app.py

print('Running')
import time
import requests
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
from matplotlib.pyplot import figure
from matplotlib.offsetbox import OffsetImage, AnnotationBbox
from scipy import stats
import matplotlib.lines as mlines
import matplotlib.transforms as mtransforms
import numpy as np
import time
#import plotly.express as px
#!pip install chart_studio
#import chart_studio.tools as tls
from bs4 import BeautifulSoup
import matplotlib.pyplot as plt
import numpy as np
import matplotlib.font_manager as font_manager
from datetime import datetime
import pytz
from matplotlib.ticker import MaxNLocator
from matplotlib.patches import Ellipse
import matplotlib.transforms as transforms
from matplotlib.gridspec import GridSpec
datetime.now(pytz.timezone('US/Pacific')).strftime('%B %d, %Y')
# Configure Notebook
#%matplotlib inline
plt.style.use('fivethirtyeight')
sns.set_context("notebook")
import warnings
warnings.filterwarnings('ignore')
# import yfpy
# from yfpy.query import YahooFantasySportsQuery
# import yahoo_oauth
import json
#import openpyxl
# from sklearn import preprocessing
from datetime import timedelta
# import dataframe_image as dfi
# from google.colab import drive
def percentile(n):
    def percentile_(x):
        return np.percentile(x, n)
    percentile_.__name__ = 'percentile_%s' % n
    return percentile_

# import os
# import praw
# import matplotlib.pyplot as plt
# import matplotlib.colors
# import matplotlib.colors as mcolors
# cmap_sum = matplotlib.colors.LinearSegmentedColormap.from_list("", ["#4285f4","#FFFFFF","#F0E442"])
#import pybaseball
import math
# import matplotlib.ticker as mtick
# import matplotlib.ticker as ticker

# colour_palette = ['#FFB000','#648FFF','#785EF0',
                #   '#DC267F','#FE6100','#3D1EB2','#894D80','#16AA02','#B5592B','#A3C1ED']
# import matplotlib.colors as mcolors
# from  matplotlib.ticker import FuncFormatter
# from matplotlib.font_manager import FontProperties

import numpy as np
# import matplotlib.pyplot as plt
import matplotlib.colors

# import undetected_chromedriver as uc
# from selenium import webdriver
# from seleniumbase import Driver

# driver = Driver(uc=True)
# driver.get('https://www.naturalstattrick.com')

#x,y,c = zip(*np.random.rand(30,3)*4-2)

#norm=plt.Normalize(-2,2)
#co = matplotlib.colors.LinearSegmentedColormap.from_list("", ["#ffffff","#F0E442"])
co = matplotlib.colors.LinearSegmentedColormap.from_list("", ["#56B4E9","#FFFFFF","#F0E442"])
#co = matplotlib.colors.LinearSegmentedColormap.from_list("", ["#FFFFFF","#56B4E9"])




# try: 
#     data_r = requests.get("https://pub-api-ro.fantasysports.yahoo.com/fantasy/v2/league/427.l.public;out=settings/players;position=ALL;start=0;count=3000;sort=rank_season;search=;out=percent_owned;out=auction_values,ranks;ranks=season;ranks_by_position=season;out=expert_ranks;expert_ranks.rank_type=projected_season_remaining/draft_analysis;cut_types=diamond;slices=last7days?format=json_f").json()
#     key_check = data_r['fantasy_content']['league']['players']

# except KeyError:
#     data_r = requests.get("https://pub-api-ro.fantasysports.yahoo.com/fantasy/v2/league/427.l.public;out=settings/players;position=ALL;start=0;count=400;sort=rank_season;search=;out=percent_owned;out=auction_values,ranks;ranks=season;ranks_by_position=season;out=expert_ranks;expert_ranks.rank_type=projected_season_remaining/draft_analysis;cut_types=diamond;slices=last7days?format=json_f").json()
#     print('key_checked')

# total_list = []

# for x in data_r['fantasy_content']['league']['players']:
#     single_list = []

#     single_list.append(int(x['player']['player_id']))
#     single_list.append(int(x['player']['player_ranks'][0]['player_rank']['rank_value']))
#     single_list.append(x['player']['name']['full'])
#     single_list.append(x['player']['name']['first'])
#     single_list.append(x['player']['name']['last'])
#     single_list.append(x['player']['draft_analysis']['average_pick'])
#     single_list.append(x['player']['average_auction_cost'])
#     single_list.append(x['player']['display_position'])
#     single_list.append(x['player']['editorial_team_abbr'])
#     if 'value' in x['player']['percent_owned']:
#         single_list.append(x['player']['percent_owned']['value']/100)
#     else:
#         single_list.append(0)
#     total_list.append(single_list)

yahoo_df = pd.read_csv('df_2023_small.csv',index_col=[0])
yahoo_df.percent_owned = yahoo_df.percent_owned.astype(float)
#yahoo_df_scrape.columns = ['yahoo_id','idx','full','first','last','average_pick','average_auction_cost','projected_auction_value','position','team','percent_owned','status']
#yahoo_df_scrape.status = yahoo_df_scrape.status.astype(str)


#yahoo_df = pd.DataFrame(total_list,columns = ['player_id','rank_value','full','first','last','average_pick','average_auction_cost','display_position','editorial_team_abbr','percent_owned'])

# yahoo_df = pd.read_csv('df_2023_small.csv',index_col=[0],usecols=range(12))
# yahoo_df.columns = ['rank_value','player_id','full','first','last','average_pick', 'average_cost','display_position','projected_auction_value','editorial_team_abbr','percent_owned']

yahoo_df_2 = yahoo_df.copy()

# # Write your code here.
# response = requests.get("https://www.naturalstattrick.com/playerlist.php?fromseason=20232024&thruseason=20232024&stype=2&sit=all&stdoi=oi&rate=n")
# soup = BeautifulSoup(response.text, 'html.parser')
# table_rows = soup.findAll('tr')
# table_rows = table_rows[1:-1]
# table_rows[0].findAll('td')

# player_name = []
# player_position = []
# player_team = []
# player_id = []

# for i in range(0,len(table_rows)-1):
#     player_name.append(str(table_rows[i].findAll('td')[0].contents[0]))
#     player_position.append(table_rows[i].findAll('td')[1].contents[0])
#     player_team.append(table_rows[i].findAll('td')[2].contents[0])
#     player_id.append(str(table_rows[i].findAll('td')[3].contents[0])[-76:][:7])

# player_id_df = pd.DataFrame({'Player':player_name,'Player ID':player_id,'Position':player_position,'Team':player_team})
# #player_id_df.index.name = 'Player Name'
# player_id_df.head()

# skater_df = player_id_df[player_id_df['Position'] != 'G']
# goalie_df = player_id_df[player_id_df['Position'] == 'G']

season = 20232024
seasontype = 2

def nat_stat_trick_range_pp_gp(rookie='n',start_date='2022-10-01',end_date=str(pd.to_datetime(datetime.now(pytz.timezone('US/Pacific')).strftime('%Y-%m-%d')).date()),sit='all',gp=1):
    time.sleep(2)
    url = f'https://www.naturalstattrick.com/playerteams.php?fromseason={season}&thruseason={season}&stype={seasontype}&sit=pp&score=all&stdoi=std&rate=y&team=ALL&pos=S&loc=B&toi=0&gpfilt=gpteam&fd=&td=&tgp='+str(gp)+'&lines=single&draftteam=ALL'

    player_list_all = []
    response = requests.get(url)
    soup = BeautifulSoup(response.text, 'html.parser')
    table_rows = soup.findAll('tr')
    table_rows = table_rows[1:]

    for j in range(0,len(table_rows)):
      p_string = [str(x).strip('<td>').strip('</') for x  in list(table_rows[j].findAll('td')) if "<td>" in str(x)]
      player_list_all.append([p_string[0]]+[str(table_rows[j].findAll('td')[1]).split('>')[2].split('<')[0]]+p_string[1:]+[str(table_rows[j].findAll('td')[1])[98:105].strip('</a></td>')])
    #table_rows[0].findAll('td')

    if soup != "":
      columns_list = [str(x).split('>')[1].split('<')[0] for x in soup.findAll('th')]+['player_id']
      df_url = pd.DataFrame(data=player_list_all,columns=columns_list)

    df_url = df_url.fillna(0)
    df_url['Shots+Hits+Blocks/60'] = df_url['Shots/60'].astype(float)+df_url['Hits/60'].astype(float)+df_url['Shots Blocked/60'].astype(float)
    df_url['Shots+Hits/60'] = df_url['Shots/60'].astype(float)+df_url['Hits/60'].astype(float)
    #print(url)
    return df_url

team_abv = pd.read_csv('team_abv.csv')
team_dict = team_abv.set_index('team_abv').to_dict()['team_name']

yahoo_nhl_df = pd.read_csv('yahoo_to_nhl.csv', encoding='unicode_escape')


player_games_df_old = pd.read_csv('player_games_cards.csv',index_col=[0]).sort_values(by='date',ascending=False)
team_games_df = pd.read_csv('team_games.csv',index_col=[0])#.sort_values(by='date',ascending=False)

player_games_df_2 = player_games_df_old.copy()

team_games_df['game_count'] = team_games_df.groupby('team')['team'].cumcount()+1
team_games_df['max_games'] = team_games_df.groupby('team').game_count.transform('max')
team_games_df['abv'] = team_games_df.team.map(team_abv.set_index('team_name')['team_abv'].to_dict())
team_games_df = team_games_df.sort_values(by='game_count',ascending=False)
#team_abv = pd.read_csv('team_abv.csv')



def nat_stat_convert(df):
    for i in range(0,len(df.columns)):
        if df.columns[i][-3:]=='/60':
            if 'ix' not in df.columns[i]:
                df[df.columns[i]] = np.round(df[df.columns[i]].astype(float)*df['TOI'].astype(float)/60,0)
                df = df.rename(columns={df.columns[i]: df.columns[i].replace('/60','')})
            else:
                df[df.columns[i]] = df[df.columns[i]].astype(float)*df['TOI'].astype(float)/60
                df = df.rename(columns={df.columns[i]: df.columns[i].replace('/60','')})

    df['Faceoffs %'] = df['Faceoffs Won']/(df['Faceoffs Won']+df['Faceoffs Lost'])

    return df

from shiny import ui, render, App
import matplotlib.image as mpimg
app_ui = ui.page_fluid(
    #ui.panel_title("Simulate a normal distribution"),

    ui.layout_sidebar(

      ui.panel_sidebar(
        #ui.input_date_range("date_range_id", "Date range input",start = statcast_df.game_date.min(), end = statcast_df.game_date.max()),
        ui.input_select("team_id", "Select Team",team_dict,width=1,size=1,selected='ANA',),
        ui.input_numeric("n_1", "Last Games x", value=1),
        ui.input_numeric("n_2", "Last Games y", value=0),
        ui.input_numeric("n_3", "Last Games z", value=0),
        ui.input_numeric("top_n", "Show top 'n'", value=10),
        ui.input_switch("x", "Drop N/A"),
        #ui.input_select("ignore_id", "Remove Columns",['Position','Roster%'],multiple=True,selectize=True),
      ),

        ui.panel_main(ui.tags.h3(""),
                      ui.div({"style": "font-size:2.7em;"},ui.output_text("txt_title")),
                      #ui.tags.h2("Fantasy Hockey Schedule Summary"),
                      ui.tags.h5("Created By: @TJStats,     Data: Natural Stat Trick, Yahoo Fantasy"),
                      ui.div({"style": "font-size:1.6em;"},ui.output_text("txt")),
                      ui.output_table("pp_roundup"),
                      #ui.tags.h5('Legend'),
                     #ui.tags.h6('An Off Night is defined as a day in which less than half the teams in the NHL are playing'),
                      #ui.tags.h6('The scores are determined by using games played, off-nights, B2B, and strength of opponents') )
        )
    ),
)




from urllib.request import Request, urlopen
from shiny import App, reactive, ui
from shiny.ui import h2, tags
# importing OpenCV(cv2) module




#print(app_ui)
def server(input, output, session):


    @output
    @render.text
    def txt():
        return   f'{team_dict[input.team_id()]} Last Games PP Summary'
      
    @output
    @render.text
    def txt_title():
    
        return f'Team Last Games PP% Leaders'





    @output
    @render.table
    def pp_roundup():

        top_n = input.top_n()
        n_1 = input.n_1()
        n_2 = input.n_2()
        n_3 = input.n_3()

        list_of_columns =  ['Player', 'Team', 'display_position','percent_owned','L'+str(n_1)+' PP TOI','L'+str(n_2)+' PP TOI','L'+str(n_3)+' PP TOI',
                                 'L'+str(n_1)+' PP%','L'+str(n_2)+' PP%','L'+str(n_3)+' PP%']

        list_of_columns_name =  ['Player', 'Team', 'Position','Roster%','L'+str(n_1)+' PP TOI','L'+str(n_2)+' PP TOI','L'+str(n_3)+' PP TOI',
                                 'L'+str(n_1)+' PP%','L'+str(n_2)+' PP%','L'+str(n_3)+' PP%']

        if type(n_1) is not int:
            n_1 = 1
            
        if (n_2 == 0) or (n_2 == n_1) or (n_2 == None):
            list_of_columns.remove(f'L{str(n_2)} PP TOI')
            list_of_columns.remove(f'L{str(n_2)} PP%')
            list_of_columns_name.remove(f'L{str(n_2)} PP TOI')
            list_of_columns_name.remove(f'L{str(n_2)} PP%')

        if (n_3 == 0) or (n_3 == n_1) or (n_3 == n_2) or (n_3 == None):
            list_of_columns.remove(f'L{str(n_3)} PP TOI')
            list_of_columns.remove(f'L{str(n_3)} PP%')
            list_of_columns_name.remove(f'L{str(n_3)} PP TOI')
            list_of_columns_name.remove(f'L{str(n_3)} PP%')

            
        
        start_date ='2023-09-01'
        end_date = '2024-05-01'
        player_games_df = player_games_df_2.copy()
        player_games_df = player_games_df.merge(right=team_games_df,left_on=['Team','date'],right_on=['abv','date'])

        df_pp_1 = player_games_df.groupby('player_id').head(n_1)
        df_pp_1 = df_pp_1[df_pp_1.game_count > df_pp_1.max_games-n_1]

        df_pp_2 = df_pp_2 = player_games_df.groupby('player_id').head(n_2)
        df_pp_2[df_pp_2.game_count > df_pp_2.max_games-n_2]

        df_pp_3 = df_pp_3 = player_games_df.groupby('player_id').head(n_3)
        df_pp_3[df_pp_3.game_count > df_pp_3.max_games-n_3]

        team_games_df_1 = team_games_df.groupby('team').head(n_1)
        team_games_df_2 = team_games_df.groupby('team').head(n_2)
        team_games_df_3 = team_games_df.groupby('team').head(n_3)

        df_all_pp_1 = df_pp_1.copy()
        df_all_pp_2 = df_pp_2.copy()
        df_all_pp_3 = df_pp_3.copy()

        df_all_pp_1_final = df_all_pp_1.groupby(['player_id','Player','Team','Position']).sum()[['TOI_pp']].reset_index()
        df_all_pp_2_final = df_all_pp_2.groupby(['player_id','Player','Team','Position']).sum()[['TOI_pp']].reset_index()
        df_all_pp_3_final = df_all_pp_3.groupby(['player_id','Player','Team','Position']).sum()[['TOI_pp']].reset_index()

        team_games_df_1_final = team_games_df_1.groupby(['abv']).sum()[['pp_toi']].reset_index()
        team_games_df_2_final = team_games_df_2.groupby(['abv']).sum()[['pp_toi']].reset_index()
        team_games_df_3_final = team_games_df_3.groupby(['abv']).sum()[['pp_toi']].reset_index()

        df_final =  df_all_pp_1_final.merge( df_all_pp_2_final,how='outer',left_on=['player_id'],right_on=['player_id'],suffixes=("","_2"))
        df_final =  df_final.merge( df_all_pp_3_final,how='outer',left_on=['player_id'],right_on=['player_id'],suffixes=("_1","_3"))

        team_final =  team_games_df_1_final.merge( team_games_df_2_final,how='outer',left_on=['abv'],right_on=['abv'],suffixes=("","_2"))
        team_final =  team_final.merge( team_games_df_3_final,how='outer',left_on=['abv'],right_on=['abv'],suffixes=("_1","_3"))


        df_final = df_final.merge(team_final,left_on='Team_1',right_on='abv')

        test = df_final[['player_id','Player_1','Team_1','Position_1','TOI_pp_1','TOI_pp_2','TOI_pp_3','pp_toi_1','pp_toi_2','pp_toi_3']]
        test.columns = ['player_id','Player','Team','Position','TOI_1','TOI_2','TOI_3','pp_toi_1','pp_toi_2','pp_toi_3']
        test = test.fillna('0')


        test['PP%_1'] = test['TOI_1'].astype(float)/ test['pp_toi_1'].astype(float)
        test['PP%_2'] = test['TOI_2'].astype(float)/ test['pp_toi_2'].astype(float)
        test['PP%_3'] = test['TOI_3'].astype(float)/ test['pp_toi_3'].astype(float)
        # test = test.fillna(0)
        test['TOI_1'] = ["%d:%02d" % (int(x),(x*60)%60) for x in test['TOI_1'].astype(float)]
        test['TOI_2'] = ["%d:%02d" % (int(x),(x*60)%60) for x in test['TOI_2'].astype(float)]
        test['TOI_3'] = ["%d:%02d" % (int(x),(x*60)%60) for x in test['TOI_3'].astype(float)]
        test = test.drop(['pp_toi_1','pp_toi_2','pp_toi_3'],axis=1)
        test.columns = ['player_id','Player','Team','Position','L'+str(n_1)+' PP TOI','L'+str(n_2)+' PP TOI','L'+str(n_3)+' PP TOI','L'+str(n_1)+' PP%','L'+str(n_2)+' PP%','L'+str(n_3)+' PP%']


        yahoo_df = yahoo_df_2.merge(yahoo_nhl_df,left_on = 'player_id',right_on='player_id_yahoo',suffixes=['','_y'])
        yahoo_df.nhl_id = yahoo_df.nhl_id.astype(float)
        test.player_id = test.player_id.astype(float)


        test = test.merge(right=yahoo_df,left_on='player_id',right_on='nhl_id',suffixes=['','_y'],how='left')
        # if len(test[test.player_id == 8478427]) > 0:
        #     return test.style
            
            #print(test[test.nhl_id == 8478427])

        test.loc[test.display_position.isna(),'display_position'] = test.loc[test.display_position.isna(),'Position']
        test.display_position = test.display_position.replace({'L':'LW','R':'RW'})
        test.percent_owned = test.percent_owned.fillna(0)

        


        print('Column List')
        print(test.columns)


        print(list_of_columns)
        test = test[list_of_columns]
        test = test.rename(columns={'percent_owned':'Roster%'})
        test = test.rename(columns={'display_position':'Position'})
        

        top_d_score = test.copy()[(test.Team==input.team_id())].sort_values(by=['L'+str(n_1)+' PP%'],ascending=False).reset_index(drop=True)
        print(top_d_score)
        if input.x():
            top_d_score = top_d_score.dropna(axis='columns')
        
        top_d_score = top_d_score.head(min(len(top_d_score),top_n))


        #top_d_score.columns = list_of_columns_name

        

        top_d_score['Deployment'] = "PP2"
        top_d_score['Deployment'][0:5] = "PP1"

        
        cols = top_d_score.columns.tolist();
        print('we made it here',cols)

        # for i in list(input.ignore_id()):
        #     print('we made it here')
        #     print(i)
        #     cols.remove(i)

        # df_style_bang = top_d_score.head(10).style.background_gradient(cmap=co, subset=['L'+str(n_1)+' PP%','L'+str(n_2)+' PP%','L'+str(n_3)+' PP%','Roster%'],vmin=0,vmax=1).hide_index().set_properties(**{'Height': '12px'},**{'text-align': 'center'}).set_table_styles([{
        #         'selector': 'caption',
        #         'props': [
        #             ('color', ''),
        #             ('fontname', 'Century Gothic'),
        #             ('font-size', '20px'),
        #             ('font-style', 'italic'),
        #             ('font-weight', ''),
        #             ('text-align', 'centre'),
        #             ]

        #             },{'selector' :'th', 'props':[('text-align', 'center'),('Height','5px')]},{'selector' :'td', 'props':[('text-align', 'center'),('font-size', '13px'),('fontname', 'Century Gothic')]}]).format(
        #                 {'L'+str(n_1)+' PP%': '{:.0%}',
        #                 'L'+str(n_2)+' PP%': '{:.0%}',
        #                 'L'+str(n_3)+' PP%': '{:.0%}',
        #                 'Roster%': '{:.0%}',
        #                 },)

        df_style_bang = top_d_score[cols].head(input.top_n()).style.set_properties(**{'border': '3 px'},overwrite=False).set_table_styles([{
            'selector': 'caption',
            'props': [
                ('color', ''),
                ('fontname', 'Century Gothic'),
                ('font-size', '20px'),
                ('font-style', 'italic'),
                ('font-weight', ''),
                ('text-align', 'centre'),
                ]

                },{'selector' :'th', 'props':[('text-align', 'center'),('Height','px'),('color','black'),(
                    'border', '1px black solid !important')]},{'selector' :'td', 'props':[('text-align', 'center'),('font-size', '18px'),('color','black')]}],overwrite=False).set_properties(
                    **{'background-color':'White','index':'White','min-width':'100px'},overwrite=False).set_properties(
                    **{'background-color':'White','index':'White','min-width':'200px'},overwrite=False,subset=cols[0]).set_table_styles(
            [{'selector': 'th:first-child', 'props': [('background-color', 'white')]}],overwrite=False).set_table_styles(
            [{'selector': 'tr:first-child', 'props': [('background-color', 'white')]}],overwrite=False).set_table_styles(
            [{'selector': 'tr', 'props': [('line-height', '35px')]}],overwrite=False).set_properties(
            **{'Height': '35px'},**{'text-align': 'center'},overwrite=False).set_properties(**{'border': '3 px','color':'black'},overwrite=False).set_properties(**{'border': '3 px','color':'black'},overwrite=False).set_properties(
                    **{'border': '1px black solid !important'},subset = ((list(top_d_score.index[:]),top_d_score.columns[:]))).set_properties(**{
                                'color': 'black'},overwrite=False).set_properties(
                    **{'border': '1px black solid !important'},subset = ((list(top_d_score.index[:]),top_d_score.columns[:]))).format(
            {
            'L'+str(n_1)+' PP%': '{:.0%}',
            'L'+str(n_2)+' PP%': '{:.0%}',
            'L'+str(n_3)+' PP%': '{:.0%}',
            'Roster%': '{:.0%}',
            },).background_gradient(cmap=co, subset=[x for x in cols if x.endswith('PP%')]).hide_index().background_gradient(cmap=co,vmin=0.0,vmax=0.7, subset=[x for x in cols if x.endswith('PP%')])

        return df_style_bang



# test = test.fillna(0)
#test['PP TOI'] = ["%d:%02d" % (int(x),(x*60)%60) if x>0 else '0:00' for x in test['PP TOI']]



app = App(app_ui, server)
#time.sleep(60)