Python-csv文件处理建议收藏 - 全栈程序员必看

Python–csv文件处理建议收藏

全栈程序员-用户IM • 2021年12月18日下午5:00 • 未分类

大家好，又见面了，我是全栈君，今天给大家准备了Idea注册码。

此处内容已经被作者隐藏，请输入验证码查看内容

请关注本站微信公众号，回复“”，获取验证码。在微信里搜索“”或者“”或者微信扫描右侧二维码都可以关注本站微信公众号。

CSV(Comma-Separator Values)逗号分割值，由于是纯文本文件，任何编辑器都可以打开。下面用csv和pandas两种方式进行csv文件操作　　

原始csv文件内容

Supplier Name,Invoice Number,Part Number,Cost,Purchase Date
Supplier X,001-1001,2341,$500.00 ,1/20/14
Supplier X,001-1001,2341,$500.00 ,1/20/14
Supplier X,001-1001,5467,$750.00 ,1/20/14
Supplier X,001-1001,5467,$750.00 ,1/20/14
Supplier Y,50-9501,7009,$250.00 ,1/30/14
Supplier Y,50-9501,7009,$250.00 ,1/30/14
Supplier Y,50-9505,6650,$125.00 ,2002/3/14
Supplier Y,50-9505,6650,$125.00 ,2002/3/14
Supplier Z,920-4803,3321,$615.00 ,2002/3/14
Supplier Z,920-4804,3321,$615.00 ,2002/10/14
Supplier Z,920-4805,3321,"$6,015.00 ",2/17/14
Supplier Z,920-4806,3321,"$1,006,015.00 ",2/24/14

1. csv包操作csv文件

#coding=utf-8

import sys
import csv
import re

read_file = sys.argv[1]
write_file = sys.argv[2]

with open(read_file, "r") as readfile:
    with open(write_file, "w") as writefile:
        reader = csv.reader(readfile, delimiter=",")
        writer = csv.writer(writefile, delimiter=",")
        header = next(reader)
        writer.writerow(header)
        for rowlist in reader:
            #通过正则表达是进行行匹配
            if re.match(r"^001-*.", str(rowlist[1])):
                print (rowlist)
                writer.writerow(rowlist)

>>> D:\Pystu>python parsecsvfile.py supplier_data.csv ceshi.csv
>>> Supplier Name,Invoice Number,Part Number,Cost,Purchase Date
>>> Supplier X,001-1001,2341,$500.00 ,1/20/14
>>> Supplier X,001-1001,2341,$500.00 ,1/20/14
>>> Supplier X,001-1001,5467,$750.00 ,1/20/14
>>> Supplier X,001-1001,5467,$750.00 ,1/20/14

2. pandas包操作csv文件

#coding=utf-8

''' 运用pandas包解析csv文件'''
import pandas
from pandas import Series,DataFrame
import sys

file_path = sys.argv[1]
write_path = sys.argv[2]
data_frame = pandas.read_csv(file_path)
#print (data_frame)

#注意str的使用
data_frame["Cost"] = data_frame["Cost"].str.replace(",", "").str.strip("$").astype(float)
#print (data_frame)

newa = data_frame.loc[data_frame["Cost"] > 600, :]
#print (newa)
newa.to_csv(write_path, index = False)

>>> D:\Pystu>python parse_csv_file_by_pandas.py supplier_data.csv ceshi.csv
>>> Supplier Name,Invoice Number,Part Number,Cost,Purchase Date
>>> Supplier X,001-1001,5467,750.0,1/20/14
>>> Supplier X,001-1001,5467,750.0,1/20/14
>>> Supplier Z,920-4803,3321,615.0,2002/3/14
>>> Supplier Z,920-4804,3321,615.0,2002/10/14
>>> Supplier Z,920-4805,3321,6015.0,2/17/14
>>> Supplier Z,920-4806,3321,1006015.0,2/24/14

发布者：全栈程序员-用户IM，转载请注明出处：https://javaforall.cn/120209.html原文链接：https://javaforall.cn

【正版授权，激活自己账号】： Jetbrains全家桶Ide使用，1年售后保障，每天仅需1毛

【官方授权正版激活】： 官方授权正版激活支持Jetbrains家族下所有IDE 使用个人JB账号...