For someone to replicate the entire process of transforming O/D data to simulated driver work shifts and aggregate VKTs, routes and feasible en-route paths must be generated using pgRouting. We need to productionize our PostgreSQL code to produce and post-process the PTC O/D data, but there isn't a straightforward procedure to do this as with Python code. We'll need to discuss:
How generalizable can we make the Postgres scripts? Can they be turned into functions, then published with some discussion on deployment, or should we instead upload our own scripts with instructions on how to rewrite them for other databases?
Should this code go in the same package as the trip linker? If so, should we rename this package to bdit_triprouter to cover all its activities (and do CI and coverage tests only on the Python code)? If not, and we create a separate bdit_triprouter repository, where should the white paper go?
For someone to replicate the entire process of transforming O/D data to simulated driver work shifts and aggregate VKTs, routes and feasible en-route paths must be generated using pgRouting. We need to productionize our PostgreSQL code to produce and post-process the PTC O/D data, but there isn't a straightforward procedure to do this as with Python code. We'll need to discuss:
bdit_triprouter
to cover all its activities (and do CI and coverage tests only on the Python code)? If not, and we create a separatebdit_triprouter
repository, where should the white paper go?