python 利用dict去重对比csv文件差异

adict=[]
need_find_list
for x in alist:
    adict[x[0]]=x # 列中每一行作为key值，dict自带去重功能，后面覆盖前面的重复值
    need_find_list.append(x[0]) # 加入list中作为key为后面提供取值查询对比
bdict也是一样，就不写了

得到了需要的两个dict 和一个查询的list后循环list在每一项循环中对dict进行get取值操作，然后是对比操作
像这样

for index,xx in enumerate(set(need_find_list)):
    keya=a.get(xx)
    keyb=b.get(xx)
    # 这里就是一些对比操作了，
对比原理挺简单的，速度也还不错，比手动用excel对比快多了

demo：

# coding:utf-8
#当前系统日期时间：2021/4/15 9:28
#用于创建文件的IDE的名称: PyCharm
import time
import pandas
start=time.time()
pd=pandas.read_csv('./new.csv',engine='python',encoding='utf-8')
ss=pd.drop_duplicates(keep='first',inplace=False)
pd1=pandas.read_csv('./old.csv',engine='python',encoding='utf-8')
ss1=pd1.drop_duplicates(keep='first',inplace=False)
a={}
''' {'pn':['pn','vender','stock']}''' # 数据格式
alist=[]
for x in ss.values:
    # print(x)
    a[x[0]]=x
    alist.append(x[0])


b={}
blist=[]
for x in ss1.values:
    b[x[0]]=x
    blist.append(x[0])

stock_unequal_count=0
stock_equal_count=0
txt=open('./out.txt','a+',encoding='utf-8')
stock_unequal=[] # stock 不相等
pn_unfind=[]
stock_equal=[]
for index,xx in enumerate(set(blist)):
    print(index)
    keya=a.get(xx)
    keyb=b.get(xx)
    try:
        if keya is not None and keyb is not None:
            if keya[2]!=keyb[2]:
                if type(keya[2]) != type(keyb[2]):
                    if int(str(keya[2]).replace(',',''))!=int(str(keyb[2]).replace(',','')):
                        print('find unequal','a=> ', keya, 'b=> ', keyb)
                        stock_unequal.append([keya[0], keya[2], keyb[2]])
                        stock_unequal_count += 1
                    else:
                        stock_equal.append([keya[0], keya[2], keyb[2]])
                        stock_equal_count += 1
                        print('值不相等')
                        # cha.append([keya[0], keya[2], keyb[2]])
                else:
                    print('类型不相等')
            else:
                print('数据不相等')
 
        else:
            pn_unfind.append(xx)
       
    except Exception as e:
        print(e)
print('a', len(a))
print('b',len(b))
print('不相等数据',stock_unequal_count)
print('相同的数据',stock_equal_count)
print('未找到',len(pn_unfind))

print(time.time()-start)