Alex A. alecxe
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from selenium import webdriver | |
| url = 'http://www.archives.com/member/Default.aspx?_act=VitalSearchResult&LastName=Smith&Country=US&State=VA&RecordType=2&DeathYear=2004&DeathYearSpan=10&pagesize=10&pageNumber=1&pagesizeAP=10&pageNumberAP=1' | |
| driver = webdriver.Firefox() | |
| driver.get(url) | |
| print driver.page_source |
alecxe
/ scraper.py
Created
March 10, 2014 06:10
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import csv | |
| import urllib2 | |
| from bs4 import BeautifulSoup | |
| def record(part): | |
| soup = BeautifulSoup(urllib2.urlopen("http://www.admision.unmsm.edu.pe/admisionsabado".format(part))) | |
| links = [link.text for link in soup.find('table').find_all('a')[2:]] | |
| t = (len(links)) / 2 |
alecxe
/ mechanize_bs_textatble.py
Last active
August 29, 2015 13:56
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import mechanize | |
| from bs4 import BeautifulSoup | |
| import texttable | |
| med = raw_input("Enter the drugname") | |
| br = mechanize.Browser() | |
| br.set_handle_robots(False) | |
| res = br.open("http://www.medindia.net/drug-price/") | |
| br.select_form("frmdruginfo_search") |
NewerOlder