Files
auto24-scraper/auto24.py
2026-02-01 14:51:55 +02:00

314 lines
9.1 KiB
Python

# coding: utf-8
# In[372]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import re
import json
from urllib.request import urlopen, Request
from bs4 import BeautifulSoup
# **Scrape results list excluding detailviews**
# In[373]:
def get_car_features(element, mylist):
"""Returns the car's features as nested list"""
car_instance = element.findChild('div',attrs={"class":"description"})
#Find ad's id
ad_link = element.findChild('a',attrs={"class":"row-link"})
id_value = str(ad_link.attrs['href']).split('/')[2].strip("}").strip("'")
mylist[0].append(id_value)
#Find car description
mylist[1].append(car_instance.findChild('a',attrs={"class":"main"}).get_text())
#Production year
year_td = car_instance.findChild('span',attrs={'class':'year'})
mylist[2].append(year_td.get_text())
#Fuel
mylist[3].append(car_instance.findChild('span',attrs={'class':'fuel'}).get_text())
#Transmission
mylist[4].append(car_instance.findChild('span',attrs={'class':'transmission'}).get_text())
#Price
price = car_instance.findChild('span',attrs={'class':'price'})
print(price)
mylist[5].append(price.get_text() if price else '')
# In[374]:
def scrape_page(url):
html = urlopen(url)
soup = BeautifulSoup(html, 'lxml')
mylist=[[], # id
[], # car description
[], # production year
[], # fuel
[], # transmission
[], # Price
]
mytable = soup.find('div',attrs={'id':'usedVehiclesSearchResult-flex'})
result_rows = mytable.findChildren('div',attrs={'class':re.compile(fr"result-row item-.")})
for row in result_rows:
get_car_features(row,mylist)
df = pd.DataFrame(mylist)
transposed = df.transpose()
return transposed
# In[375]:
def click_next(driver):
try:
myelement = WebDriverWait(driver, 10).until(
EC.presence_of_element_located((By.CLASS_NAME, "next-page"))
)
myelement.click()
except:
print('Did not find the element.')
# In[376]:
import time
timeout = time.time() + 60*1
dataframes=[]
#url = "https://www.auto24.ee/kasutatud/nimekiri.php?bn=2&a=100&aj=&ssid=46274853&b=4&bw=21&ae=8&af=200&otsi=otsi"
url = "https://www.auto24.ee/kasutatud/nimekiri.php?bn=2&a=101&aj=&f1=2018&g2=12000&k1=60&ae=2&af=50&ag=0&ag=1&otsi=otsi"
# driver = webdriver.Firefox()
# driver.get(url)
hdr = {'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.80 Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9',
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.3',
'Accept-Encoding': 'none',
'Accept-Language': 'en-US,en;q=0.9,et;q=0.8',
'Connection': 'keep-alive',
# 'Cookie': 'CID=1621884104633711; my_searches_notif=1; PHPSESSID=sc66u21lot8te95edda73985tu; OptanonAlertBoxClosed=2022-01-26T10:18:23.206Z; __utma=13167336.877242740.1643192318.1643192318.1643192318.1; __utmc=13167336; __utmz=13167336.1643192318.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); __cf_bm=QLTYnfz3iFPTGXscwrRdyJrkacwt8enoFU4EqQ4w33E-1644415671-0-AVspsg+4OGWPbPvqTxKtvECVzzHBpAluh0vUtjvbrznme9mZ7QRZP5mO28qeaeZ3rvWRlU8ES4oxAzZ+Z8Wy1S4=; OptanonConsent=isIABGlobal=false&datestamp=Wed+Feb+09+2022+16%3A11%3A38+GMT%2B0200+(Eastern+European+Standard+Time)&version=6.16.0&hosts=&landingPath=NotLandingPage&groups=C0004%3A1%2CC0003%3A1%2CC0002%3A1%2CC0001%3A1&AwaitingReconsent=false&geolocation=%3B',
# 'Cookie': 'CID=1621884104633711; my_searches_notif=1; PHPSESSID=sc66u21lot8te95edda73985tu; OptanonAlertBoxClosed=2022-01-26T10:18:23.206Z; __utma=13167336.877242740.1643192318.1643192318.1643192318.1; __utmc=13167336; __utmz=13167336.1643192318.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); OptanonConsent=isIABGlobal=false&datestamp=Wed+Feb+09+2022+16%3A19%3A48+GMT%2B0200+(Eastern+European+Standard+Time)&version=6.16.0&hosts=&landingPath=NotLandingPage&groups=C0004%3A1%2CC0003%3A1%2CC0002%3A1%2CC0001%3A1&AwaitingReconsent=false&geolocation=%3B; __cf_bm=cp_7820d1SeeDxv37StQH7HMg0KZgMVEHbcD_JFYmQ0-1644416588-0-AWYK6qWmiAJwRtqvbtwuBgDJU2FauySHu6Mn9R6vYiUIrvWkcgs1khFFJcpvWmFP9o9m1LkwHNRsRSvgFPTo/bY=',
'Cookie': 'CID=1621884104633711; my_searches_notif=1; PHPSESSID=sc66u21lot8te95edda73985tu; OptanonAlertBoxClosed=2022-01-26T10:18:23.206Z; __utma=13167336.877242740.1643192318.1643192318.1643192318.1; __utmc=13167336; __utmz=13167336.1643192318.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); __cf_bm=cp_7820d1SeeDxv37StQH7HMg0KZgMVEHbcD_JFYmQ0-1644416588-0-AWYK6qWmiAJwRtqvbtwuBgDJU2FauySHu6Mn9R6vYiUIrvWkcgs1khFFJcpvWmFP9o9m1LkwHNRsRSvgFPTo/bY=; OptanonConsent=isIABGlobal=false&datestamp=Wed+Feb+09+2022+16%3A27%3A50+GMT%2B0200+(Eastern+European+Standard+Time)&version=6.16.0&hosts=&landingPath=NotLandingPage&groups=C0004%3A1%2CC0003%3A1%2CC0002%3A1%2CC0001%3A1&AwaitingReconsent=false&geolocation=%3B',
'Referer': 'https://www.auto24.ee/kasutatud/nimekiri.php?bn=2&a=100&b=4&bw=21&ae=8&af=50&ssid=46274853&ak=50'}
autoreq = Request(url, headers=hdr)
#nextlink = soup.find('a',attrs={'class':"input-link item","rel":"next"})
#url = "https://www.auto24.ee/" + nextlink.attrs.get('href')
has_nextpage=True
while(has_nextpage):
dataframes.append(scrape_page(autoreq))
html = urlopen(autoreq)
soup = BeautifulSoup(html, 'lxml')
nextlink = soup.find('a',attrs={'class':"input-link item","rel":"next"})
has_nextpage = bool(nextlink)
if has_nextpage:
url = "https://www.auto24.ee/" + nextlink.attrs.get('href')
# while True:
# dataframes.append(scrape_page(soup)
# if bool(not soup.find('div',attrs={'class':'next-page'})) or time.time() > timeout:
# break
# click_next(driver)
# In[377]:
df_all = pd.concat(dataframes,ignore_index=True)
df_all.columns = ['id', 'description', 'year','fuel','gearbox','price']
df_all['id'] = pd.to_numeric(df_all['id'])
df_all.dtypes
# In[378]:
type(df_all.iloc[0,0])
# In[379]:
def scrape_detailview(id_string, list_in):
url = 'https://www.auto24.ee/used/' + id_string
html = urlopen(Request(url, headers=hdr))
soup = BeautifulSoup(html, 'lxml')
list_in[0].append(str(id_string))
body_type = np.nan
manuf_date = np.nan
km = np.nan
main_data = soup.find('table',attrs={"class":"section main-data"})
if main_data:
tr_body = main_data.findChild('tr',attrs={'class':'field-keretyyp'})
tr_date = main_data.findChild('tr',attrs={'class':'field-month_and_year'})
tr_km = main_data.findChild('tr',attrs={'class':'field-labisoit'})
body_type = tr_body.findChild('span',attrs={'class':'value'}).get_text() if tr_body else body_type
manuf_date = tr_date.findChild('span',attrs={'class':'value'}).get_text() if tr_date else manuf_date
km = tr_km.findChild('span',attrs={'class':'value'}).get_text() if tr_km else km
list_in[1].append(body_type)
list_in[2].append(manuf_date)
list_in[3].append(km)
# In[380]:
adslist = [[], #0 id
[], #1 body_type
[], #2 manufacturing date
[] #3 kilometers driven
]
for adid in list(df_all['id']):
scrape_detailview(str(adid), adslist)
# In[381]:
pd.options.display.max_rows = 999
df_ads = pd.DataFrame(adslist).transpose()
df_ads.columns = ['id','body','man_date','km']
df_ads['id'] = pd.to_numeric(df_ads['id'])
df_ads.head()
# In[382]:
df_all.head()
# df_all.drop_duplicates(subset=['id'],inplace=True)
# df_ads.drop_duplicates(subset=['id'],inplace=True)
# In[383]:
df = df_ads.merge(df_all,how='left',on=['id'])
# In[384]:
df.head()
# In[385]:
df['model'] = df['description'].apply(lambda x: x.split(' ')[1])
df['manufacturer'] = df['description'].apply(lambda x: x.split()[0])
df.head(3)
# In[390]:
mypattern = re.compile('(^\d{2,3} ?\d*)')
# In[391]:
matchobject = mypattern.findall('19 000 sis. KM')[0]
entries = re.search('(^\d{2,3} ?\d{3})','11 900 sis. KM')
#entries.group(0)
(matchobject)
# In[392]:
import unicodedata
#new_str = unicodedata.normalize("NFKD", unicode_str)
df['km'].fillna("0", inplace=True)
df['km']= df['km'].apply(lambda x: unicodedata.normalize("NFKD",str(x)))
df['km']= df['km'].apply(lambda x: str(x).strip(' km').replace(' ',''))
df['km'] = pd.to_numeric(df['km'])
# In[401]:
mypattern = re.compile('(^\d{2,3} ?\d*)')
df['price']= df['price'].apply(lambda x: unicodedata.normalize("NFKD",str(x)))
# df['price'] = df['price'].apply(lambda x: mypattern.findall(x)[0].replace(" ",""))
#df['price']=pd.to_numeric(df['price'])
#df['price'].unique()
# In[ ]:
df['year'] = pd.to_numeric(df['year'])
df.dtypes
# In[ ]:
df.sort_values('description', inplace=True)
df.head()
# In[ ]:
df[df['manufacturer']=='Volkswagen']
# In[ ]:
df[df['year']==2011].groupby(['manufacturer','model']).count()
# In[ ]:
df[(df['year']==2009)
& (df['manufacturer']=='Ford')]['km'].plot.hist(bins=20)
# In[ ]:
df[['manufacturer','model']].value_counts()
filename = 'file.json'
with open(filename, 'w') as file:
json.dump(df.to_json(), file)