The cloud-native orchestrator for the whole development lifecycle, with integrated lineage and observability.

dagster

If I have 2 assets, one a daily partitioned asset (asset1), and the other a daily partitioned Time Window asset (asset2) with a start date of -1, and end date of 0, dependent on asset1. How would I get the return data from the 2 upstream partitions on asset1?

Hey <@U04N73MPZUM> assuming you want to get the previous two days of data from asset1 as input to asset2, you can do something similar to <https://docs.dagster.io/concepts/partitions-schedules-sensors/partitions#partitionmappings|this example > but instead of
```partition_mapping=TimeWindowPartitionMapping(
                start_offset=-1, end_offset=-1
            ),```
you could do
```partition_mapping=TimeWindowPartitionMapping(
                start_offset=-2, end_offset=-1 # start_offset is modified
            ),```
to specify the previous 2 days of data

What would the `events` input for `yesterday_event_stats` look like? How would I specify getting the different partitions data?

If I am returning a dataframe from the `events` asset

it depends on the io manager you are using. If you are using the default io manager, i believe it will be a list of dataframes. If you are using something like the snowflake io manager it would be a single dataframe with all of the data

Thanks <@U030M2AL48M>! that clears it up. I am just having trouble organizing my partitions. Is there a such thing as `JobIn` like `AssetIn`? I setup my partition as a job because I wanted to view ops to track the progress of each partition. But now how would I add a `TimeWindowPartitionMapping` an analytics job? I could change my partition to an asset, but I don't need every op to be an asset, which is why I like the job partition.