@[toc]
1. MySQL與python庫(kù)準(zhǔn)備
這里推薦使用一款A(yù)pache Web服務(wù)器、PHP解釋器以及MySQL數(shù)據(jù)庫(kù)的整合軟件包——WampServer,其會(huì)自動(dòng)將一些設(shè)置配置好,無(wú)需配置環(huán)境變量,而且能夠使用MySQL的數(shù)據(jù)庫(kù)管理平臺(tái)phpMyAdmin,安裝過(guò)程很簡(jiǎn)單,大家可以到下面的鏈接地址下載相應(yīng)的軟件即可。
WampServer下載地址https://sourceforge.net/projects/wampserver/files/WampServer%203/WampServer%203.0.0/
為了實(shí)現(xiàn)Python和MySQL數(shù)據(jù)庫(kù)的交互,需要安裝PyMySQL庫(kù)。常規(guī)的安裝方法,pip install pymysql
2. 用python連接數(shù)據(jù)庫(kù)
通過(guò)以下代碼就能實(shí)現(xiàn)數(shù)據(jù)庫(kù)連接
import pymysql
db = pymysql.connect(host='localhost', port=3308, user='root', password='',
database='spider', charset='utf8')
# 其中spider為已創(chuàng)建的數(shù)據(jù)庫(kù)名稱
3. 用python存儲(chǔ)數(shù)據(jù)到數(shù)據(jù)庫(kù)
在python中模擬執(zhí)行SQL語(yǔ)句需要首先引入一個(gè)會(huì)話指針cursor,只有這樣才能調(diào)用SQL語(yǔ)句。之后編寫(xiě)SQL語(yǔ)句,并通過(guò)cursor的execute函數(shù)執(zhí)行SQL語(yǔ)句。之后需要通過(guò)db.commit()函數(shù)來(lái)更新數(shù)據(jù)表,最后關(guān)閉之前引入的會(huì)話指針cur和數(shù)據(jù)庫(kù)連接。
cur = db.cursor() # 獲取會(huì)話指針,用來(lái)調(diào)用SQL語(yǔ)句
sql = 'INSERT INTO test(company, title, href, date, source) VALUES (%s, %s, %s, %s, %s)'
cur.execute(sql, (company, title, href, date, source))
db.commit()
cur.close()
db.close()
上面的第三行代碼中,第一個(gè)參數(shù)為SQL語(yǔ)句,第二個(gè)參數(shù)用來(lái)把具體的內(nèi)容傳遞到各個(gè)%s的位置上。插入語(yǔ)句的整體代碼如下所示:
# 預(yù)定義參數(shù)
company = '阿里巴巴'
title = '測(cè)試標(biāo)題'
href = '測(cè)試鏈接'
source = '測(cè)試來(lái)源'
date = '測(cè)試時(shí)間'
db = pymysql.connect(host='localhost', port=3308, user='root', password='',
database='spider', charset='utf8')
cur = db.cursor() # 獲取會(huì)話指針,用來(lái)調(diào)用SQL語(yǔ)句
sql = 'INSERT INTO test(company, title, href, date, source) VALUES (%s, %s, %s, %s, %s)'
cur.execute(sql, (company, title, href, date, source))
db.commit()
cur.close()
db.close()
4. 用python在數(shù)據(jù)庫(kù)中查找并提取數(shù)據(jù)
這里的操作代碼與數(shù)據(jù)插入的代碼類似。
db = pymysql.connect(host='localhost', port=3308, user='root', password='',
database='spider', charset='utf8')
cur = db.cursor() # 獲取會(huì)話指針,用來(lái)調(diào)用SQL語(yǔ)句
sql = 'SELECT * FROM `test` WHERE company = %s'
cur.execute(sql, company)
data = cur.fetchall()
print(data)
db.commit()
cur.close()
db.close()
這里的db.commit()操作可以省略,因?yàn)椴簧婕皵?shù)據(jù)更新的操作。數(shù)據(jù)查詢中最重要的操作是cur.fetchall(),提取所有數(shù)據(jù),返回的結(jié)果為元組形式。
5. 用python從數(shù)據(jù)庫(kù)中刪除數(shù)據(jù)
操作類似,不再贅述。
db = pymysql.connect(host='localhost', port=3308, user='root', password='',
database='spider', charset='utf8')
cur = db.cursor() # 獲取會(huì)話指針,用來(lái)調(diào)用SQL語(yǔ)句
sql = 'DELETE FROM `test` WHERE company = %s'
cur.execute(sql, company)
db.commit()
cur.close()
db.close()
6. 把金融數(shù)據(jù)存入數(shù)據(jù)庫(kù)
這里結(jié)合本章內(nèi)容和之前的內(nèi)容(python金融大數(shù)據(jù)挖掘與分析(二)——新聞數(shù)據(jù)挖掘),實(shí)現(xiàn)將網(wǎng)絡(luò)獲取的新聞?lì)A(yù)料內(nèi)容存入數(shù)據(jù)庫(kù)。
數(shù)據(jù)庫(kù)操作類:
"""
作者:Aidan
時(shí)間:01/02/2020
功能:python與mysql數(shù)據(jù)庫(kù)交互
"""
import pymysql
class database_execute(object):
"""
數(shù)據(jù)庫(kù)操作類
"""
def __init__(self, db_name):
self.db_name = db_name
def connect_db(self):
db = pymysql.connect(host='localhost', port=3308, user='root', password='',
database=self.db_name, charset='utf8')
return db
def insert_db(self, company, title, href, date, source):
db = self.connect_db()
cur = db.cursor()
sql = 'INSERT INTO test(company, title, href, date, source) VALUES (%s, %s, %s, %s, %s)'
cur.execute(sql, (company, title, href, date, source))
db.commit()
cur.close()
db.close()
百度新聞爬蟲(chóng)函數(shù)
"""
作者:Aidan
時(shí)間:01/02/2020
功能:提取百度新聞標(biāo)題、網(wǎng)址、日期和來(lái)源
新增功能,批量獲取多家公司的百度新聞并生成數(shù)據(jù)報(bào)告
"""
import requests
import re
def baidu_news(company):
"""
獲取網(wǎng)頁(yè)源碼,并提取百度新聞標(biāo)題、網(wǎng)址、日期和來(lái)源
:param company: 公司名稱
:return: 網(wǎng)頁(yè)源碼
"""
url = 'https://www.baidu.com/s?rtt=1&bsst=1&cl=2&tn=news&word=' + company
# 百度新聞網(wǎng)站只認(rèn)可瀏覽器發(fā)送的請(qǐng)求,所以需要設(shè)置headers參數(shù),
# 以模擬瀏覽器的發(fā)送請(qǐng)求,chrome瀏覽器可以通過(guò)about:version獲取
headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
'AppleWebKit/537.36 (KHTML, like Gecko) '
'Chrome/77.0.3865.120 Safari/537.36'}
res = requests.get(url, headers=headers)
web_text = res.text
# 獲取新聞的來(lái)源和日期
pattern = '<p class="c-author">(.*?)</p>'
info = re.findall(pattern, web_text, re.S) # re.S用于考慮換行符,因?yàn)?和*不包含換行符
# print(info)
# 獲取新聞的網(wǎng)址和標(biāo)題
pattern_herf = '<h3 class="c-title">.*?<a href="(.*?)"'
href = re.findall(pattern_herf, web_text, re.S)
# print(href)
pattern_title = '<h3 class="c-title">.*?>(.*?)</a>'
title = re.findall(pattern_title, web_text, re.S)
# print(title)
# title 數(shù)據(jù)清洗
for i in range(len(title)):
title[i] = title[i].strip()
title[i] = re.sub('<.*?>', '', title[i])
# print(title)
# 新聞來(lái)源和日期清洗
source = []
date = []
for i in range(len(info)):
info[i] = re.sub('<.*?>', '', info[i]) # 清洗<img>標(biāo)簽信息
source.append(info[i].split(' ')[0]) # 將新聞來(lái)源和日期分開(kāi)
date.append(info[i].split(' ')[1])
source[i] = source[i].strip()
date[i] = date[i].strip()
return title, href, date, source
主函數(shù)
import spider
import python_database
def main():
companys = ['華能信托', '騰訊', '阿里巴巴']
title = []
href = []
date = []
source = []
db_connect = python_database.database_execute('spider')
for company in companys:
try:
title, href, date, source = spider.baidu_news(company)
for i in range(len(title)):
db_connect.insert_db(company, title[i], href[i], date[i], source[i])
print(company + '百度新聞爬取成功!')
except:
print(company + '百度新聞爬取失敗!')
if __name__ == '__main__':
main()
執(zhí)行結(jié)果如下: