一个用Python写的 CSDN 批量文章分类修改脚本

昨天写Java写的有点累了，看着自己的博客分类乱七八糟的，遂心生了整理一下博客文章分类的想法。

手动整理了一会发现好累啊，重复的工作让程序来做就好了嘛。

于是就写了这个小脚本。

user_request_to_change.py

import requests
from lxml import etree
import re


# 获取指定类别的文章的 url
def get_article_url_list(page_url, session_obj):
    page_article_list = session_obj.get(page_url).text
    my_selector = etree.HTML(page_article_list)
    article_url_list = my_selector.xpath('//td[@class="tdleft"]/a/@href')
    return article_url_list


# 获取包含指定关键字的文章的 url
def get_spcify_article_url_list(page_url, session_obj):
    page_article_list = session_obj.get(page_url).text
    my_selector = etree.HTML(page_article_list)
    article_url_list = my_selector.xpath('//a[contains(text(),"基于《PythonCookbook》")]/@href')
    return article_url_list


if __name__ == "__main__":
    session = requests.Session()

    initial_category = 'Django'
    target_categhory = '------Python------'

    # 设置 cookie
    session.cookies.set('Cookie',
                        'yourCookie')

    # 这个 option 标签的文本内容是用 js 动态生成的...
    # 获取初始目录和目标目录的 id
    page_source_code = session.get('http://write.blog.csdn.net/postlist').text
    selector = etree.HTML(page_source_code)
    initial_category_id = selector.xpath('//option[contains(text(),"{}")]/@value'.format(initial_category))
    target_categhory_id = selector.xpath('//option[contains(text(),"{}")]/@value'.format(target_categhory))

    print({initial_category: initial_category_id, target_categhory: target_categhory_id})

    # 获取该分类的总页数
    page_source_code = session.get('http://write.blog.csdn.net/postlist/{}/all'.format(initial_category_id)).text
    selector = etree.HTML(page_source_code)
    dirty_page_nums = selector.xpath('//div[@class="page_nav"]/span/text()')[0]
    page_nums = int(re.findall('\d+', dirty_page_nums)[-1])
    print(dirty_page_nums)

    all_article_url_list = []  # 存储该分类下全部文章的 url
    article_id_list = []  # 存储该分类下全部文章的 id

    # 获取该分类全部文章的 url
    for i in range(page_nums):
        url = 'http://write.blog.csdn.net/postlist/{}/0/enabled/{}'.format(initial_category_id, i + 1)
        # all_article_url_list += get_article_url_list(url, session)
        all_article_url_list += get_spcify_article_url_list(url, session)

    print(all_article_url_list)

    # 获取该分类全部文章的 id
    for article_url in all_article_url_list:
        article_id_list.append(int(re.findall('\d+', article_url)[0]))

    print(article_id_list)

    initial_category_id = 7324441
    target_categhory_id = 6897792

    # 设置分类
    for article_id in article_id_list:
        url = 'http://write.blog.csdn.net/article/setcategory?id={}&cats={}%2C{}'.format(article_id,
                                                                                         target_categhory_id,
                                                                                         initial_category_id)
        session.get(url)

        # url = 'http://write.blog.csdn.net/article/setcategory?id={}&cats={}'.format(article_id,
        #                                                                             target_categhory_id,
        #                                                                             )
        # session.get(url)

整理的效果如下图

变得整齐美观又大方了有木有！有木有！！

---

其实还有另外一个基于 Selenium 的失败的程序

csdnChangeClass.py

from selenium import webdriver

from xpath_map_dict import XPATH_MAP_DICT
from lxml import etree
import re

driver = webdriver.Chrome()

initial_category = 'Python基础'
target_categhory = '------Python------'


try:
    csdn_login_page = "https://passport.csdn.net/account/login?from=http://my.csdn.net/my/mycsdn"
    driver.get(csdn_login_page)

    driver.maximize_window()

    # 登录用户
    username = "username"
    password = "password"
    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_login_username_input']).clear()
    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_login_username_input']).send_keys(username)

    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_login_password_input']).clear()
    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_login_password_input']).send_keys(password)

    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_login_remberMe_input']).click()

    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_login_button']).click()

    # 跳转到文章管理页面
    driver.get('http://write.blog.csdn.net/postlist')

    # 获取页面源码
    page_source = driver.page_source
    selector = etree.HTML(page_source)

    # 选择需要修改的分类
    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_select_category']).click()
    driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_option_cat_1'].format(initial_category)).click()

    # 获取此页文章数量
    page_article_list = selector.xpath("//tbody/tr")
    page_article_nums = len(page_article_list) - 1
    # print(page_article_nums)

    # 获取该分类的总页数
    dirty_page_nums = selector.xpath('//div[@class="page_nav"]/span/text()')
    page_nums = int(re.findall('\d+', dirty_page_nums)[-1])
    # print(page_nums)

    scroll_js = "var q=document.documentElement.scrollTop=10000"

    for page in range(page_nums):
        # 修改此页文章的分类
        for page_article in range(page_article_nums):
            if page_article == 8:
                driver.execute_script(scroll_js)

            driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_a_cat'].format(target_categhory)).click()
            driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_input_category'].format((page_article + 1))).click()
            driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_input_close']).click()

        # 翻页
        driver.find_element_by_xpath(XPATH_MAP_DICT['xpath_a_next_page']).click()

except:
    pass
finally:
    driver.quit()
    driver.close()

xpath_map_dict.py

XPATH_MAP_DICT = {
    'xpath_login_username_input':       '//input[@id="username"]',
    'xpath_login_password_input':       '//input[@id="password"]',
    'xpath_login_remberMe_input':       '//input[@id="rememberMe"]',
    'xpath_login_button':               '//input[@value="登 录"]',
    'xpath_select_category':            '//select[@value="selCat"]',
    'xpath_option_cat_1':               '//option[contians(text(),"{}")]',
    'xpath_a_cat':                      '(//a[@class="cat"])[{}]',
    'xpath_input_category':             '//div[@id="setcat_box"]/label/span[contains(text(),"{}")]/../input',
    'xpath_input_close':                '//*[@id="setcat_div"]/div/input',
    'xpath_a_next_page':                '//a[contains(text(),"下一页")]',
}

失败原因，在测试程序的时候，总是报错说是“目标计算机积极拒绝连接”balabala的。