check if a key exists in a bucket in s3 using boto3

I’m not a big fan of using exceptions for control flow. This is an alternative approach that works in boto3:

import boto3

s3 = boto3.resource('s3')
bucket = s3.Bucket('my-bucket')
key = 'dootdoot.jpg'
objs = list(bucket.objects.filter(Prefix=key))
keys = set(o.key for i in objs)
if path_s3 in keys:
    print("Exists!")
else:
    print("Doesn't exist")

Check out

bucket.get_key(
    key_name, 
    headers=None, 
    version_id=None, 
    response_headers=None, 
    validate=True
)

Check to see if a particular key exists within the bucket. This method
uses a HEAD request to check for the existence of the key. Returns: An
instance of a Key object or None

from Boto S3 Docs

You can just call bucket.get_key(keyname) and check if the returned object is None.

In Boto3, if you’re checking for either a folder (prefix) or a file using list_objects. You can use the existence of ‘Contents’ in the response dict as a check for whether the object exists. It’s another way to avoid the try/except catches as @EvilPuppetMaster suggests

import boto3
client = boto3.client('s3')
results = client.list_objects(Bucket='my-bucket', Prefix='dootdoot.jpg')
return 'Contents' in results

The easiest way I found (and probably the most efficient) is this:

import boto3
from botocore.errorfactory import ClientError

s3 = boto3.client('s3')
try:
    s3.head_object(Bucket='bucket_name', Key='file_path')
except botocore.exceptions.ClientError as e:
    if e.response['Error']['Code'] == "404":
        # The key does not exist.
        ...
    elif e.response['Error']['Code'] == 403:
        # Unauthorized, including invalid bucket
        ...
    else:
      # Something else has gone wrong.
      raise e

Not only client but bucket too:

import boto3
import botocore
bucket = boto3.resource('s3', region_name='eu-west-1').Bucket('my-bucket')

try:
  bucket.Object('my-file').get()
except botocore.exceptions.ClientError as ex:
  if ex.response['Error']['Code'] == 'NoSuchKey':
    print('NoSuchKey')

If you have less than 1000 in a directory or bucket you can get set of them and after check if such key in this set:

files_in_dir = {d['Key'].split('/')[-1] for d in s3_client.list_objects_v2(
Bucket='mybucket',
Prefix='my/dir').get('Contents') or []}

Such code works even if my/dir is not exists.

http://boto3.readthedocs.io/en/latest/reference/services/s3.html#S3.Client.list_objects_v2

S3_REGION="eu-central-1"
bucket="mybucket1"
name="objectname"

import boto3
from botocore.client import Config
client = boto3.client('s3',region_name=S3_REGION,config=Config(signature_version='s3v4'))
list = client.list_objects_v2(Bucket=bucket,Prefix=name)
for obj in list.get('Contents', []):
    if obj['Key'] == name: return True
return False

There is one simple way by which we can check if file exists or not in S3 bucket. We donot need to use exception for this

sesssion = boto3.Session(aws_access_key_id, aws_secret_access_key)
s3 = session.client('s3')

object_name = 'filename'
bucket = 'bucketname'
obj_status = s3.list_objects(Bucket = bucket, Prefix = object_name)
if obj_status.get('Contents'):
    print("File exists")
else:
    print("File does not exists")

import boto3
client = boto3.client('s3')
s3_key = 'Your file without bucket name e.g. abc/bcd.txt'
bucket = 'your bucket name'
content = client.head_object(Bucket=bucket,Key=s3_key)
    if content.get('ResponseMetadata',None) is not None:
        print "File exists - s3://%s/%s " %(bucket,s3_key) 
    else:
        print "File does not exist - s3://%s/%s " %(bucket,s3_key)

FWIW, here are the very simple functions that I am using

import boto3

def get_resource(config: dict={}):
    """Loads the s3 resource.

    Expects AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY to be in the environment
    or in a config dictionary.
    Looks in the environment first."""

    s3 = boto3.resource('s3',
                        aws_access_key_id=os.environ.get(
                            "AWS_ACCESS_KEY_ID", config.get("AWS_ACCESS_KEY_ID")),
                        aws_secret_access_key=os.environ.get("AWS_SECRET_ACCESS_KEY", config.get("AWS_SECRET_ACCESS_KEY")))
    return s3


def get_bucket(s3, s3_uri: str):
    """Get the bucket from the resource.
    A thin wrapper, use with caution.

    Example usage:

    >> bucket = get_bucket(get_resource(), s3_uri_prod)"""
    return s3.Bucket(s3_uri)


def isfile_s3(bucket, key: str) -> bool:
    """Returns T/F whether the file exists."""
    objs = list(bucket.objects.filter(Prefix=key))
    return len(objs) == 1 and objs[0].key == key


def isdir_s3(bucket, key: str) -> bool:
    """Returns T/F whether the directory exists."""
    objs = list(bucket.objects.filter(Prefix=key))
    return len(objs) > 1

For boto3, ObjectSummary can be used to check if an object exists.

Contains the summary of an object stored in an Amazon S3 bucket. This object doesn’t contain contain the object’s full metadata or any of its contents

import boto3
from botocore.errorfactory import ClientError
def path_exists(path, bucket_name):
    """Check to see if an object exists on S3"""
    s3 = boto3.resource('s3')
    try:
        s3.ObjectSummary(bucket_name=bucket_name, key=path).load()
    except ClientError as e:
        if e.response['Error']['Code'] == "404":
            return False
        else:
            raise e
    return True

path_exists('path/to/file.html')

In ObjectSummary.load

Calls s3.Client.head_object to update the attributes of the ObjectSummary resource.

This shows that you can use ObjectSummary instead of Object if you are planning on not using get(). The load() function does not retrieve the object it only obtains the summary.

If you seek a key that is equivalent to a directory then you might want this approach

session = boto3.session.Session()
resource = session.resource("s3")
bucket = resource.Bucket('mybucket')

key = 'dir-like-or-file-like-key'
objects = [o for o in bucket.objects.filter(Prefix=key).limit(1)]    
has_key = len(objects) > 0

This works for a parent key or a key that equates to file or a key that does not exist. I tried the favored approach above and failed on parent keys.

I noticed that just for catching the exception using botocore.exceptions.ClientError we need to install botocore. botocore takes up 36M of disk space. This is particularly impacting if we use aws lambda functions. In place of that if we just use exception then we can skip using the extra library!

• I am validating for the file extension to be ‘.csv’
• This will not throw an exception if the bucket does not exist!
• This will not throw an exception if the bucket exists but object does not exist!
• This throws out an exception if the bucket is empty!
• This throws out an exception if the bucket has no permissions!

The code looks like this. Please share your thoughts:

import boto3
import traceback

def download4mS3(s3bucket, s3Path, localPath):
    s3 = boto3.resource('s3')

    print('Looking for the csv data file ending with .csv in bucket: ' + s3bucket + ' path: ' + s3Path)
    if s3Path.endswith('.csv') and s3Path != '':
        try:
            s3.Bucket(s3bucket).download_file(s3Path, localPath)
        except Exception as e:
            print(e)
            print(traceback.format_exc())
            if e.response['Error']['Code'] == "404":
                print("Downloading the file from: [", s3Path, "] failed")
                exit(12)
            else:
                raise
        print("Downloading the file from: [", s3Path, "] succeeded")
    else:
        print("csv file not found in in : [", s3Path, "]")
        exit(12)

Try This simple

import boto3
s3 = boto3.resource('s3')
bucket = s3.Bucket('mybucket_name') # just Bucket name
file_name = 'A/B/filename.txt'      # full file path
obj = list(bucket.objects.filter(Prefix=file_name))
if len(obj) > 0:
    print("Exists")
else:
    print("Not Exists")

Here is a solution that works for me. One caveat is that I know the exact format of the key ahead of time, so I am only listing the single file

import boto3

# The s3 base class to interact with S3
class S3(object):
  def __init__(self):
    self.s3_client = boto3.client('s3')

  def check_if_object_exists(self, s3_bucket, s3_key):
    response = self.s3_client.list_objects(
      Bucket = s3_bucket,
      Prefix = s3_key
      )
    if 'ETag' in str(response):
      return True
    else:
      return False

if __name__ == '__main__':
  s3  = S3()
  if s3.check_if_object_exists(bucket, key):
    print "Found S3 object."
  else:
    print "No object found."

You can use S3Fs, which is essentially a wrapper around boto3 that exposes typical file-system style operations:

import s3fs
s3 = s3fs.S3FileSystem()
s3.exists('myfile.txt')

you can use Boto3 for this.

import boto3
s3 = boto3.resource('s3')
bucket = s3.Bucket('my-bucket')
objs = list(bucket.objects.filter(Prefix=key))
if(len(objs)>0):
    print("key exists!!")
else:
    print("key doesn't exist!")

Here key is the path you want to check exists or not

It’s really simple with get() method

import botocore
from boto3.session import Session
session = Session(aws_access_key_id='AWS_ACCESS_KEY',
                aws_secret_access_key='AWS_SECRET_ACCESS_KEY')
s3 = session.resource('s3')
bucket_s3 = s3.Bucket('bucket_name')

def not_exist(file_key):
    try:
        file_details = bucket_s3.Object(file_key).get()
        # print(file_details) # This line prints the file details
        return False
    except botocore.exceptions.ClientError as e:
        if e.response['Error']['Code'] == "NoSuchKey": # or you can check with e.reponse['HTTPStatusCode'] == '404'
            return True
        return False # For any other error it's hard to determine whether it exists or not. so based on the requirement feel free to change it to True/ False / raise Exception

print(not_exist('hello_world.txt')) 

Just following the thread, can someone conclude which one is the most efficient way to check if an object exists in S3?

I think head_object might win as it just checks the metadata which is lighter than the actual object itself

Assuming you just want to check if a key exists (instead of quietly over-writing it), do this check first. Will also check for errors:

import boto3

def key_exists(mykey, mybucket):
    s3_client = boto3.client('s3')
    try:
        response = s3_client.list_objects_v2(Bucket=mybucket, Prefix=mykey)
        for obj in response['Contents']:
            if mykey == obj['Key']:
                return 'exists'
        return False  # no keys match
    except KeyError:
        return False  # no keys found
    except Exception as e:
        # Handle or log other exceptions such as bucket doesn't exist
        return e

key_check = key_exists('someprefix/myfile-abc123', 'my-bucket-name')
if key_check:
    if key_check == 'exists':
        print("key exists!")
    else:
        print(f"S3 ERROR: {e}")
else:
    print("safe to put new bucket object")
    # try:
    #     resp = s3_client.put_object(Body="Your string or file-like object",
    #                                 Bucket=mybucket,Key=mykey)
    # ...check resp success and ClientError exception for errors...

This could check both prefix and key, and fetches at most 1 key.

def prefix_exits(bucket, prefix):
    s3_client = boto3.client('s3')
    res = s3_client.list_objects_v2(Bucket=bucket, Prefix=prefix, MaxKeys=1)
    return 'Contents' in res

Using objects.filter and checking the resultant list is the by far fastest way to check if a file exists in an S3 bucket. .

Use this concise oneliner, makes it less intrusive when you have to throw it inside an existing project without modifying much of the code.

s3_file_exists = lambda filename: bool(list(bucket.objects.filter(Prefix=filename)))

The above function assumes the bucket variable was already declared.

You can extend the lambda to support additional parameter like

s3_file_exists = lambda filename, bucket: bool(list(bucket.objects.filter(Prefix=filename)))

You can use awswrangler to do it in 1 line.

awswrangler.s3.does_object_exist(path_of_object_to_check)

https://aws-data-wrangler.readthedocs.io/en/stable/stubs/awswrangler.s3.does_object_exist.html

The does_object_exist method uses the head_object method of the s3 client and checks if there is a ClientError raised. If the error code is 404 than False is returned.

It’s 2023 and none of the above worked for me. Here is the version that did for me:

import boto3
import botocore
s3 = boto3.client('s3')
try:  
    s3.head_object(Bucket='YOUR_BUCKET_NAME', Key=object_name)
except botocore.exceptions.ClientError as error:
    if error.response['Error']['Code']:
        print("Object does not exist!")
    else:
        print("Object exists!")

If you do have permission and everything is right with the request, in case the object does not exist, you will get 404. The request also returns 400 and 403 so if you want to be more specific with the error handling you can check the error for those errors.