You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Some plans are forming around the Covid Recovery Dash:
In the immediate term, we'd like to automatically update it daily
In the medium term, we want to port its functionality over to the System Ridership tab of the Data Dashboard
In either case, the first step is to generate the CRD's data.json file from our ScheduledServiceDaily and Ridership Dynamo tables. An example of the file is here, and the code that generates it is here. With the daily ridership and service level data already available in Dynamo, there's a lot of logic from the CRD codebase that can be dropped or ignored — the important thing is that the structure of the file should remain the same.
In the end, we should have a cron job in this repo that automatically updates this file daily, running shortly after the tables mentioned above are updated, and drops it somewhere in S3. We will probably want to beef up the lambda_timeout and lambda_memory_size for the job to make sure it can reliably complete — it's got a lot of work to do. We should also consider parallelizing Dynamo reads if possible, since it'll need to look up values for ~100 routes.
Nice-to-haves:
We'd prefer to have the file contain pre-covid history too, perhaps back to 2016
If the file is getting very large, we should consider chunking it by route
If you can add ridership and service breakdowns by mode while you're in here, please do! There's already a spot for system-wide values that this can slot into nearby.
The text was updated successfully, but these errors were encountered:
Some plans are forming around the Covid Recovery Dash:
In either case, the first step is to generate the CRD's
data.json
file from ourScheduledServiceDaily
andRidership
Dynamo tables. An example of the file is here, and the code that generates it is here. With the daily ridership and service level data already available in Dynamo, there's a lot of logic from the CRD codebase that can be dropped or ignored — the important thing is that the structure of the file should remain the same.In the end, we should have a cron job in this repo that automatically updates this file daily, running shortly after the tables mentioned above are updated, and drops it somewhere in S3. We will probably want to beef up the
lambda_timeout
andlambda_memory_size
for the job to make sure it can reliably complete — it's got a lot of work to do. We should also consider parallelizing Dynamo reads if possible, since it'll need to look up values for ~100 routes.Nice-to-haves:
The text was updated successfully, but these errors were encountered: