Дз парсинг

This commit is contained in:
inweems
2023-03-30 23:06:11 +03:00
parent 709fc0d694
commit 56d6a99a71
2 changed files with 48 additions and 7 deletions
+7 -7
View File
@@ -4,7 +4,7 @@
<option name="autoReloadType" value="SELECTIVE" />
</component>
<component name="ChangeListManager">
<list default="true" id="e5615680-c090-497d-a104-1337ed85f850" name="Changes" comment="parser">
<list default="true" id="e5615680-c090-497d-a104-1337ed85f850" name="Changes" comment="Первый коммит">
<change beforePath="$PROJECT_DIR$/.idea/workspace.xml" beforeDir="false" afterPath="$PROJECT_DIR$/.idea/workspace.xml" afterDir="false" />
</list>
<option name="SHOW_DIALOG" value="false" />
@@ -20,7 +20,7 @@
</option>
</component>
<component name="Git.Rebase.Settings">
<option name="NEW_BASE" value="master" />
<option name="NEW_BASE" value="origin/master" />
</component>
<component name="Git.Settings">
<option name="RECENT_GIT_ROOT_PATH" value="$PROJECT_DIR$" />
@@ -35,11 +35,11 @@
<option name="hideEmptyMiddlePackages" value="true" />
<option name="showLibraryContents" value="true" />
</component>
<component name="PropertiesComponent"><![CDATA[{
"keyToString": {
"last_opened_file_path": "C:/Users/71332/PycharmProjects/c22712"
<component name="PropertiesComponent">{
&quot;keyToString&quot;: {
&quot;last_opened_file_path&quot;: &quot;C:/Users/71332/PycharmProjects/c22712&quot;
}
}]]></component>
}</component>
<component name="RunManager" selected="Python.zaparsil">
<configuration name="main" type="PythonConfigurationType" factoryName="Python" nameIsGenerated="true">
<module name="22712" />
@@ -169,7 +169,7 @@
<breakpoints>
<line-breakpoint enabled="true" suspend="THREAD" type="python-line">
<url>file://$PROJECT_DIR$/main.py</url>
<line>8</line>
<line>5</line>
<option name="timeStamp" value="1" />
</line-breakpoint>
</breakpoints>
+41
View File
@@ -0,0 +1,41 @@
import psycopg2
import wget
from bs4 import BeautifulSoup
from selenium import webdriver
from selenium.webdriver.chrome.service import Service
connection = psycopg2.connect(host='localhost', dbname='dbdata', user='postgres', password='Q1w2e3r4')
cursor = connection.cursor()
create_table = """ create table dresses
(id serial primary key, name varchar(200), sale varchar(50), price varchar(50), info varchar(50), picture varchar(200))"""
cursor.execute(create_table)
connection.commit()
driver = Service('D:\data\chromedriver.exe')
browser = webdriver.Chrome(service=driver)
browser.get('https://www.ozon.ru/category/platya-zhenskie-7502/')
html_code = browser.page_source
soup = BeautifulSoup(html_code, 'lxml')
name = soup.find_all('span', class_="m2e e3m m3e m5e tsBodyL k7l l7k")
sale = soup.find_all('div', class_="eg1 g3e")
price = soup.find_all('span', class_="a2a-a2")
info = soup.find_all('span', class_="je4")
picture = soup.find_all('div', class_="k1m")
for i in range(20):
url = picture[i].find('img').attrs['src']
file_name = f"C:\\Users\\Honor\\Desktop\\учеба\\прога\\pictures\\{i}.jpg"
wget.download(url, file_name)
ins_table = f"""insert into public.dresses(name, sale, price, info, picture) values ('{name[i].text}', '{sale[i].text}', '{price[i].text}', '{info[i].text}', '{file_name}')"""
cursor.execute(ins_table)
connection.commit()
cursor.close()
connection.close()