###### tags: `web crawler`,`網路爬蟲`
# web crawler 實作--查找自由時報即時新聞(結合Tkinter)

**先來張成果圖(雖然版面有點陽春,但功能也還行!)
好吧!看在阿湯哥這麼帥,今天版面就決定是他了!**


import tkinter as tk
from tkinter import messagebox
import requests
from bs4 import BeautifulSoup 
import time
import csv
import json

def _hit1(): #蒐集即時新聞
    global a1,a2,a3,news
    urL="https://news.ltn.com.tw/list/breakingnews" #自由時報電子新聞
    rq=requests.get(urL).text #發出請求
    soup=BeautifulSoup(rq,"html5lib") #BeautifulSoup解析網頁
    
    for rows in soup.find_all("li"): #找到所有html裡面的li
        try:
            a1=(rows.span.text.strip())  #strip()把其餘空白丟掉
            a2=(rows.a["title"].strip()) #列出新聞title
            a3=(rows.a["href"].strip()) #列出新聞連結
            a4=("-"*30)
            news=[a1,a2,a3,a4]#把資訊都塞進list裡面
            for i in news:
                listbox.insert(tk.END,[i])   #在listbox一條條列出資訊
        except:
            continue  

def _hit2(): #清除功能
    listbox.delete(0,tk.END) #清空List資料

def _hit3(): #儲存csv檔功能
    global day #用日期記錄檔名
    csvfile=open(day+"news.csv","w",newline="",encoding="utf-8-sig")
    writercsv=csv.writer(csvfile)
    urL="https://news.ltn.com.tw/list/breakingnews"
    rq=requests.get(urL).text
    soup=BeautifulSoup(rq,"html5lib")
    writercsv.writerow(["時間","標題","超連結"])#標題
    for rows in soup.find_all("li"):
        try:
            a1=(rows.span.text.strip())
            a2=(rows.a["title"].strip())
            a3=(rows.a["href"].strip())
            #writeR.writerow([mySoup.span.text.strip(),mySoup.a["title"].strip(),mySoup.a["href"].strip()])
            writercsv.writerow([a1,a2,a3])
        except:
            continue
    csvfile.close()#要記得關檔
    
def _hit4(): #儲存json檔功能
    global day,a1,a2,a3
    urL="https://news.ltn.com.tw/list/breakingnews"
    rq=requests.get(urL).text
    soup=BeautifulSoup(rq,"html5lib")
    news1=[]
    for rows in soup.find_all("li"):
        try:
            a1=(rows.span.text.strip())
            a2=(rows.a["title"].strip())
            a3=(rows.a["href"].strip())
            news1.append([a1,a2,a3])
            
        except:
            continue 
    with open(day+"news.json","w",encoding="utf-8") as jsonFile1:
        json.dump(news1,jsonFile1,ensure_ascii=False,indent=4)
    
def _hit5(): #離開功能
     que=tk.messagebox.askokcancel("Hint","確定離開???")
     if que:
         win1.destroy()


win1 = tk.Tk()
win1.title("自由時報即時新聞!!!")
win1.geometry("700x500")

tt=time.localtime()
day=str(tt.tm_year)+"-"+str(tt.tm_mon)+"-"+str(tt.tm_mday)

#Tkinter button ,每個bt要對應一個pack()
bt1 = tk.Button(win1, text="收集即時新聞",fg="Ivory",bg="OliveDrab", font=("Arial", 13), width=20, height=1, command=_hit1)
bt1.pack() 
bt2 = tk.Button(win1, text="清除",bg="Khaki", font=("Arial", 13), width=20, height=1, command=_hit2)
bt2.pack() 
bt3 = tk.Button(win1, text="儲存為csv檔",fg="Ivory",bg="OliveDrab", font=("Arial", 13), width=20, height=1, command=_hit3)
bt3.pack() 
bt4 = tk.Button(win1, text="儲存為json檔",bg="Khaki", font=("Arial", 13), width=20, height=1, command=_hit4)
bt4.pack()
bt5 = tk.Button(win1, text="離開",fg="Ivory",bg="OliveDrab", font=("Arial", 13), width=20, height=1, command=_hit5)
bt5.pack() 

#Tkinter 卷軸
s_bar=tk.Scrollbar(win1)
s_bar.pack(side=tk.RIGHT,fill=tk.Y)

listbox=tk.Listbox(win1, width=40, height=15,bg="Khaki",font=("Arial", 14),yscrollcommand=s_bar.set)
listbox.pack(side=tk.BOTTOM, fill=tk.BOTH)
s_bar.config(command=listbox.yview)

win1.mainloop()