http_cache.py 文件源码

def keep_redirecting(r, my_pub):
    # don't read r.content unless we have to, because it will cause us to download the whole thig instead of just the headers

    # 10.5762/kais.2016.17.5.316
    if ("content-length" in r.headers):
        # manually follow javascript if that's all that's in the payload
        file_size = int(r.headers["content-length"])
        if file_size < 500:
            matches = re.findall(ur"<script>location.href='(.*)'</script>", r.content, re.IGNORECASE)
            if matches:
                redirect_url = matches[0]
                if redirect_url.startswith(u"/"):
                    redirect_url = get_link_target(redirect_url, r.url)
                return redirect_url

    # 10.1097/00003643-201406001-00238
    if my_pub and my_pub.is_same_publisher("Ovid Technologies (Wolters Kluwer Health)"):
        matches = re.findall(ur"OvidAN = '(.*?)';", r.content, re.IGNORECASE)
        if matches:
            an_number = matches[0]
            redirect_url = "http://content.wkhealth.com/linkback/openurl?an={}".format(an_number)
            return redirect_url

    return None