Information marked in three ways, compare and extraction method

　　1, flag information

　　　　Flag information may be formed after the tissue structure information, dimension information increases

　　　　The tag information is used for communication, storage or display

　　　　Structure and information as the mark of great value

　　　　Information after the mark is more conducive to the understanding and application procedures

　　2, HTML tag information

　　　　HTML is the message of the organization www.

　　　　HTML is organized by different types of information <> .... </> tag predefined form

　　3, information flag three ways

　　　　1)、XML

　　　　If the tag has no content, then we can use angle brackets, he said:

　　　　Notes can also be embedded in content:

　　　　2)、JSON

　　　　　　There are types of keys on key: value

　　　　　　note:

　　　　　　　　Whether it is the key or value, if it is a string, we need to use double quotation marks to indicate, if the value is written directly on the line.

　　　　　　　　If the value has multiple values, we need to use [,] to represent.

　　　　　　　　Key-value pairs nested when we need {,} be represented as:

　　　　3) yaml

　　　　　　　No type of key-key: value, such as:

　　　　　　　While using indentation way to represent your relationship:

　　　　　　　By - expression parallel relationship:

　　　　　　With | the expression of entire blocks of data # denotes a comment

　　Example 4, three kinds of information marks

　　　　1), XML instance

　　　　2), JSON examples

　　　　3), YAML examples

　　Comparison 5, three kinds of information labeled form

　　　　The earliest general-purpose XML markup language information, scalability is good, but cumbersome

　　　　JSON typed information for program processing, compact than XML

　　　　YAML information untyped, the highest proportion of text messages and readable.

　　　　　　The information exchange and transmission on the Internet is in XML format (HTML also fall into this category)

　　　　　　JSON is used in the information communication nodes and mobile applications Drive, no comment,

　　· YAML is readable annotation system used in all types of configuration files, there is

　　6, general information extraction method

　　　　Method One: full marks in the form of analytical information, and then extract key information

　　　　即使用标记解析器去解析三种信息标记格式，然后将所需要的信息提取出来。如：bs4库中的标签树遍历

　　　　　　优点：信息解析准确

　　　　　　缺点：提取过程繁琐，速度慢

　　　　方法二：无视标记形式，直接搜索关键信息。

　　　　搜索：　

　　　　　　对信息的文本查找函数即可

　　　　　　优点：提取过程简洁，速度较快

　　　　　　缺点：提取结果准确性与信息内容相关

　　　　方法三：融合方法

　　　　　　结合形式解析与搜索方法，提取关键信息

　　　　　　　　需要标记解析器及文本查找函数。

　　　　　　实例：

　　　　　　　　提取HTML中所有URL链接

　　　　　　思路：

　　　　　　　　搜索到所有<a>标签

　　　　　　　　解析<a>标签格式，提取href后的链接内容

import requests
from bs4 import BeautifulSoup
#BeautifulSoup是一个类
r = requests.get('http://python123.io/ws/demo.html')

# print(r.text)
demo = r.text
#解析demo的解释器
soup = BeautifulSoup(demo,'html.parser')

for link in soup.find_all('a'):
    print(link.get('href'))

Information marked in three ways, compare and extraction method

Guess you like