python tarfile创建了一个额外的@PaxHeader文件，并导致错误无法utime：不允许操作

问题描述

我有一个lambda函数，用于解压缩tgz文件，替换内容和文件名，然后将其重新打包为新的tgz文件。如果我从s3下载新文件并在Windows上使用7z或在Ubuntu上使用tar xvzf <new_file_name>.tgz在本地解压缩，则一切正常。但是，如果我将新的tgz文件sftp到客户，他们会收到utime错误。他们向我发送了这样的日志

cd /usr/.../; tar xvzf </usr/.../new_file_name>.tgz;
./
./A/
./A/a.jpg
./A/b.jpg
./A/c.jpg
./something.xml
tar: .:Cannot utime: Operation not permitted.
tar: Exiting with failure status due to prevIoUs error

我想知道额外的@paxHeader文件是否引起utime错误，因为我从日志文件中可以看到内容已经解压缩。

这是我的lambda函数代码。

import os
import sys
import tarfile
import shutil
import boto3
from botocore.exceptions import ClientError

s3_client = boto3.client('s3')
accepted_date = os.environ["Acceptable_Date"]
work_dir = '/tmp'  # os.getcwd() returns "/var/task"
new_dir = '/tmp/new'
current_date = ''

def lambda_handler(event,context):
    for record in event['Records']:
        bucket = record['s3']['bucket']['name']
        key = record['s3']['object']['key']
        tmpkey = key.replace('/','')
        if accepted_date in tmpkey:
            print('Already has the correct date. No processing')
        else:
            download_path = '{}/{}'.format(work_dir,tmpkey)

            current_date = get_current_date(tmpkey)
            newkey = tmpkey.replace(current_date,accepted_date)

            upload_path = '{}/{}'.format(work_dir,newkey.replace('tgz','tar.gz'))
            print(work_dir)
            s3_client.download_file(bucket,key,download_path)
            process(upload_path)
            s3_client.upload_file(upload_path,bucket,newkey)

            cleanup(download_path)
            print('aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa')
            for entry in os.scandir(work_dir):
                print(entry.name)
            print('bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb') 

    
def process(upload_path):
    for entry in os.scandir(work_dir):
        if entry.path.endswith(".tgz") and entry.is_file:
            print(entry.path)
            current_date = get_current_date(entry.name)
            decompress(entry.path)
            new_name = replace_content(current_date)
            compress2(upload_path)      

def get_current_date(file_name):
    return file_name.split('_')[1]

def decompress(file_name):
    tar = tarfile.open(file_name,"r:gz")
    tar.extractall(new_dir)
    tar.close()

def compress2(output_filename):
    # if using tgz as file extension,you cannot decompress it using 7z on Windows.
    # Do not add entry by entry. Add the source fold instead.
    for entry in os.scandir(new_dir):
        print('Included: '+entry.path)
    with tarfile.open(output_filename,"w:gz") as tar:
        tar.add(new_dir,arcname='.')

def cleanup(output_filename):
    for entry in os.scandir(work_dir):
        if entry.is_file:
            remove(entry.path,False)
        else:
            remove(entry.path,True)
            
def replace_content(current_date): 
    for entry in os.scandir(new_dir):
        if entry.path.endswith(".XML") and entry.is_file():
            print(entry.path)
            new_path = entry.path.replace(current_date,accepted_date)
            with open(entry.path,"rt") as old:
                with open(new_path,"wt") as new:
                    for line in old:
                        new.write(line.replace(current_date,accepted_date))
            remove(entry.path,False)
            return new_path.replace('XML','tgz')

def remove(path,is_dir):
    if os.path.exists(path):
        if is_dir:
            shutil.rmtree(path)
        else:
            os.remove(path)
        print('Removed: '+path)
    else:
        print("The file does not exist: "+path)
 

知道我做错了什么吗？

PS 所有内容文件和tgz文件的文件许可权均为-rwxrwxrwx，如果我在本地下载并解压缩，则文件所有者为我。

解决方法

试试这个解决方案：

tar = tarfile.open('sample.tar.gz','w:gz',format=tarfile.GNU_FORMAT)