Tileverse¶

Tileverse provides a toolkit of independent Java libraries for building cloud-native geospatial applications.

These libraries are designed to be composable: pick the ones you need for your specific problem—whether that's reading from cloud storage, handling vector tiles, or calculating tile grids—without pulling in a monolithic framework.

Libraries¶

Library	Artifact	Role
Storage	`tileverse-storage-all`	Object Storage Abstraction: Full container API (list, range/streaming reads, atomic writes, deletes, server-side copy, presigned URLs) across local files, HTTP, S3 (GP + Express), Azure (Blob + Data Lake Gen2), and GCS (flat + HNS). Includes the byte-range `RangeReader` API used by PMTiles, COG, and single-file Parquet.
PMTiles	`tileverse-pmtiles`	Archive Format: Read/write support for PMTiles v3, leveraging Storage / RangeReader for cloud access.
Vector Tiles	`tileverse-vectortiles`	Codec: High-performance encoding and decoding of Mapbox Vector Tiles (MVT) to/from JTS Geometries.
Tile Matrix Set	`tileverse-tilematrixset`	Math & Logic: Implementation of the OGC Tile Matrix Set standard for calculating tile pyramids and grids.

Ecosystem¶

While tileverse-pmtiles naturally uses the other libraries, Storage, Vector Tiles, and Tile Matrix Set are completely standalone.

Building a GeoParquet datastore that needs to list and read partitioned files in S3 / Azure / GCS? Use Storage.
Reading COGs (Cloud Optimized GeoTIFFs) with byte-range requests? Use Storage (its RangeReader API) directly.
Building a tile server from PostGIS? Use Vector Tiles and Tile Matrix Set.
Need a standard grid definition? Use Tile Matrix Set.

graph TD
    App[Your App]

    subgraph "I/O"
        ST[Storage<br/>+ RangeReader API]
    end

    subgraph "Formats"
        VT[Vector Tiles]
        PMT[PMTiles]
    end

    subgraph "Spatial"
        TMS[Tile Matrix Set]
    end

    App --> ST
    App --> VT
    App --> PMT
    App --> TMS

    PMT -.-> ST
    PMT -.-> VT
    PMT -.-> TMS