Python PhantomJS.set_script_timeout方法代码示例

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import set_script_timeout [as 别名]
class PagesCrawler(BaseSpider):

    name = 'pages'
    link_extractor = RegexpLinkExtractor(canonicalize=False, deny_extensions=[])
    ignored_exts = set(['.' + e for e in IGNORED_EXTENSIONS])

    def __init__(self, **kw):
        args = DEFAULT_INPUT.copy()
        args.update(kw)
        self.args = args
        self.start_urls = to_list(args['start_urls'])
        self.maxdepth = int(args['maxdepth'])
        self.follow_prefixes = to_list(args['follow_prefixes'])
        self.nofollow_prefixes = to_list(args['nofollow_prefixes'])
        self.discover_prefixes = [url_to_lru_clean("http%s://%s" % (https, u.replace('http://', '').replace('https://', ''))) for u in to_list(args['discover_prefixes']) for https in ['', 's']]
        self.resolved_links = {}
        self.user_agent = args['user_agent']
        self.phantom = 'phantom' in args and args['phantom'] and args['phantom'].lower() != "false"
        if self.phantom:
            self.ph_timeout = int(args.get('phantom_timeout', PHANTOM['TIMEOUT']))
            self.ph_idle_timeout = int(args.get('phantom_idle_timeout', PHANTOM['IDLE_TIMEOUT']))
            self.ph_ajax_timeout = int(args.get('phantom_ajax_timeout', PHANTOM['AJAX_TIMEOUT']))
        self.errors = 0
        dispatcher.connect(self.closed, spider_closed)
        dispatcher.connect(self.crashed, spider_error)

    def start_requests(self):
        self.log("Starting crawl task - jobid: %s" % self.crawler.settings['JOBID'], log.INFO)
        self.log("ARGUMENTS : "+str(self.args), log.INFO)
        if self.phantom:
            self.init_phantom()
        for url in self.start_urls:
            yield self._request(url)

    def init_phantom(self):
        self.prefixfiles = os.path.join(
            scrapyd_config().get('logs_dir'),
            HYPHE_PROJECT,
            self.name,
            self.crawler.settings['JOBID']
        )
        self.log("Using path %s for PhantomJS crawl" % self.prefixfiles, log.INFO)
        phantom_args = []
        if PROXY and not PROXY.startswith(':'):
            phantom_args.append('--proxy=%s' % PROXY)
        phantom_args.append('--cookies-file=%s-phantomjs-cookie.txt' % self.prefixfiles)
        phantom_args.append('--ignore-ssl-errors=true')
        phantom_args.append('--load-images=false')
        self.capabilities = dict(DesiredCapabilities.PHANTOMJS)
        self.capabilities['phantomjs.page.settings.userAgent'] = self.user_agent
        self.capabilities['takesScreenshot'] = False
        self.capabilities['phantomjs.page.settings.javascriptCanCloseWindows'] = False
        self.capabilities['phantomjs.page.settings.javascriptCanOpenWindows'] = False
        self.phantom = PhantomJS(
            executable_path=PHANTOM['PATH'],
            service_args=phantom_args,
            desired_capabilities=self.capabilities,
            service_log_path="%s-phantomjs.log" % self.prefixfiles
        )
        self.phantom.implicitly_wait(10)
        self.phantom.set_page_load_timeout(60)
        self.phantom.set_script_timeout(self.ph_timeout + 15)

    def crashed(self, spider):
        self.errors += 1
        self.closed("CRASH")

    def closed(self, reason):
        if self.errors:
            self.log("%s error%s encountered during the crawl." %
                (self.errors, 's' if self.errors > 1 else ''), log.ERROR)
        if self.phantom:
            self.phantom.quit()
            if not self.errors:
                for f in ["phantomjs-cookie.txt", "phantomjs.log"]:
                    fi = "%s-%s" % (self.prefixfiles, f)
                    if os.path.exists(fi) and not self.errors:
                        os.remove(fi)

    def handle_response(self, response):
        lru = url_to_lru_clean(response.url)

        if self.phantom:
            self.phantom.get(response.url)

          # Collect whole DOM of the webpage including embedded iframes
            with open(os.path.join(PHANTOM["JS_PATH"], "get_iframes_content.js")) as js:
                get_bod_w_iframes = js.read()
            bod_w_iframes = self.phantom.execute_script(get_bod_w_iframes)
            response._set_body(bod_w_iframes.encode('utf-8'))

          # Try to scroll and unfold page
            self.log("Start PhantomJS scrolling and unfolding", log.INFO)
            with open(os.path.join(PHANTOM["JS_PATH"], "scrolldown_and_unfold.js")) as js:
                try:
                    signal.signal(signal.SIGALRM, timeout_alarm)
                    signal.alarm(self.ph_timeout + 30)
                    timedout = self.phantom.execute_async_script(
                        js.read(), self.ph_timeout,
                        self.ph_idle_timeout, self.ph_ajax_timeout)
#.........这里部分代码省略.........