Why does the "AWS S3 deploy pipe" always deploy all files to S3?

[Jussi Sarén]

In the description the AWS S3 deploy pipe promises to: "Recursively copies new and updated files from the source local directory to the destination."

However the pipe always copies all files from the VCS to the S3 bucket even if they are untouched. Is there something I'm doing wrong?

[davidstanton]

+1 on this , it ends up massively increasing your s3 put costs as well.  There really needs to be a way to only Sync "changed" files, unless anyone has found a way around it?

[Jussi Sarén]

I see that other people have been asking the same question. Could it be possible to create the changed file detection based on the GIT change log? GIT knows exactly which files have been changed and thus there would be an ability to improve the pipe to fulfill its promises.