Delta Lake

Delta Lake tables are the storage layer under most lakehouse platforms. This source reads them where they live — as files in a bucket — so you can scan a warehouse path without a Databricks workspace, a Spark cluster, or a catalog service in front of it.

What you need to connect

A bucket and, unless the scanning environment already has ambient cloud credentials, an access key pair. For MinIO, Cloudflare R2, Backblaze B2 and other S3-compatible providers, add the custom endpoint URL and region.

Then either give a prefix and let table discovery find every Delta table under it (recognised by its _delta_log directory), or list exact table paths when you know them.

What Classifyre reads

One asset per table. Its metadata is the table’s real structure — columns and types, row count, current version — and its content is a sample of rows drawn through the sampling strategy you chose.

Metadata on every asset

Lineage

Worth knowing

• Reads are non-destructive and version-aware — the table’s current snapshot is read; nothing is compacted, vacuumed or written.
• Discovery costs list calls. On a very large bucket, a narrower prefix or an explicit table list is much faster than scanning from the root.

Configuration

Beyond the fields below, every source also has the settings shared by all of them: the sampling strategy, the detectors to run, the scan schedule, and the compute limits for its scan jobs.

Required

Without these, the source will not save.

Secrets

Stored encrypted and never shown again after you save them. See Configuration & Fields.

Optional

Everything you can tune. Sensible defaults apply when you leave them alone.