Datasets
Blockchain tables over S3, queried with your own project credentials.
Coming soon
Iceberg catalog
An Apache Iceberg REST catalog will let DuckDB, Spark, Trino and PyIceberg discover the Datasets tables and their current snapshots. Your query engine will read the Parquet files from S3 using your own Pinax Project ID and API key.
Discover tables
Browse namespaces and table definitions through the REST catalog.
Follow the snapshot
Use the current table metadata so files replaced by compaction are excluded.
Read with your project
Catalog metadata is separate from S3 data access, which uses your project credentials.
The public catalog and its tables are not available yet. Until then, read through the Delta log with DuckDB delta_scan or Polars scan_delta. See SQL & Connect for working examples, and Schemas for table definitions.