I know previously there was no good pre-seeding capabilities a couple of years ago, but according to the 2.10 docs, it seems that the feature was finally fleshed out as a stand-alone application bundled in the /bin of each release now. I'm really glad that this was done, but I had a question about replicating what the seeder does from the context in source code, rather than using a .earth file as input from the command line with the osgearth_cache script. Our application has deprecated the usage of .earth files as we opted to do everything from source which includes:
1. Creating the Map
2. Creating the MapNode
3. Creating the Image layers and Elevation layers and adding them to the Map
What I want to do is have the ability for GeoTiff, DTED, CADRG, and ECRG layers to be pre-seeded at all levels of detail so when a new user uses the application for the first time they don't have to wait for their cache to initialize naturally, since it can take a decent amount of time for so many differing layers. If this still isn't possible from a source code perspective, is there any plans to implement doing so from code using a MapNode as the input and have all of the same options available as the command line osgearth_cache script has?
Re: Pre-Seeding a cache without using a .earth file
After a bit of digging I found the solution to my own question by looking for the osgearth_cache application source.
Essentially this file describes exactly what I want to do on the map node and even has a super convenient callback mechanism for tracking the progress on a front-end widget if I wanted through polling.