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
When the field.set processor is used, then some fields can be lost, like in the pipeline below that uses a generator source, field.set to add a new field, and a file destination.
When the field.set processor is not present, then the generated records have the fields as expected.
Yeah, I changed the steps a bit, thinking they could be simplified. I'll update the steps in a bit to what they were (and they consistently reproduce this behavior). It looks like the issue is in our schema registry and the way we generate IDs.
Bug description
When the
field.set
processor is used, then some fields can be lost, like in the pipeline below that uses a generator source,field.set
to add a new field, and a file destination.When the
field.set
processor is not present, then the generated records have the fields as expected.Steps to reproduce
destination.txt
withtail -f destination.txt | jq
. All the records will look similar to this:field.set
processor, like below:destination.txt
withtail -f destination.txt | jq
. All the records will look similar to this:Expected result:
Field
scheduledDeparture
has a non-zero value,city
is present and set toBedrock
.Actual result:
Field
scheduledDeparture
has a zero value,city
is not present.Version
Commit 4c3c375
The text was updated successfully, but these errors were encountered: