import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
data = pd.read_csv("referendum-categories.csv")
data['respondent'] = data['respondent'].str.strip()
data['comment'] = data['comment'].str.strip()
cv = data['comment'].value_counts() # a list of each category and respective counts
cat_and_value = pd.DataFrame() # a dataframe for categories of count > 1
ones =  pd.DataFrame() # a dataframe for categories of count = 1
for ind, row in cv.items(): # iterating through comment and count
    if row == 1: # sorting categories where count = 1
        new_row = pd.DataFrame({'comment': [ind], 'count' : [row]})
        ones = pd.concat([ones, new_row],ignore_index = True) 
    else: # all other categories
        new_row = pd.DataFrame({'comment': [ind], 'count' : [row]})
        cat_and_value = pd.concat([cat_and_value, new_row],ignore_index = True) 

ones.to_csv("ones.csv")
cat_and_value.to_csv("comments.csv")

print(len(ones))

# A plot for category counts where the count > 1
plt.plot(cat_and_value['count'], marker='o')
plt.title("Ranked most popular category count")
plt.ylabel("Category Count")
plt.show()