data.read_collection

Fetch multiple objects matching a URI prefix or simple glob (single * in the trailing segment). Backend is inferred from the scheme. Use max_items to guard against accidentally listing millions of keys.

Kind: standard
Network: yes

Inputs🔗

Name	Type	Description
`uri`	string	URI prefix or glob (e.g. `s3://bucket/prefix/`, `file:///abs/path/.csv`). `*` is rejected explicitly because recursive listing is easy to misuse.
`format`	string	Optional format hint recorded for downstream reference.
`max_items`	number	Maximum number of items to fetch. `0` means unlimited. If the backend returns more than this, the block errors out so users must opt in to larger fetches.

Outputs🔗

Name	Type	Description
`files`	collection of file	Fetched objects, one file per listed key. Keys with slashes are flattened (slashes become underscores).

Example🔗

- id: 01890000-0000-0000-0000-000000000101
  name: data.read_collection
  args:
    uri: "s3://my-bucket/tiles/*.tif"
    max_items: 100