click_page.py 文件源码

def _click_page(total_posts, pool_size, group_index):
    _log.info('?{}?: starting...'.format(group_index + 1))
    if group_index > 0 and total_posts < pool_size * group_index:
        return
    # ????????
    _driver = webdriver.PhantomJS()
    _driver.get('https://www.xncoding.com/archives/')

    global TRY_COUNT
    for k in range(1, TRY_COUNT + 1):
        # _log.info('?{}?: ?{}???...'.format(group_index + 1, k))
        for i in range(pool_size * group_index, min(pool_size * (group_index + 1), total_posts)):
            l_xpath = '(//article/header/h1[@class="post-title"]/a[@class="post-title-link"])[{}]'.format(i + 1)
            ele = WebDriverWait(_driver, 2).until(
                EC.presence_of_element_located((By.XPATH, l_xpath))
            )
            ele.click()
            WebDriverWait(_driver, 5).until(
                EC.presence_of_element_located((By.XPATH, '//div[@class="post-body"]'))
            )
            _driver.back()

    _log.info('?{}?: finished.'.format(group_index + 1))
    _driver.close()