无法清除熊猫数据帧

Question

这是我第一次使用panda，我正在尝试制作一个forloop，从网站中提取mp3链接并将其放入csv文件中。对于每个相册链接，它将创建一个新文件夹和一个新的csv文件，然后将mp3放入csv中。

一切正常但我有一个主要问题-数据帧不断将上一个循环中的数据附加到当前循环中。所以我的数据帧的list越来越大。

以下是代码的样子：

from bs4 import BeautifulSoup
import urllib.request
import re
import pandas as pd
import os
import csv
def mpull():
albumlist()
baseurl = "https://downloads.khinsider.com"
alist = albumlist.albums_str
llist = albumlist.link_str
fullsoup = []
for l, ab in zip(llist, alist):
os.mkdir(ab)
url = urllib.request.urlopen(l)
content = url.read()
soup = BeautifulSoup(content, features="html.parser")
for a in soup.findAll('a',href=re.compile('/*.mp3')):
df = pd.DataFrame([])
fullsoup.append(baseurl+a['href'])
remove_dup(fullsoup)
df = pd.DataFrame(fullsoup)
df.to_csv(ab+"/"+ab+".csv", index=False, header=False) 
print(fullsoup)
mpull()

我想要的是：

007 everything or nothing:
https://downloads.khinsider.com/game-soundtracks/album/007-everything-or-nothing/EON-01-James-Bond-Theme.mp3
https://downloads.khinsider.com/game-soundtracks/album/007-everything-or-nothing/EON-02-Russian-Liar.mp3
#MORE 007 everything or nothing songs

我得到的是：

007 everything or nothing:
#songs from the last loop appear first for some reason
https://downloads.khinsider.com/game-soundtracks/album/007-blood-stone/01-%2520James%2520Bond-Blood%2520Stone%2520Theme%2520Song.mp3
https://downloads.khinsider.com/game-soundtracks/album/007-blood-stone/02-%2520M%2520Puts%2520Her%2520Trust%2520in%2520Bond.mp3
#Then the right songs appear afterwards 
https://downloads.khinsider.com/game-soundtracks/album/007-everything-or-nothing/EON-01-James-Bond-Theme.mp3
https://downloads.khinsider.com/game-soundtracks/album/007-everything-or-nothing/EON-02-Russian-Liar.mp3
#MORE 007 everything or nothing songs

我尝试过的：我尝试将del df添加到循环的末尾，如下所示：

from bs4 import BeautifulSoup
import urllib.request
import re
import pandas as pd
import os
import csv
def mpull():
albumlist()
baseurl = "https://downloads.khinsider.com"
alist = albumlist.albums_str
llist = albumlist.link_str
fullsoup = []
for l, ab in zip(llist, alist):
os.mkdir(ab)
url = urllib.request.urlopen(l)
content = url.read()
soup = BeautifulSoup(content, features="html.parser")
for a in soup.findAll('a',href=re.compile('/*.mp3')):
df = pd.DataFrame([])
fullsoup.append(baseurl+a['href'])
remove_dup(fullsoup)
df = pd.DataFrame(fullsoup)
df.to_csv(ab+"/"+ab+".csv", index=False, header=False) 
del def
print(fullsoup)
mpull()

但这似乎没有起到任何作用——它仍然将最后一个循环的数据帧附加到当前的csv迭代中。

任何想法都会很棒。谢谢

Answer 1

我想通了！！！而不是尝试删除数据帧df。我需要在每个循环迭代中删除fullsoup列表，这样它就不会在每个循环中保留列表的数据。

def mpull():
albumlist()
baseurl = "https://downloads.khinsider.com"
alist = albumlist.albums_str
llist = albumlist.link_str
for l, ab in zip(llist, alist):
fullsoup = []
os.mkdir(ab)
url = urllib.request.urlopen(l)
content = url.read()
soup = BeautifulSoup(content, features="html.parser")
for a in soup.findAll('a',href=re.compile('/*.mp3')):
df = pd.DataFrame([])
fullsoup.append(baseurl+a['href'])
# remove_dup(fullsoup)
df = pd.DataFrame(fullsoup)
print(fullsoup)
del fullsoup
df.to_csv(ab+"/"+ab+".csv", index=False, header=False) 
mpull()