I have a service that will upload files related to some products to an S3 bucket. In order to keep the bucket organized, each files related to a specific product will have its own subfolder in the bucket (named after the prod. #). Theorically those should be pretty standard, by since that prod # is user-created, I can’t be 100% certain it won’t contain weird characters that do not fit within the bucket naming restrictions.
Do those restrictions apply to subfolders in the bucket as well? I can’t seem to find a good source for that info. THanks.
>Solution :
The world-wide uniqueness of bucket names don’t apply to folders, you can have identical folder names across multiple buckets:
This means that the following combinations are legal/valid:
s3://foo-nfts/foo
s3://foo-eths/foo
s3://foo-btcs/foo
s3://foo-btcs/foo/foo/foo
The concept of folder don’t also exists in S3. The prefixes are just used by AWS console to give us the perceived foldering structure. The rules and restrictions of S3 object taxonomy are defined here