odin_fires

The odin_fires application domain crate provides a data model and SpaService to display operational fire data such as name, incident identifiers, location, perimeters and more.

This was started as an attempt to provide visualization for the “Fire Progression” database from https://data-nifc.opendata.arcgis.com (which is succeeded by the “Historical Operational Data” database). This data was not directly displayable as it contains GeoJSON FeatureCollections with all perimeter Polygons in one file. Our intention is to extend the functionality to a full data model for wildfire incidents that can be dynamically updated, i.e. that is applicable to both historical and ongoing incidents. This model should

• be extensible
• be backed by a hierarchical file system in which each incident has its own subdirectory
• use standard external file formats such as JSON and GeoJSON
• use a summary (JSON) file in each incident directory that contains links to all available incident data such as perimeters

Since incident data is mainly produced either manually or by external systems and applications odin_fires does not have a classic import actor and operates solely on its file system in ODIN_ROOT/data/odin_fires. This file system is currently analyzed during initialization of the FireService but eventually will be monitored continuously for changes that are to be distributed as summary updates to connected clients.

Before such content is sent to clients odin_fires has to make sure the data conforms to our internal data model, i.e. all data has to be parsed into our model before sending it out so that our odin_fires.js Javascript module will be able to process the data.

Consequently the main construct of the odin_fires crate is the FireSummary data model in its lib.rs. The crate also provides a FireService that is a simple odin_server::spa::SpaService implementation.

The crate also contains tools and generators for related incident files such as the opendata_splitter command line executable.