airbnb_superhost_photo_crawler.py 文件源码

def get_name(uid):
    url = 'https://www.airbnb.com/users/show/' + uid + '?locale=en'
    cj = cookielib.MozillaCookieJar()
    opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
    req = urllib2.Request(url)
    req.add_header("User-agent", 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) \
                    Chrome/50.0.2661.102 Safari/537.36')
    try:
        response = opener.open(req, timeout=10)
    except Exception as inst:
        print type(inst)
        print inst.args
        print inst
        print('-----fail to get name data')
        return ''
    data = response.read()
    name = re.findall('Hey, I’m (.*?)!', data)
    if len(name):
        return name[0]
    else:
        return ''