A cute little Royal Opera House website parser using Selenium

roh_parser.py

#!/usr/bin/env python
from email.mime.text import MIMEText
from pyvirtualdisplay.smartdisplay import SmartDisplay
from selenium import webdriver
import difflib
import os.path
import random
import smtplib
import time

def parse_roh():
    from_addr = '[email protected]'
    to_addr = '[email protected]'
    # Start sleep
    #time.sleep(random.randint(1,60))
    display = SmartDisplay(visible=0, size=(1024, 768), bgcolor='white', backend='xvfb')
    display.start()
    browser = webdriver.Chrome('/usr/bin/chromedriver')
    event_ids = [
                 ('19766', u"La Boheme on Sat 9th March"),
                 ('20351', u'Tosca on Sat 16th March'),
                 ('20353', u'Tosca on Sat 23rd March'),
                 ]
    output = []
    num_diffs = 0
    for event_id, event_name in event_ids:
        # Sleep between gets
        time.sleep(random.randint(2,10))
        url = u'http://www.roh.org.uk/events/%s/tickets/syos' % event_id
        browser.get(url)
        #img = display.waitgrab()
        #img.save('img_%s_%s.jpeg' % (event_id, int(time.time())), 'JPEG')
        import ipdb
        ipdb.set_trace()
        sections = browser.find_elements_by_xpath("//li[@class='panel-section on']/div[@class='sections']/ul/li")
        full_stat = []
        for section in sections:
            section_name = section.find_element_by_tag_name('h1').text
            prices = section.find_element_by_tag_name('h2').text
            status = section.find_element_by_tag_name('h3').text
            full_stat.append(u"Section: %s, Price: %s, Status: %s ... URL=%s" % (section_name, prices, status, url))
        full_stat = u'\n'.join(full_stat)
        old_stat = u''
        fname = 'roh_status_%s.txt' % event_id
        if os.path.exists(fname):
            with open(fname, 'r') as f:
                old_stat = f.read().decode('utf-8')
        if full_stat != old_stat:
            with open(fname, 'w') as f:
                f.write(full_stat.encode('utf-8'))
            num_diffs += 1
            output.append(u"Found differences for event: %s" % event_name)
            diff = difflib.ndiff(old_stat.splitlines(), full_stat.splitlines())
            output += diff
            output.append(u'')
            output.append(u'')
    browser.quit()
    display.stop()
    if output:
        msg = MIMEText((u'\n'.join(output)).encode('utf-8'), 'plain', 'UTF-8')
        msg['Subject'] = 'Royal Opera House - Found %s differences' % num_diffs
        msg['From'] = from_addr
        msg['To'] = to_addr
        s = smtplib.SMTP('localhost')
        s.sendmail(from_addr, [to_addr], msg.as_string())
        s.quit()

if __name__ == '__main__':
    parse_roh()