Add files via upload

Author: rajat4665

flipkart_scraping.ipynb

@@ -0,0 +1,115 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "pop-up closed\n",
+      "Task completed\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os \n",
+    "from selenium import webdriver\n",
+    "from bs4 import BeautifulSoup\n",
+    "import time\n",
+    "import csv\n",
+    "\n",
+    "\n",
+    "class Flipkart():\n",
+    "    \n",
+    "    def __init__(self):\n",
+    "        self.current_path= os.getcwd()\n",
+    "        self.url= 'https://www.flipkart.com'\n",
+    "        self.driver_path= os.path.join(os.getcwd(),'chromedriver')\n",
+    "        self.driver= webdriver.Chrome(self.driver_path)\n",
+    "        \n",
+    "    def page_load(self):\n",
+    "        self.driver.get(self.url)\n",
+    "        try:\n",
+    "            login_pop = self.driver.find_element_by_class_name('_29YdH8')\n",
+    "            login_pop.click()\n",
+    "            print('pop-up closed')\n",
+    "        except:\n",
+    "            pass\n",
+    "        search_field = self.driver.find_element_by_class_name('LM6RPg')\n",
+    "        search_field.send_keys('smartphone'+ '\\n')\n",
+    "        time.sleep(2)\n",
+    "        page_html = self.driver.page_source\n",
+    "        self.soup = BeautifulSoup(page_html,'html.parser')\n",
+    "    \n",
+    "    def create_csv_file(self):\n",
+    "        rowHeaders=[\"Name\",\"Storage_details\",\"Screen_size\",\"Camera_details\",\"Battery_details\",\"Processor\",\n",
+    "                    \"Warranty\",\"Price in Rupees\"]\n",
+    "        self.file_csv = open('Flipkart_output.csv', 'w', newline='', encoding='utf-8')\n",
+    "        self.mycsv = csv.DictWriter(self.file_csv, fieldnames=rowHeaders)\n",
+    "        self.mycsv.writeheader()\n",
+    "    \n",
+    "    def data_scrap(self):\n",
+    "        first_page_mobiles = (self.soup.find_all('div',class_='_3O0U0u'))   \n",
+    "        for i in first_page_mobiles:\n",
+    "            Name = i.find('img',class_ ='_1Nyybr')['alt']\n",
+    "            price = i.find('div',class_ ='_1vC4OE _2rQ-NK')\n",
+    "            details = i.find_all(\"li\")\n",
+    "            storage = details[0].text\n",
+    "            screen_size = details[1].text\n",
+    "            camera_details = details[2].text\n",
+    "            battery_details =details[3].text\n",
+    "            processor = details[4].text\n",
+    "            try:\n",
+    "                warranty_details = [j.text for j in details if j.text[:14] == \"Brand Warranty\"][0]\n",
+    "            except:\n",
+    "                warranty_details = \"No data available\"\n",
+    "            price = price.text[1:]\n",
+    "            self.mycsv.writerow({\"Name\":Name, \"Storage_details\":storage, \"Screen_size\":screen_size, \"Camera_details\":camera_details,\"Battery_details\":battery_details, \"Processor\":processor, \"Warranty\":warranty_details, \"Price in Rupees\":price})\n",
+    "            \n",
+    "    def tearDown(self):\n",
+    "        self.driver.quit()\n",
+    "        self.file_csv.close()\n",
+    "\n",
+    "if __name__==\"__main__\":\n",
+    "    \n",
+    "    Flipkart=Flipkart()\n",
+    "    Flipkart.page_load()\n",
+    "    Flipkart.create_csv_file()\n",
+    "    Flipkart.data_scrap()\n",
+    "    Flipkart.tearDown()\n",
+    "    print(\"Task completed\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.5.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}