When I was describing Jayvee at TU Berlin, I felt like we are missing a stronger, domain-specific example.
This PR uses the new composite block capabilities and some of the GTFS domain knowledge we developed to start a "domain extension" for mobility data to the stdLib. Using this extension, we can shorten the GTFS example quite a bit from 358 lines to 91 (and the result should be better because we are using domain specific value types).
Further progress could be made with configurable blocks (to reduce the boilerplate when creating SQLite sinks that only differ in a table name) and better ways to describe DAGs (to reduce the initial bloat with creating many different lines instead of one graph).
Let me know what you think of this in general, I'd like to create something similar for SWC (materials science domain). We can slowly move value types that are more generic than the domain they are initially defined in to a higher level (e.g. I anticipate the Latitude and Longitude value types are not mobility specific).
When I was describing Jayvee at TU Berlin, I felt like we are missing a stronger, domain-specific example.
This PR uses the new composite block capabilities and some of the GTFS domain knowledge we developed to start a "domain extension" for mobility data to the stdLib. Using this extension, we can shorten the GTFS example quite a bit from 358 lines to 91 (and the result should be better because we are using domain specific value types).
Further progress could be made with configurable blocks (to reduce the boilerplate when creating SQLite sinks that only differ in a table name) and better ways to describe DAGs (to reduce the initial bloat with creating many different lines instead of one graph).
Let me know what you think of this in general, I'd like to create something similar for SWC (materials science domain). We can slowly move value types that are more generic than the domain they are initially defined in to a higher level (e.g. I anticipate the Latitude and Longitude value types are not mobility specific).