Pythonで日本のあまりにも酷過ぎる超悲惨な年齢別人口比を考察

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import re
import urllib.request, urllib.parse, urllib.error
from bs4 import BeautifulSoup

ctx = ssl.create_default_context()
ctx.check_hostname = False
ctx.verify_mode = ssl.CERT_NONE

# Read the HTML from the URL and pass on to BeautifulSoup
url = 'https://www.cia.gov/library/publications/the-world-factbook/'
print("Opening the file connection...")
uh= urllib.request.urlopen(url, context=ctx)
print("HTTP status",uh.getcode())
html =uh.read().decode()
print(f"Reading done. Total {len(html)} characters read.")

soup = BeautifulSoup(html, 'html.parser')

country_codes=[]
country_names=[]
for tag in soup.find_all('option'):
    country_codes.append(tag.get('value')[8:10])
    country_names.append(tag.text)

temp=country_codes.pop(0) # To remove the first entry 'World'
temp=country_names.pop(0) # To remove the first entry 'World'

print('COUNTRY NAMES\n'+'-'*30)
for country in country_names[1:]:
    print(country,end=',')
print('\n\nCOUNTRY CODES\n'+'-'*30)
for country in country_codes[1:]:
    print(country,end=',')

# Base URL
urlbase = 'https://www.cia.gov/library/publications/the-world-factbook/geos/'
demographics1=[]
demographics2=[]
demographics3=[]
demographics4=[]
demographics5=[]

offset = len('65 years and over: ')

# Iterate over every country
for i in range(1,len(country_names)-1):
    country_html=country_codes[i]+'.html'
    url_to_get=urlbase+country_html
    # Read the HTML from the URL and pass on to BeautifulSoup
    html = urllib.request.urlopen(url_to_get, context=ctx).read()
    soup = BeautifulSoup(html, 'html.parser')
    
    txt=soup.get_text()
    pos1=txt.find('0-14 years:')
    pos2=txt.find('15-24 years:')
    pos3=txt.find('25-54 years:')
    pos4=txt.find('55-64 years:')
    pos5=txt.find('65 years and over:')
    
    if pos1==-1:
        print(f"**0-14 years % data not found for {country_names[i]}!**")
        demographics1.append(np.nan)
    else:
        text=txt[pos1+12:pos1+18]
        end=re.search('%',text).start()
        a=float((txt[pos1+12:pos1+12+end]))
        demographics1.append(a)
        print(f"0-14 years % data extraction complete for {country_names[i]}!")
    
    if pos2==-1:
        print(f"**15-24 years % data not found for {country_names[i]}!**")
        demographics2.append(np.nan)
    else:
        text=txt[pos2+13:pos2+19]
        end=re.search('%',text).start()
        a=float((txt[pos2+13:pos2+13+end]))
        demographics2.append(a)
        print(f"15-24 years % data extraction complete for {country_names[i]}!")
        
    if pos3==-1:
        print(f"**25-54 years % data not found for {country_names[i]}!**")
        demographics3.append(np.nan)
    else:
        text=txt[pos3+13:pos3+19]
        end=re.search('%',text).start()
        a=float((txt[pos3+13:pos3+13+end]))
        demographics3.append(a)
        print(f"25-54 years % data extraction complete for {country_names[i]}!")
    
    if pos4==-1:
        print(f"**55-64 years % data not found for {country_names[i]}!**")
        demographics4.append(np.nan)
    else:
        text=txt[pos4+13:pos4+19]
        end=re.search('%',text).start()
        a=float((txt[pos4+13:pos4+13+end]))
        demographics4.append(a)
        print(f"55-64 years % data extraction complete for {country_names[i]}!")
        
    if pos5==-1:
        print(f"**65 years and beyond % data not found for {country_names[i]}!**")
        demographics5.append(np.nan)
    else:
        text=txt[pos5+offset:pos5+offset+6]
        end=re.search('%',text).start()
        a=float((txt[pos5+offset:pos5+offset+end]))
        demographics5.append(a)
        print(f"65 years and beyond % data extraction complete for {country_names[i]}!")

data={
    '0-14 years old %':demographics1,
    '15-24 years %':demographics2,
    '25-54 years %':demographics3,
    '55-64 years %':demographics4,
    '65 years and above %':demographics5
}
df=pd.DataFrame(data=data,index=\
 country_names[1:len(country_codes)-1])
df.index.name='Country'
index=df.index.str.strip()
df=pd.DataFrame(data=data,index=index)
df.dropna(inplace=True)
df.head()

df['Japan':].head(1)

df1=df[df['65 years and above %']>20]
df1.sort_values(by=\
 ['65 years and above %'],ascending=False)

df2=df1.iloc[:,4]
df3=df2.sort_values(ascending=False).plot(kind='bar',legend=None,title="old population by Country")

df2=df1.iloc[:,4]
df2=df2.drop(['Monaco', 'Saint Pierre and Miquelon','Isle of Man','Puerto Rico'])
df3=df2.sort_values(ascending=False).plot(kind='bar',legend=None,title="old population by Country")

df3=df2.sort_values(ascending=True).plot(kind='barh',legend=None,title="old population by Country")
df3.set_xlabel('65 years and above %')

df1=df1.drop(['Monaco', 'Saint Pierre and Miquelon','Isle of Man','Puerto Rico'])

df4=df1.iloc[:,0]+df1.iloc[:,1]

df4=df4.sort_values(ascending=True).plot(kind='barh',legend=None,title="young population by Country")
df4.set_xlabel('24 years and below %')