How to use AWS CLI to only copy files in S3 bucket that match a given string pattern

Question

I'm using the AWS CLI to copy files from an S3 bucket to my R machine using a command like below:

  system(
    "aws s3 cp s3://my_bucket_location/ ~/my_r_location/ --recursive --exclude '*' --include '*trans*' --region us-east-1"
    )

This works as expected, i.e. it copies all files in my_bucket_location that have "trans" in the filename at that location.

The problem that I am facing is that I have other files with similar naming conventions that I don't want to import in this step. As an example, in the list below I only want to copy the first two files, not the last two:

File list
trans_120215.csv
trans_130215.csv
sum_trans_120215.csv
sum_trans_130215.csv

If I was using regex I could make it more specific like "^trans_\\d+" to bring in just the first two files, but this doesn't seem possible using AWS CLI. So my question is there a way to have more complex pattern matching using AWS CLI like below?

  system(
    "aws s3 cp s3://my_bucket_location/ ~/my_r_location/ --recursive --exclude '*' --include '^trans_\\d+' --region us-east-1"
    )

Please note that I can only use information about the file in question, i.e. that I want to import a file with pattern "^trans_\\d+", I can't use the fact that the other unwanted files contain sum_ at the start, because this is only an example there could be other files with similar names like "check_trans_120215.csv".

I have considered other alternatives like below, but hoping there is a way to adjust the copy command to avoid going down either of these routes:

• Listing all items in the bucket > using regex in R to specify the files that I want > Only importing those files
• Keeping the copy command as it is > delete unwanted files on the R machine after the copy

Answer 1

The alternatives that you have listed are the best options because S3 CLI doesn't support regex.

Use of Exclude and Include Filters:

Currently, there is no support for the use of UNIX style wildcards in a command's path arguments. However, most commands have --exclude "" and --include "" parameters that can achieve the desired result. These parameters perform pattern matching to either exclude or include a particular file or object. The following pattern symbols are supported.

*: Matches everything
?: Matches any single character
[sequence]: Matches any character in sequence
[!sequence]: Matches any character not in sequence

Answer 2

Putting this here for others to find, since I just had to figure this out. Here's what I came up with:

s3cmd del $(s3cmd ls s3://[BUCKET]/ | grep '.*s3://[BUCKET]/[FILENAME]' | cut -c 41-)

You can put the regex in the grep search string. For instance, I was searching for specific files to delete (hence the s3cmd del). My regex looked like: '2016-11-04.*s3.*[DN][RS].*'. You may have to adjust the cut for your use. Should also work with s3cmd get.

Answer 3

here is the same solution for deletion, you can replace rm with cp you can do it using aws cli : https://aws.amazon.com/cli/ and some unix command.

this aws cli commands should work:

aws s3 rm s3://<your_bucket_name> --exclude "*" --include "<your_regex>"

if you want to include sub-folders you should add the flag --recursive

or with unix commands:

aws s3 ls s3://<your_bucket_name>/ | awk '{print $4}' | xargs -I%  <your_os_shell>   -c 'aws s3 rm s3:// <your_bucket_name>/% $1'

explanation:

1. list all files on the bucket --pipe-->
2. get the 4th parameter(its the file name) --pipe--> // you can replace it with linux command to match your pattern
3. run delete script with aws cli