment == 3)


class NegativeTreatment_lowdiff(Page):
    form_model = "player"
    form_fields = ["T_low_asset_chosen"]

    def before_next_page(player, timeout_happened):
        player.RoundNumber = player.RoundNumber + 1
        player.Pages_Treatment = player.Pages_Treatment + 2

    def is_displayed(player): # Show this page as either first or second trading round only in the negative treatment case
        return player.TreatmentGroup == 0 and (player.Pages_Treatment == 1 or player.Pages_Treatment == 4)


class NegativeTreatment_highdiff(Page):
    form_model = "player"
    form_fields = ["T_high_asset_chosen"]

    @staticmethod
    def before_next_page(player, timeout_happened):
        player.RoundNumber = player.RoundNumber + 1
        player.Pages_Treatment = player.Pages_Treatment + 2

    def is_displayed(player): # Show this page as either first or second trading round only in the negative treatment case
        return player.TreatmentGroup == 0 and (player.Pages_Treatment == 2 or player.Pages_Treatment == 3)


class Results(Page):
    form_model = 'player'
    form_fields = ['email', 'donation_info']


page_sequence = [
    Welcome,
    Introduction,
    Introduction2,
    Questionnaire,
    NeutralTreatment_lowdiff,
    NeutralTreatment_highdiff,
    NeutralTreatment_lowdiff,
    NeutralTreatment_highdiff,
    SocialPreferences,
    SocialPreferencesResult_Neg,
    NegativeTreatment_lowdiff,
    NegativeTreatment_highdiff,
    NegativeTreatment_lowdiff,
    NegativeTreatment_highdiff,
    SocialPreferencesResult_Pos,
    PositiveTreatment_lowdiff,
    PositiveTreatment_highdiff,
    PositiveTreatment_lowdiff,
    PositiveTreatment_highdiff,
    Demographics,
    Results,
    EndPage
]