import pandas as pd
from tabulate import tabulate
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

# Standard plotly imports
import chart_studio.plotly as py
import plotly.graph_objs as go
from plotly.offline import iplot, init_notebook_mode
# Using plotly + cufflinks in offline mode
import cufflinks
data=pd.read_csv(r"Data/combineFinal.csv",encoding="unicode_escape")
def processTypeOfSofa(answer):
    if answer==1:
        return(data[data.loc[:,"TYPE"]=="Sectional sofa"])
    elif answer==2:
        return(data[data.loc[:,"TYPE"]=="Sleeper sofa"])
    elif answer==3:
        return(data[data.loc[:,"TYPE"]=="Loveseat"])
    elif answer==4:
        return(data[data.loc[:,"TYPE"]=="Chaise"])
    elif answer==5:
        return(data[data.loc[:,"TYPE"]=="Sofa Bed"])
    elif answer==6:
        return(data[data.loc[:,"TYPE"]=="Sleeper sofa"])
    elif answer==7:
        return(data[data.loc[:,"TYPE"]=="Recliner"])
    elif answer==8:
        return(data[data.loc[:,"TYPE"]=="Sofa"])
    else:
        return("Invalid input")
def processSource(choiceOfSource):
    if(choiceOfSource==1):
        return(data[data.loc[:,"Website"]=="Craigslist"])
    elif(choiceOfSource==2):
        return(data[data.loc[:,"Website"]=="Ikea"])
    elif(choiceOfSource==3):
        return(data[data.loc[:,"Website"]=="Ebay"])

def giveStatistics(dataFrame):
    #df = pd.dataframe(dataFrame)

    print("")
    print('The lowest price available is: ' + str((dataFrame.Price.min())))
    print('The highest price available is: ' + str((dataFrame.Price.max())))
    print('The average price available is: ' + str((dataFrame.Price.mean())))

def dataframe_Table(dataframe):

    #df1 = dataframe['Name', 'Price', 'TYPE','Website']
    print('trial')
    cols=list(dataframe.columns)
    cols.remove("Product URL")
    cols.remove("REVIEW")
    pdtabulate = lambda dataframe:tabulate(dataframe[cols], headers = 'keys', tablefmt = 'psql',showindex=False)
    print(pdtabulate(dataframe))

def statistics_graph(dataframe):

    cufflinks.go_offline(connected=True)
    init_notebook_mode(connected=True)
    # Read data from file 'filename.csv'
    # (in the same directory that your python process is based)
    # Control delimiters, rows, column names with read_csv (see later)

    print(data)
    print(data.TYPE)

    type(data.Price[0])
    data['Price'] = data['Price'].astype(str)
    data['Price'] = data.Price.apply(lambda x: x.strip('$'))  # removing $ sign
    data['Price'] = data.Price.apply(lambda x: x.replace(",", ""))  # removing , sign
    print(data.Price)

    data['TYPE'] = data['TYPE'].astype(str)
    # data['TYPE'] = [x[7:] for x in data.TYPE]

    data['Price'] = data['Price'].astype(float, errors='ignore')
    data['Price'] = pd.to_numeric(data['Price'], errors='coerce')



    ax = sns.countplot(x="Website", data=data)

    # plot data
    fig, ax = plt.subplots(figsize=(15, 7))
    # use unstack()
    data.groupby(['Website', 'TYPE']).count()['Price'].unstack().plot(ax=ax)



    sns.set(
        rc={'figure.figsize': (18, 6)},
        style="white"
    )
    sns.violinplot(
        x='Website',
        y='Price',
        hue='TYPE',
        data=data
    )
    sns.despine()
    import plotly.express as px
    fig = px.scatter(
        data_frame=data,
        x="TYPE",
        y="Price",
        animation_group="TYPE",
        size="Price",
        color="Website",
        hover_name="TYPE",
        facet_col="Website",
        size_max=45
    )
    fig.show()

    filtered_class = data[data['Website'] != 'AVANTGARDE']
    chart = sns.lmplot("TYPE", "Price", data=filtered_class, hue="TYPE", fit_reg=False, col='Website', col_wrap=2)
    for ax in chart.axes.flat:
        for label in ax.get_yticklabels():
            label.set_rotation(0)

    data.sort_values(by=['Price'], inplace=True)
    g = sns.FacetGrid(data, col="Website", col_wrap=3, height=4)
    g = (g.map(plt.hist, "Price", bins=np.arange(0, 2300, 500)))

    g1 = sns.FacetGrid(data, col='TYPE', col_wrap=3, height=4)
    g1 = (g1.map(plt.hist, "Price", bins=np.arange(0, 2300,
                                                   500)))  # From the graph, it seems like companies chose not to categorize sofas either explicitly or in the name. It may be to increase visibility of certain sofas which can be suggested to users irrespective of the filter they applied.

    data_pivot = data.pivot(columns='Website', values='TYPE').iplot(
        kind='box',
        yTitle='Website',
        title='Type of sofas by website')