mjdpollo opened a new issue, #863: URL: https://github.com/apache/sedona-db/issues/863
### Is your feature request related to a problem? Please describe. Currently, Apache Sedona provides a highly robust suite of Raster Functional APIs (`RS_` functions) within the JVM and Python ecosystems, relying heavily on GeoTools and distributed Spark/Flink architectures. However, as the ecosystem expands toward native Rust performance and lighter-weight deployments, there is a visible gap in pixel-level raster manipulation and map algebra functionality within the `sedona-rust` crate. For developers looking to build highly performant, native spatial pipelines in Rust or WebAssembly without a JVM dependency, the lack of raster analytical capabilities (e.g., raster loading, map algebra, metadata extraction) limits the utility of Sedona's Rust bindings compared to its Java/Python counterparts. ### Describe the solution you'd like I would like to inquire if there is an official roadmap or an open tracking issue for porting Sedona's Raster Functional APIs to the native Rust codebase. Specifically, it would be incredibly beneficial to understand: 1. **Core Milestones:** Are there plans to incrementally port fundamental raster types and basic operations (e.g., `RS_LoadRaster`, `RS_Value`, metadata accessors)? 2. **Underlying Dependencies:** Is the community considering leveraging native Rust crates like `gdal` or `geo-raster` to handle the underlying pixel-level logic, or is an entirely native implementation envisioned? 3. **Contribution Points:** Are there specific architectural bottlenecks (such as memory management or geometry conversion boundaries) preventing this, and how can external contributors best help jumpstart this effort? ### Describe alternatives you've considered Currently, the primary workarounds involve: * Falling back to the JVM/Python Sedona environment when heavy raster manipulation is required, which negates the deployment advantages of a native Rust environment. * Interoperating with external Rust GIS crates (like `gdal` or `geo`) manually alongside `sedona-rust` for vector-only tasks, which introduces serialization overhead and fragmented workflows. ### Additional context Thank you all for the incredible work on Apache Sedona and the continuous push toward bringing spatial analytics to the Rust ecosystem! I am eager to hear the core maintainers' thoughts on where raster capabilities stand in the long-term vision of `sedona-rust`. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
