Python PhantomJS.find_elements_by_css_selector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 find_elements_by_css_selector [as 别名]
def main(argv=sys.argv[1:]):
    parser = argparse.ArgumentParser()
    parser.add_argument('--url', default='http://127.0.0.1:8000/static/index.html')
    args = parser.parse_args(argv)

    url = args.url

    browser = WebDriver()
    browser.get(url)
    tags = browser.find_elements_by_css_selector('li')
    for tag in tags:
        print(tag.text)
    browser.close()

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import find_elements_by_css_selector [as 别名]
class WeixinPhantomjs(Base):
    all_uids = {docs['uid'] for docs in in_collection.find({}, {'uid': 1}) if 'uid' in docs}

    def __init__(self):
        self.start_page = START_PAGE
        self.end_page = END_PAGE
        self.weixin_url = REFER_FIRST

        # self.driver = Firefox()
        if hasattr(config, 'PHANTOMJS_PATH'):
            self.driver = PhantomJS(executable_path=getattr(config, 'PHANTOMJS_PATH'))
        else:
            self.driver = PhantomJS()

    def open_weixin_browser(self, word):
        try:
            self.driver.get(self.weixin_url)
            self.driver.set_page_load_timeout(3)

            self.driver.find_element_by_id('upquery').send_keys(word)
            self.driver.find_element_by_class_name('swz').click()
            time.sleep(3)

            urls_uids = self.extract_urls_uids(word=word)
            Article(urls_uids=urls_uids, word=word).extract()
        except Exception as e:
            storage_word.append([word, 0])
            self.logger.info('Open weixin error: type <{}>, mag <{}>'.format(e.__class__, e))
            self.close_browser()
            return True
        return False

    def get_total_pages_to_word(self):
        pages = []
        page_id_css = 'pagebar_container'

        try:
            e = self.driver.find_element_by_id(page_id_css)
            for _p in e.text.split():
                _p = _p.strip()

                if not _p.isdigit():
                    return DEFAULT_PAGES if DEFAULT_PAGES <= pages[-1] else pages[-1]
                else:
                    pages.append(int(_p))
            return 1
        except (NoSuchElementException, NoSuchWindowException, TypeError, IndexError):
            pass
        return 1

    def extract_urls_uids(self, word):
        urls_uids = []
        timestamp = [_t.get_attribute('t') for _t in self.driver.find_elements_by_css_selector('div.s-p')]
        urls_tits = [(t.get_attribute('href'), self.trim(t.text))
                     for t in self.driver.find_elements_by_css_selector('h4 a')]

        if len(urls_tits) != len(timestamp):
            return urls_uids

        for index, url_tit in enumerate(urls_tits):
            try:
                uid = self.md5(timestamp[index] + url_tit[1] + word)

                if uid not in self.__class__.all_uids:
                    self.__class__.all_uids.add(uid)
                    urls_uids.append({'url': url_tit[0], 'uid': uid})
            except (TypeError, IndexError):
                pass
        return urls_uids

    @property
    def is_forbidden(self):
        css_id = 'seccodeForm'

        try:
            if self.driver.find_element_by_id(css_id):
                return True
        except NoSuchElementException:
            pass
        return False

    def appear_element(self, by):
        try:
            # Have `click` function to specified element
            tag = WebDriverWait(self.driver, 20).until(lambda driver: driver.find_element_by_id(by))
            tag.click()
            return True
        except (TimeoutException, NoSuchWindowException, NoSuchElementException):
            pass
        return False

    def crawl_single(self, word=None, go=0):
        is_go = True
        go_page = int(go)
        next_page_css = 'sogou_page_%s'

        is_break = self.open_weixin_browser(word)
        pages = self.get_total_pages_to_word()

        for page in range(self.start_page + 1, (pages or self.end_page) + 1):
#.........这里部分代码省略.........

# 需要导入模块: from selenium.webdriver import PhantomJS [as 别名]
# 或者: from selenium.webdriver.PhantomJS import find_elements_by_css_selector [as 别名]
class WeixinPhantomjs(Base):
    def __init__(self):
        self.start_page = START_PAGE
        self.end_page = END_PAGE
        self.weixin_url = REFER_FIRST

        # self.driver = Firefox()
        if hasattr(config, 'PHANTOMJS_PATH'):
            self.driver = PhantomJS(executable_path=getattr(config, 'PHANTOMJS_PATH'))
        else:
            self.driver = PhantomJS()

        self.client = MongoClient(HOST, PORT)
        self.collection = self.client[DB][COLLECTION]
        self.all_uids = self.uids

    def open_weixin_browser(self, word):
        try:
            self.driver.get(self.weixin_url)
            self.driver.set_page_load_timeout(3)

            self.driver.find_element_by_id('upquery').send_keys(word)
            self.driver.find_element_by_class_name('swz').click()
            time.sleep(3)

            urls_uids = self.extract_urls_uids(word=word)
            Article(urls_uids=urls_uids, word=word).extract()
        except Exception as e:
            storage_word.append([word, 0])
            self.logger.info('Open weixin error: type <{}>, mag <{}>'.format(e.__class__, e))
            self.close_browser()
            return True
        return False

    def get_total_pages_to_word(self):
        pages = []
        page_id_css = 'pagebar_container'

        try:
            e = self.driver.find_element_by_id(page_id_css)
            for _p in e.text.split():
                _p = _p.strip()

                if not _p.isdigit():
                    return DEFAULT_PAGES if DEFAULT_PAGES <= pages[-1] else pages[-1]
                else:
                    pages.append(int(_p))
            return 1
        except (NoSuchElementException, NoSuchWindowException, TypeError, IndexError):
            pass
        return 1

    def get_query_words(self, word):
        query_words = []

        for docs in self.collection.find({}, {'rel': 1, 'conp': 1}).sort([('_id', 1)]):
            w = docs['conp']

            if w not in query_words:
                query_words.append(w)

            for item in docs['rel']:
                if item not in query_words:
                    query_words.append(item)

        self.client.close()

        return self.query_index(query_words, word)

    @property
    def uids(self):
        return {docs['uid'] for docs in in_collection.find({}, {'uid': 1}) if 'uid' in docs}

    def extract_urls_uids(self, word):
        urls_uids = []
        timestamp = [_t.get_attribute('t') for _t in self.driver.find_elements_by_css_selector('div.s-p')]
        urls_tits = [(t.get_attribute('href'), self.trim(t.text))
                     for t in self.driver.find_elements_by_css_selector('h4 a')]

        if len(urls_tits) != len(timestamp):
            return urls_uids

        for index, url_tit in enumerate(urls_tits):
            try:
                uid = self.md5(timestamp[index] + url_tit[1] + word)

                if uid not in self.all_uids:
                    self.all_uids.add(uid)
                    urls_uids.append({'url': url_tit[0], 'uid': uid})
            except (TypeError, IndexError):
                pass
        return urls_uids

    @staticmethod
    def query_index(words, cut_word):
        temp_words = words[START_INDEX:END_INDEX]

        try:
            index = temp_words.index(cut_word)
            return temp_words[index:], index + START_INDEX
#.........这里部分代码省略.........