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
PR #427 introduced changes which made some readers able to handle multiple scan times, as opposed to a single scan time. This is needed for certain applications at CIRA (such as OCTANE), but introduces different challenges. To make sure that both single scan time or multi-scan time reads work for these updated readers, we should add a suite of integration tests which address these new changes. Specifically, we should put together a few test datasets which have multiple scan times, add an algorithm which can handle this new 'time' dimension, and produce some output to test against.
We should keep these datasets as small as possible, so it's probably best to use M01 for GOES16 ABI, LA for GEOKOMPSAT AMI,
JP01 for Himawari9 AHI, and other small sectors for Meteosat[X]-SEVIRI.
Background and Motivation
This comes from a conversation that @jsolbrig and I had when testing #427.
Code to demonstrate issue
These readers are able to handle multiple scan time:
Checklist for Completion
Upload applicable test datasets to CIRA's NextCloud instance (see readers above for what data to upload)
Update download test scripts to include these new datasets
Create new products for multiple scan times
Create a new algorithm which can handle 2 times or more
Add new integration tests which test this functionality
The text was updated successfully, but these errors were encountered:
Requested Update
Description
PR #427 introduced changes which made some readers able to handle multiple scan times, as opposed to a single scan time. This is needed for certain applications at CIRA (such as OCTANE), but introduces different challenges. To make sure that both single scan time or multi-scan time reads work for these updated readers, we should add a suite of integration tests which address these new changes. Specifically, we should put together a few test datasets which have multiple scan times, add an algorithm which can handle this new 'time' dimension, and produce some output to test against.
We should keep these datasets as small as possible, so it's probably best to use M01 for GOES16 ABI, LA for GEOKOMPSAT AMI,
JP01 for Himawari9 AHI, and other small sectors for Meteosat[X]-SEVIRI.
Background and Motivation
This comes from a conversation that @jsolbrig and I had when testing #427.
Code to demonstrate issue
These readers are able to handle multiple scan time:
Checklist for Completion
The text was updated successfully, but these errors were encountered: