fl | Lightweight Bayesian Filtering Library with real-time | Recommender System library

from logging import StreamHandler, exception
from sqlalchemy import create_engine
from re import search
from pandas._config.config import options
from pandas.core.frame import DataFrame
from pandas.core.tools import numeric
import streamlit as st       
import plotly.express as pt
import pandas as pd
import numpy as np
import os
import base64

#function
def filedownloader(database):
    csv = database.to_csv(index=False)
    b64 = base64.b64encode(csv.encode()).decode() 
    href = f'Download csv file'
    st.markdown(href,unsafe_allow_html=True)
    return href




#basic titles

st.title("Institution's Section")
st.sidebar.subheader("Settings")

# file uploader

fl=st.sidebar.file_uploader(label="Uplaod File in csv or xlsx format", type=['csv','xlsx'])
global df
if fl is not None:
    st.title("Performance Graph")
    try:
        df=pd.read_csv(fl)
    except Exception as e:
        print(e)
        df=pd.read_excel(fl)

#write table in web page

global columns
try:
    st.write(df)
    columns = list(df.columns)
except Exception as e:
    print(e)
    st.write("Please upload a file")

#Chart select

st.sidebar.subheader("Analysis section")
ch=st.sidebar.selectbox(
    label="Select the chart type",
    options=['Scatterplots','Lineplots','Histogram','Boxplot'],
    key="chart"
)

#Plot Settings

if ch =='Scatterplots':
    st.sidebar.subheader("Scatterplot Settings")
    
    try:
        x_values=st.sidebar.selectbox('X axis',options=columns)
        y_values=st.sidebar.selectbox('Y axis',options=columns)
        plot=pt.scatter(data_frame=df,x=x_values,y=y_values)
        st.plotly_chart(plot)
    except Exception as e:
        print(e)
elif ch=='Lineplots':
    st.sidebar.subheader("Lineplot Settings")
    
    try:
        x_values=st.sidebar.selectbox('X axis',options=columns)
        y_values=st.sidebar.selectbox('Y axis',options=columns)
        plot=pt.line(data_frame=df,x=x_values,y=y_values)
        st.plotly_chart(plot)
    except Exception as e:
        print(e)
elif ch=='Histogram':
    st.sidebar.subheader("Histogram Settings")
    
    try:
        x_values=st.sidebar.selectbox('X axis',options=columns)
        y_values=st.sidebar.selectbox('Y axis',options=columns)
        plot=pt.histogram(data_frame=df,x=x_values,y=y_values)
        st.plotly_chart(plot)
    except Exception as e:
        print(e)
elif ch=='Boxplot':
    st.sidebar.subheader("Boxplot Settings")
    
    try:
        x_values=st.sidebar.selectbox('X axis',options=columns)
        y_values=st.sidebar.selectbox('Y axis',options=columns)
        plot=pt.box(data_frame=df,x=x_values,y=y_values)
        st.plotly_chart(plot)
    except Exception as e:
        print(e)

#rank select


st.sidebar.subheader("Quick Search Rank")
radio=st.sidebar.radio(
    "What you want to see?",("Top five","Bottom five")
)

#rank settings

numeric_column = df.select_dtypes(include=np.number).columns.tolist()
if(radio=='Top five'):
    st.sidebar.subheader("Rank Settings")
    try:
        val=st.sidebar.selectbox('Select rank category',options=numeric_column)
        st.title("Quick Search Result")
        st.write(df.nlargest(5,val))
    except Exception as e:
        print(e)
elif(radio=='Bottom five'):
    st.sidebar.subheader("Rank Settings")
    try:
        val=st.sidebar.selectbox('Select rank category',options=numeric_column)
        st.title("Quick Search Result")
        st.write(df.nsmallest(5,val))
    except Exception as e:
        print(e)

#search

st.sidebar.subheader("Search")
val=st.sidebar.selectbox('Search desired column',options=columns)
st.sidebar.subheader("Search keyword")
user_input = st.sidebar.text_area("Type Keyword here", 0)
try: 
    dl = df[df[val] == type(df[val][1])(user_input)]
except Exception as e:
    pass

st.sidebar.subheader("Custom plot settings")
choose=st.sidebar.selectbox(
    label="Choose plot for selected data",
    options=['Scatterplot','Linearplot','Histogram','Boxplot',]
)
if(choose == 'Scatterplot'):
    try:
        st.sidebar.subheader("Scatterplot Settings")
        x_val=st.sidebar.selectbox('X axis',options=columns,key="scatter_x")
        y_val=st.sidebar.selectbox('Y axis',options=columns,key="scatter_y")
        plots=pt.scatter(data_frame=dl,x=x_val,y=y_val)
    except Exception as e:
        pass
elif (choose == 'Linearplot'):
    try:
        st.sidebar.subheader("Scatterplot Settings")
        x_val=st.sidebar.selectbox('X axis',options=columns,key="scatter_x")
        y_val=st.sidebar.selectbox('Y axis',options=columns,key="scatter_y")
        plots=pt.line(data_frame=dl,x=x_val,y=y_val)
    except Exception as e:
        pass
elif (choose == 'Histogram'):
    try:
        st.sidebar.subheader("Scatterplot Settings")
        x_val=st.sidebar.selectbox('X axis',options=columns,key="scatter_x")
        y_val=st.sidebar.selectbox('Y axis',options=columns,key="scatter_y")
        plots=pt.histogram(data_frame=dl,x=x_val,y=y_val)
    except Exception as e:
        pass
elif (choose == 'Boxplot'):
    try:
        st.sidebar.subheader("Scatterplot Settings")
        x_val=st.sidebar.selectbox('X axis',options=columns,key="scatter_x")
        y_val=st.sidebar.selectbox('Y axis',options=columns,key="scatter_y")
        plots=pt.box(data_frame=dl,x=x_val,y=y_val)
    except Exception as e:
        pass

#chart and search button

bt=st.sidebar.button("Search and Apply")
if(bt):
    st.title("Search Results")
    st.write(dl)
    st.title("Searh Analysis")
    st.plotly_chart(plots)

#Modify database
st.sidebar.subheader("Select Column to drop null values")
val1=st.sidebar.multiselect(
    label="Selet Columns",
    options=columns
)
bt_null_r=st.sidebar.button("Drop Null values")
if(bt_null_r):
    mdf=df.dropna(subset=val1, how='all')
    filedownloader(mdf)

# drop selected column
st.sidebar.subheader("Select Column to drop null values")
val2=st.sidebar.multiselect(
    label="Selet Columns",
    options=columns,key="column_multi"
)
bt_c=st.sidebar.button("Drop Columns")
if(bt_c):
    mdf=df.drop(val2,axis=1)
    filedownloader(mdf)
**#creating sql engine
engine=create_engine('mysql://root:Soumik18@@localhost:3306/college')
df.to_sql('Students',con=engine)**