python数据分析——网络爬虫和API

HTML

什么是超文本标记语言？

它包含由标签标记的多层内容，包括开始标签和带有'/'的结束标签

"head"：用于浏览器特定信息

"style"：层叠样式表（CSS）用于设置HTML页面的样式

"body"：用于可见内容

html 复制代码

<html>
<head>
	<title>My First Web Page</title>
	<style>
		.highlight {
			background-color: lightblue;
		}
		div.section h1 {
			color: red;
			font-size: 24px;
		}
		#important {
			font-weight: bold;
		}
		</style>
</head>
<body>
	<h1>This is my first web page!</h1>
	<p>I'm excited to learn HTML.</p>
</html>

数据检索

使用requests.get()下载整个网页的全部内容
使用BeautifulSoup导航并提取精确信息（位于开始标签和结束标签之间）
逐步遍历网页的层次结构从而到达目标位置

python 复制代码

import requests
from bs4 import BeautifulSoup

# store the response from the website in a varieble
response = requests.get("https://en.wikipedia.org/wiki/...")
# extract the actual content of the web page
content = response.content

# create a BeautifulSoup object
# 'html.parser' is the default parser for BeautifulSoup
soup = BeautifulSoup(content, 'html.parser')

# get the head element of BeautifulSoup object
body = soup.head
# get the title element of the head element
t = body.title

print(t.text)

选取元素

find_all()

定位并提取网页中所有带有某一tag的元素

id: unique

python 复制代码

# find_all() with an id attribute passed
links = soup.find_all('li', id = "toc-Computing")

for link in links:
	href = link.get('href') # get the URL
	text = link.text
	print(f"URL: {href}\\nText: {text}\\n")

class_: not unique

python 复制代码

import requests
from bs4 import BeautifulSoup
import pandas as pd

url = "https://en.wikipedia.org/wiki/List_of_Nobel_laureates"

response = requests.get(url)

soup = BeautifulSoup(response.text, "html.parser")

# scrape a table from web page
# find an HTML table with the specified class_ attribute
table = soup.find("table", class_ = "wikitable sortable")

# read the HTML table into a DataFrame
# select the first DataFrame from the list
df = pd.read_htnl(str(table))[0]

CSS selectors

tag selector
.class selector
#ID selector

python 复制代码

# select all elements with the class "highlight"
highlighted_elements = soup.select(".highlight")
for element in highlighted_elements:
	print("Highlighted Element Text:", element.text)

# select <h2> elements inside a <div> with class "section"	
section_h2_elements = soup.select("div.section h2")
for element in section_h2_elements:
	print("Section <h2> Text:", element.text)
	
# select the element with the ID "important"
important_elements = soup.select("#important")
print("Important Element Text:", important_elements[0].text)

API

它是一种软件组件之间相互交互的方式

它可以用来从外部源（如数据库、Web服务和云存储）提取数据

获取一个API密钥来向API发送请求

端点（endpoint）

一个用于从API访问特定资源或功能的URL

Google Maps API

/geocode/json: get the latitude and longitude of a given address

/directions/json: get directions between two points

/places/nearby: get a list of places nearby a given location
GitHub API

/users/{username}/repos: get a list of repositories for a specified user

/repos/{owner}/{repo}/commits: get a list of commits for a repository

/repos/{owner}/{repo}/issues: get a list issues for a repository
OpenWeatherMap API

/weather: get current weather data for a specified location

/forecast/hourly: get hourly weather forecast for 4 days of a specified location

/history/city: get hourly historical weather data for specified location

向API发送请求

使用HTTP客户端：一个可以发送和接收HTTP请求的软件应用程序

requests.get(url)：向URL发送HTTP请求，并从API端点检索数据，其中URL作为参数传入

python 复制代码

import requests

url = 'https://api.openweathermap.org/data/2.5/weather?q=Singapore&APPID=YOUR_API_KEY'

response = requests.get(url)

# check whether the request is successful
# 200 indicates successful
# 400 indicates the request was invalid
# 500 indicates an error occureed on the surver
if response.status_code == 200:
	# convert response content into a dictionary or list
	weather_data = response.json() # output
	
	weather_description = weather_data['weather'][0]['description']
	print(weather_description)
else:
	print("An error occurred.")

JSON

JavaScript Object Notation

一种轻量级数据交换格式（无需额外标签）
基于文本，且与平台无关，在不同应用程序之间交换数据时非常流行
由对象（{}）和数组（[]）组成，以层次化的树状格式进行结构化

response data

python 复制代码

import json
import requests
import pandas as pd

url = 'https://api.openweathermap.org/data/2.5/weather?q=Singapore&APPID=YOUR_API_KEY'

response = requests.get(url)

# parse the JSON response data into a dictionary
weather_data = json.loads(response.content)

# convert JSON data into a DataFrame
df = pd.json_normalize(weather_data)
print(pd)

pass URL parameters

在?后添加URL参数
q=value: 特定的查询条件
appid=your_api_key: 传入你的API key
&: 不同参数用&分隔开

python 复制代码

import requests

URL = 'https://api.openweathermap.org/data/2.5/weather'

# parameters are stored in key-value pair structure
PARAMETERS = {
		"q": "Singapore",
		"appid": "your_api_key",
		"units": "imperial"
}

# params require a dictionary
response = requests.get(URL, params=PARAMETERS)

# parse the JSON response data into a dictionary
data = response.json()

# convert temperature form Kelvins to Celsius
temperature = data['main']['temp'] - 273.15

print(f"The current temperature in Singapore is {round(temperatire,2)} degrees Celsius.")