headless.py

import re

from pyvirtualdisplay import Display
from selenium import webdriver

count_re = re.compile('(\d+) people')

class selenium_firefox():
    def __enter__(self):
        display, browser = self.setup()
        self._display = display
        self._browser = browser
        return self

    def __exit__(self, type, value, traceback):
        self.shutdown(self._display, self._browser)

    def setup(self):
        display = Display(visible=0, size=(800, 600))
        display.start()

        browser = webdriver.Firefox()

        return display, browser


    def parse_count(self, url):
        self._browser.get(url)

        iframe = self._browser.find_element_by_tag_name('iframe')
        if iframe is not None:
            name =  iframe.get_attribute('name')
            self._browser.switch_to_frame(name)
            matches = count_re.search(self._browser.page_source)
            return matches.group(1)
        else:
            return None


    def shutdown(self, display, browser):
        browser.quit()
        display.stop()



if __name__ == '__main__':
    with selenium_firefox() as sf:
        print sf.parse_count('http://dump.zoranzaric.de/plus1.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus2.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus1.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus2.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus1.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus2.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus1.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus2.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus1.html')
        print sf.parse_count('http://dump.zoranzaric.de/plus2.html')