# WVLottopy: Matteo DiBiagio
from fractions import Fraction as frac
import pandas as pd
import requests
from collections import Counter


def get_data():
    url = 'https://gateway.loyalty.wvlottery.com/services/jackpot/api/v1/jackpot-results?gameId=9&jackpotStatus=PAYABLE&size=500&sort=externalId,drawDate,desc&page='
    header = {
        "User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.75 Safari/537.36",
    }
    # Get web data, the new site pulls past draws from json 500 at a time
    web = {'Date': [], 'Numbers': []}
    page = 0
    while True:
        r = requests.get(url + str(page), headers=header)
        data = r.json()
        for x in data['content']:
            web['Date'].append(x['drawingDate'])
            web['Numbers'].append('–'.join(str(n['data']) for n in x['resultData'] if n['type'] == 'REGULAR'))
        if data['last']:
            break
        page += 1
    dfs = [pd.DataFrame(web)]
    pd.set_option('display.max_rows', None)
    # Specifies no max rows, otherwise only shows 10 records
    df = dfs[0]
    df2 = df[['Date', 'Numbers']]
    date = list(df2['Date']) 
    nums = list(df2['Numbers'].astype('str')) 
    return date, nums

date, nums = get_data()

# Formatting 
hyphenfree = []
for x in nums:
    hyphenfree.append(x.replace('–',', ')) 
splitlist = ", ".join(hyphenfree)
sep = splitlist.split(", ")

for n in range(0, 10):
    most_common= Counter(sep).most_common(3)
    likely_nums = [v[0] for v in most_common]
    frequency = [v[-1] for v in most_common]

sorted_nums = sorted(likely_nums, key=lambda x: (len(x), x))
total_freq = sum(frequency) 
Chance  = frac(total_freq, 15000) # Chance = number call freq / all possible numbers i.e. 1000
Forecast = str(" - ".join(sorted_nums))

print(f"Likely numbers are . . .  {Forecast} \n"
f"With percent chance of winning being {Chance} \n"
f"DISCLAIMER: This set's chance is innacurate due to how many numbers are drawn per day.")