为了处理大型键列表(即当目录列表大于1000项时)，我使用以下代码将多个列表中的键值(即文件名)累积起来(感谢上面的阿梅里奥的第一行)。代码是针对python3的:

    from boto3  import client
    bucket_name = "my_bucket"
    prefix      = "my_key/sub_key/lots_o_files"

    s3_conn   = client('s3')  # type: BaseClient  ## again assumes boto.cfg setup, assume AWS S3
    s3_result =  s3_conn.list_objects_v2(Bucket=bucket_name, Prefix=prefix, Delimiter = "/")

    if 'Contents' not in s3_result:
        #print(s3_result)
        return []

    file_list = []
    for key in s3_result['Contents']:
        file_list.append(key['Key'])
    print(f"List count = {len(file_list)}")

    while s3_result['IsTruncated']:
        continuation_key = s3_result['NextContinuationToken']
        s3_result = s3_conn.list_objects_v2(Bucket=bucket_name, Prefix=prefix, Delimiter="/", ContinuationToken=continuation_key)
        for key in s3_result['Contents']:
            file_list.append(key['Key'])
        print(f"List count = {len(file_list)}")
    return file_list

使用cloudpathlib

cloudpathlib提供了一个方便的包装器，这样您就可以使用简单的pathlib API与AWS S3(以及Azure blob存储、GCS等)进行交互。你可以用pip install "cloudpathlib[s3]"来安装。

像pathlib一样，你可以使用glob或iterdir来列出目录的内容。

下面是一个带有公共AWS S3桶的示例，您可以复制并过去运行该桶。

from cloudpathlib import CloudPath

s3_path = CloudPath("s3://ladi/Images/FEMA_CAP/2020/70349")

# list items with glob
list(
    s3_path.glob("*")
)[:3]
#> [ S3Path('s3://ladi/Images/FEMA_CAP/2020/70349/DSC_0001_5a63d42e-27c6-448a-84f1-bfc632125b8e.jpg'),
#>   S3Path('s3://ladi/Images/FEMA_CAP/2020/70349/DSC_0002_a89f1b79-786f-4dac-9dcc-609fb1a977b1.jpg'),
#>   S3Path('s3://ladi/Images/FEMA_CAP/2020/70349/DSC_0003_02c30af6-911e-4e01-8c24-7644da2b8672.jpg')]

# list items with iterdir
list(
    s3_path.iterdir()
)[:3]
#> [ S3Path('s3://ladi/Images/FEMA_CAP/2020/70349/DSC_0001_5a63d42e-27c6-448a-84f1-bfc632125b8e.jpg'),
#>   S3Path('s3://ladi/Images/FEMA_CAP/2020/70349/DSC_0002_a89f1b79-786f-4dac-9dcc-609fb1a977b1.jpg'),
#>   S3Path('s3://ladi/Images/FEMA_CAP/2020/70349/DSC_0003_02c30af6-911e-4e01-8c24-7644da2b8672.jpg')]

创建于2021-05-21 20:38:47 PDT由reprexlite v0.4.2创建

如果你想传递ACCESS和SECRET密钥(你不应该这样做，因为这是不安全的):

from boto3.session import Session

ACCESS_KEY='your_access_key'
SECRET_KEY='your_secret_key'

session = Session(aws_access_key_id=ACCESS_KEY,
                  aws_secret_access_key=SECRET_KEY)
s3 = session.resource('s3')
your_bucket = s3.Bucket('your_bucket')

for s3_file in your_bucket.objects.all():
    print(s3_file.key)

下面是一个简单的函数，它返回所有文件的文件名或具有特定类型的文件，如'json'， 'jpg'。

def get_file_list_s3(bucket, prefix="", file_extension=None):
            """Return the list of all file paths (prefix + file name) with certain type or all
            Parameters
            ----------
            bucket: str
                The name of the bucket. For example, if your bucket is "s3://my_bucket" then it should be "my_bucket"
            prefix: str
                The full path to the the 'folder' of the files (objects). For example, if your files are in 
                s3://my_bucket/recipes/deserts then it should be "recipes/deserts". Default : ""
            file_extension: str
                The type of the files. If you want all, just leave it None. If you only want "json" files then it
                should be "json". Default: None       
            Return
            ------
            file_names: list
                The list of file names including the prefix
            """
            import boto3
            s3 = boto3.resource('s3')
            my_bucket = s3.Bucket(bucket)
            file_objs =  my_bucket.objects.filter(Prefix=prefix).all()
            file_names = [file_obj.key for file_obj in file_objs if file_extension is not None and file_obj.key.split(".")[-1] == file_extension]
            return file_names

从lambda函数运行aws cli命令也是一个不错的选择

import subprocess
import logging

logger = logging.getLogger()
logger.setLevel(logging.INFO)

def run_command(command):
    command_list = command.split(' ')

    try:
        logger.info("Running shell command: \"{}\"".format(command))
        result = subprocess.run(command_list, stdout=subprocess.PIPE);
        logger.info("Command output:\n---\n{}\n---".format(result.stdout.decode('UTF-8')))
    except Exception as e:
        logger.error("Exception: {}".format(e))
        return False

    return True

def lambda_handler(event, context):
    run_command('/opt/aws s3 ls s3://bucket-name')

我的s3键实用函数本质上是@Hephaestus的答案的优化版本:

import boto3


s3_paginator = boto3.client('s3').get_paginator('list_objects_v2')


def keys(bucket_name, prefix='/', delimiter='/', start_after=''):
    prefix = prefix.lstrip(delimiter)
    start_after = (start_after or prefix) if prefix.endswith(delimiter) else start_after
    for page in s3_paginator.paginate(Bucket=bucket_name, Prefix=prefix, StartAfter=start_after):
        for content in page.get('Contents', ()):
            yield content['Key']

在我的测试(boto3 1.9.84)中，它比等效的(但更简单)代码要快得多:

import boto3


def keys(bucket_name, prefix='/', delimiter='/'):
    prefix = prefix.lstrip(delimiter)
    bucket = boto3.resource('s3').Bucket(bucket_name)
    return (_.key for _ in bucket.objects.filter(Prefix=prefix))

由于S3保证UTF-8二进制排序结果，因此在第一个函数中添加了start_after优化。