Vitaly Terentyev comments

Results 67 comments of


                                            Vitaly Terentyev

Support withFormatRecordOnFailureFunction() for BigQuery STORAGE_WRITE_API and STORAGE_API_AT_LEAST_ONCE methods

> @Amar3tto can we add a comment in CHANGES.md calling out this new feature? Done

Support withFormatRecordOnFailureFunction() for BigQuery STORAGE_WRITE_API and STORAGE_API_AT_LEAST_ONCE methods

Run Java_GCP_IO_Direct PreCommit

[Feature Request]: UDF supports lambda & RecordType parameter

There are some difficulties in supporting UDFs with `@Parameter(...) Row` and `@Parameter(...) Predicate1`: 1. Each parameter type should be [translatable](https://github.com/apache/beam/blob/90d3f8a177ff935a09852fd85892bb1399b85c91/sdks/java/extensions/sql/src/main/java/org/apache/beam/sdk/extensions/sql/impl/utils/CalciteUtils.java#L340) to Calcite `RelDataType`. 2. If we want to translate `Row`...

[Feature Request]: UDF supports lambda & RecordType parameter

@gabrywu FYI

[Bug]: PubsubMessage should be renamed to PubSubMessage to reduce confusion

`PubSubMessage` naming will be not consistent, because: 1. We have `PubsubIO` and other classes (not `PubSubIO`) 2. `com.google.pubsub.v1` has `PubsubMessage` class. `PubsubMessage` naming also wouldn't be consistent, because: 1. `com.google.cloud.pubsublite.proto`...

[Bug]: PubsubMessage should be renamed to PubSubMessage to reduce confusion

FYI @aaltay

[Feature Request]: Handle failed inserts in Storage API sink

Can this issue be closed or is there something else left?

[Bug]: apache.calcite CannotPlanException: All the inputs have relevant nodes, however the cost is still infinite.

I tested locally different queries and here are the results. I guess `total` and `avg1`, `avg2` columns are somehow related and together lead to an error in certain cases. Successful:...

Add KMS_KEY_NAME to ImageSpecParameterType

Note: fix UI for the KMS_KEY_NAME parameter before merging this PR

The PostCommit TransformService Direct job is flaky

Stabilized