The Geoparquet encoding defines how vector data compliant to Vecorel must be published. The generic GeoParquet format is defined in the OGC GeoParquet specification, either version v1.0.0 or v1.1.0. We aim to support any future version of GeoParquet, too.
Each Vecorel Feature corresponds to a row in a GeoParquet file.
The properties defined for Vecorel Features are made available as individual columns in the GeoParquet file.
Properties that are optional can be omitted if all values are null values, i.e. the column can be missing from the GeoParquet file.
Properties can also be stored at the collection-level if all values in a column have the same value.
This de-duplicates data for more efficient resource usage and simplifies the sturcture of the Parquet file.
The GeoParquet file must embed the properties in the Parquet metadata in a property named collection.
The metadata must be JSON-encoded.
The mapping between the Parquet data types and the Vecorel SDL data types, can be found in the data type mapping.
Related documents:
For data with a lot of repetition, brotli compression is recommended. This applies particularly for merged datasets that don't deduplicate properties to the collection-level.