python learning to capture blog park news

前言

　　说到python，对它有点耳闻的人，第一反应可能都是爬虫~

　　这两天看了点python的皮毛知识，忍不住想写一个简单的爬虫练练手，JUST DO IT

准备工作

　　要制作数据抓取的爬虫，对请求的源页面结构需要有特定分析，只有分析正确了，才能更好更快的爬到我们想要的内容。

　　浏览器访问570973/，右键“查看源代码”，初步只想取一些简单的数据（文章标题、作者、发布时间等），在HTML源码中找到相关数据的部分：

　　1）标题（url）：90f4f834e5b06d6096bb681240960ddba69c86dd6867f396eeb12803a320ba93SpaceX重复使用的“龙”飞船成功与国际空间站对接5db79b134e9f6b82c0b36e0489ee08ed16b28748ea4df4d9c2150843fecfba68

#　2) Author: 9df2630932accbf34ebd32561ffefb37itwriter5db79b134e9f6b82c0b36e0489ee08ed54bdf357c58b8a65c66d7c19c8e4d114

　3) Release time: 26481bc0d1e3a64b3d556a664a12a9d8 Published on 2017-06-06 14:5354bdf357c58b8a65c66d7c19c8e4d114

#　 4) Current news ID: ##6459ef0b0d65a859ec0ffa39053271c2

##　

Of course, if you want to follow the lead, the structure of the "previous article" and "next article" links is very important; but I found a problem, the two 3499910bf9dac5ae3c52d5ede7383485 tags in the page, their links and text content , is rendered through js, what should I do? Try to find information (python executes js and the like), but for python novices, it may be a bit ahead of the curve and I plan to find another solution. Although these two links are rendered through js, in theory, the reason why js can render the content should be by initiating a request and getting the response. Then is it possible to monitor the web page? Check out the loading process to see what useful information there is? I would like to give a thumbs up to browsers such as chrome/firefox. Developer Tools/Network can clearly see the request and response status of all resources. 　

Their request addresses are:

1) Previous news ID:

2) Next news ID:

The content of the response is JSON

The ContentID here is what we need. Based on this value, we can know the previous or next article of the current news News URL, because the page address of news releases has a fixed format:

{{ContentID}}

/ (The red content is the replaceable ID)

Tools

##　1) python 3.6 (install pip at the same time during installation, and add environment variables)

　2) PyCharm 2017.1.3

　3) Third-party python library (installation: cmd -> pip install name)

a) pyperclip: used to read and write the clipboard

b) requests: an HTTP library based on urllib and using the Apache2 Licensed open source protocol. It is more convenient than urllib and can save us a lot of work

c) beautifulsoup4: Beautifulsoup provides some simple, python-style functions to handle navigation, search, modify parse trees, etc. Function. It is a toolbox that provides users with the data they need to crawl by parsing documents

Source code

Personally I think the codes are very basic and easy to understand (after all, novices can’t write advanced code). If you have any questions or suggestions, please feel free to let me know

#! python3
# coding = utf-8
# get_cnblogs_news.py
# 根据博客园内的任意一篇新闻，获取所有新闻（标题、发布时间、发布人）
# 

# 这是标题格式 ：<div id="news_title"><a href="//news.cnblogs.com/n/570973/">SpaceX重复使用的“龙”飞船成功与国际空间站对接</a></div>
# 这是发布人格式 ：<span class="news_poster">投递人 <a href="//home.cnblogs.com/u/34358/">itwriter</a></span>
# 这是发布时间格式 ：<span class="time">发布于 2017-06-06 14:53</span>

# 当前新闻ID ：<input type="hidden" value="570981" id="lbContentID">

# html中获取不到上一篇和下一篇的直接链接，因为它是使用ajax请求后期渲染的
# 需要另外请求地址，获取结果，JSON
# 上一篇 
# 下一篇 

# 响应内容
# ContentID : 570971
# Title : "Mac支持外部GPU VR开发套件售599美元"
# Submitdate : "/Date(1425445514)"
# SubmitdateFormat : "2017-06-06 14:47"

import sys, pyperclip
import requests, bs4
import json

# 解析并打印（标题、作者、发布时间、当前ID）
# soup : 响应的HTML内容经过bs4转化的对象
def get_info(soup):
    dict_info = {&#39;curr_id&#39;: &#39;&#39;, &#39;author&#39;: &#39;&#39;, &#39;time&#39;: &#39;&#39;, &#39;title&#39;: &#39;&#39;, &#39;url&#39;: &#39;&#39;}

    titles = soup.select(&#39;div#news_title > a&#39;)
    if len(titles) > 0:
        dict_info[&#39;title&#39;] = titles[0].getText()
        dict_info[&#39;url&#39;] = titles[0].get(&#39;href&#39;)

    authors = soup.select(&#39;span.news_poster > a&#39;)
    if len(authors) > 0:
        dict_info[&#39;author&#39;] = authors[0].getText()

    times = soup.select(&#39;span.time&#39;)
    if len(times) > 0:
        dict_info[&#39;time&#39;] = times[0].getText()

    content_ids = soup.select(&#39;input#lbContentID&#39;)
    if len(content_ids) > 0:
        dict_info[&#39;curr_id&#39;] = content_ids[0].get(&#39;value&#39;)

    # 写文件
    with open(&#39;D:/cnblognews.csv&#39;, &#39;a&#39;) as f:
        text = &#39;%s,%s,%s,%s\n&#39; % (dict_info[&#39;curr_id&#39;], (dict_info[&#39;author&#39;] + dict_info[&#39;time&#39;]), dict_info[&#39;url&#39;], dict_info[&#39;title&#39;])
        print(text)
        f.write(text)
    return dict_info[&#39;curr_id&#39;]

# 获取前一篇文章信息
# curr_id : 新闻ID
# loop_count : 向上多少条，如果为0，则无限向上，直至结束
def get_prev_info(curr_id, loop_count = 0):
    private_loop_count = 0
    try:
        while loop_count == 0 or private_loop_count < loop_count:
            res_prev = requests.get(&#39;https://news.cnblogs.com/NewsAjax/GetPreNewsById?contentId=&#39; + curr_id)
            res_prev.raise_for_status()
            res_prev_dict = json.loads(res_prev.text)
            prev_id = res_prev_dict[&#39;ContentID&#39;]

            res_prev = requests.get(&#39;https://news.cnblogs.com/n/%s/&#39; % prev_id)
            res_prev.raise_for_status()
            soup_prev = bs4.BeautifulSoup(res_prev.text, &#39;html.parser&#39;)
            curr_id = get_info(soup_prev)

            private_loop_count += 1
    except:
        pass

# 获取下一篇文章信息
# curr_id : 新闻ID
# loop_count : 向下多少条，如果为0，则无限向下，直至结束
def get_next_info(curr_id, loop_count = 0):
    private_loop_count = 0
    try:
        while loop_count == 0 or private_loop_count < loop_count:
            res_next = requests.get(&#39;https://news.cnblogs.com/NewsAjax/GetNextNewsById?contentId=&#39; + curr_id)
            res_next.raise_for_status()
            res_next_dict = json.loads(res_next.text)
            next_id = res_next_dict[&#39;ContentID&#39;]

            res_next = requests.get(&#39;https://news.cnblogs.com/n/%s/&#39; % next_id)
            res_next.raise_for_status()
            soup_next = bs4.BeautifulSoup(res_next.text, &#39;html.parser&#39;)
            curr_id = get_info(soup_next)

            private_loop_count += 1
    except:
        pass


# 参数从优先从命令行获取，如果无，则从剪切板获取
# url是博客园新闻版块下，任何一篇新闻
if len(sys.argv) > 1:
    url = sys.argv[1]
else:
    url = pyperclip.paste()

# 没有获取到有地址，则抛出异常
if not url:
    raise ValueError

# 开始从源地址中获取新闻内容
res = requests.get(url)
res.raise_for_status()
if not res.text:
    raise ValueError

#解析Html
soup = bs4.BeautifulSoup(res.text, &#39;html.parser&#39;)
curr_id = get_info(soup)
print(&#39;backward...&#39;)
get_prev_info(curr_id)
print(&#39;forward...&#39;)
get_next_info(curr_id)
print(&#39;done&#39;)

Run

Save the above source code to D:/get_cnblogs_news.py, under the windows platform Open the command line tool cmd:

Enter the command: py.exe D:/get_cnblogs_news.py Enter

Analysis: No need to explain py.exe, the second parameter is the python script file , the third parameter is the source page that needs to be crawled (there is another consideration in the code. If you copy this url to the system clipboard, you can run it directly: py.exe D:/get_cnblogs_news.py

　Command line output interface (print)

　

　Content saved to csv file

　

Recommended python learning bookbox or materials for rookies:

1) Liao Xuefeng’s Python tutorial, very basic and easy to understand:

2 ) Get started with Python programming quickly and automate tedious work.pdf

　

The article is just a diary for myself to learn python. Please criticize and correct me if it is misleading (no Please don’t spray), I would be honored if it helped you.

　

The above is the detailed content of python learning to capture blog park news. For more information, please follow other related articles on the PHP Chinese website!