Skip to content

only expand update graph if needed #38332

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
reuvenlax merged 1 commit into from
Apr 30, 2026
Merged

only expand update graph if needed #38332

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
206 changes: 107 additions & 99 deletions ...platform/src/main/java/org/apache/beam/sdk/io/gcp/bigquery/StorageApiConvertMessages.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -147,109 +147,117 @@ public PCollectionTuple expand(PCollection<KV<DestinationT, ElementT>> input) {
.get(patchTableSchemaTag)
.setCoder(KvCoder.of(destinationCoder, ProtoCoder.of(TableSchema.class)));
result.get(elementsWaitingForSchemaTag).setCoder(KvCoder.of(destinationCoder, elementCoder));
if (!hasSchemaUpdateOptions) {
// Don't expand the update graph if it's not needed.
return result;
} else {
final int numShards =
input
.getPipeline()
.getOptions()
.as(BigQueryOptions.class)
.getSchemaUpgradeBufferingShards();

final int numShards =
input
.getPipeline()
.getOptions()
.as(BigQueryOptions.class)
.getSchemaUpgradeBufferingShards();
// Throttle the stream to the patch-table function so that only a single update per table per
// two seconds gets processed (to match quotas). The combiner merges incremental schemas, so
// we
// won't miss any updates.
PCollection<KV<ShardedKey<DestinationT>, ElementT>> tablesPatched =
result
.get(patchTableSchemaTag)
.apply(
"rewindow",
Window.<KV<DestinationT, TableSchema>>configure()
.triggering(
Repeatedly.forever(
AfterProcessingTime.pastFirstElementInPane()
.plusDelayOf(Duration.standardSeconds(2))))
.discardingFiredPanes())
.apply("merge schemas", Combine.fewKeys(new MergeSchemaCombineFn()))
.setCoder(KvCoder.of(destinationCoder, ProtoCoder.of(TableSchema.class)))
.apply(
"Patch table schema",
ParDo.of(
new PatchTableSchemaDoFn<>(operationName, bqServices, dynamicDestinations)))
.setCoder(KvCoder.of(destinationCoder, NullableCoder.of(elementCoder)))
// We need to make sure that all shards of the buffering transform are notified.
.apply(
"fanout to all shards",
FlatMapElements.via(
new SimpleFunction<
KV<DestinationT, ElementT>,
Iterable<KV<ShardedKey<DestinationT>, ElementT>>>() {
@Override
public Iterable<KV<ShardedKey<DestinationT>, ElementT>> apply(
KV<DestinationT, ElementT> elem) {
return IntStream.range(0, numShards)
.mapToObj(
i ->
KV.of(
StorageApiConvertMessages.AssignShardFn.getShardedKey(
elem.getKey(), i, numShards),
elem.getValue()))
.collect(Collectors.toList());
}
}))
.setCoder(
KvCoder.of(ShardedKey.Coder.of(destinationCoder), NullableCoder.of(elementCoder)))
.apply(
Window.<KV<ShardedKey<DestinationT>, ElementT>>configure()
.triggering(DefaultTrigger.of()));

// Throttle the stream to the patch-table function so that only a single update per table per
// two seconds gets processed (to match quotas). The combiner merges incremental schemas, so we
// won't miss any updates.
PCollection<KV<ShardedKey<DestinationT>, ElementT>> tablesPatched =
result
.get(patchTableSchemaTag)
.apply(
"rewindow",
Window.<KV<DestinationT, TableSchema>>configure()
.triggering(
Repeatedly.forever(
AfterProcessingTime.pastFirstElementInPane()
.plusDelayOf(Duration.standardSeconds(2))))
.discardingFiredPanes())
.apply("merge schemas", Combine.fewKeys(new MergeSchemaCombineFn()))
.setCoder(KvCoder.of(destinationCoder, ProtoCoder.of(TableSchema.class)))
.apply(
"Patch table schema",
ParDo.of(
new PatchTableSchemaDoFn<>(operationName, bqServices, dynamicDestinations)))
.setCoder(KvCoder.of(destinationCoder, NullableCoder.of(elementCoder)))
// We need to make sure that all shards of the buffering transform are notified.
.apply(
"fanout to all shards",
FlatMapElements.via(
new SimpleFunction<
KV<DestinationT, ElementT>,
Iterable<KV<ShardedKey<DestinationT>, ElementT>>>() {
@Override
public Iterable<KV<ShardedKey<DestinationT>, ElementT>> apply(
KV<DestinationT, ElementT> elem) {
return IntStream.range(0, numShards)
.mapToObj(
i ->
KV.of(
StorageApiConvertMessages.AssignShardFn.getShardedKey(
elem.getKey(), i, numShards),
elem.getValue()))
.collect(Collectors.toList());
}
}))
.setCoder(
KvCoder.of(ShardedKey.Coder.of(destinationCoder), NullableCoder.of(elementCoder)))
.apply(
Window.<KV<ShardedKey<DestinationT>, ElementT>>configure()
.triggering(DefaultTrigger.of()));
// Any elements that are waiting for a schema update are sent to this stateful DoFn to be
// buffered.
// Note: we currently do not provide the DynamicDestinations object access to the side input
// in
// this path.
// This is because side inputs are not currently available from timer callbacks. Since side
// inputs are generally
// used for getSchema and in this case we read the schema from the table, this is unlikely to
// be
// a problem.
PCollection<KV<ShardedKey<DestinationT>, ElementT>> shardedWaitingElements =
result
.get(elementsWaitingForSchemaTag)
// TODO: Consider using GroupIntoBatchs.withShardingKey to get auto sharding here
// instead of fixed sharding.
.apply("assignShard", ParDo.of(new AssignShardFn<>(numShards)))
.setCoder(
KvCoder.of(
ShardedKey.Coder.of(destinationCoder), NullableCoder.of(elementCoder)));

// Any elements that are waiting for a schema update are sent to this stateful DoFn to be
// buffered.
// Note: we currently do not provide the DynamicDestinations object access to the side input in
// this path.
// This is because side inputs are not currently available from timer callbacks. Since side
// inputs are generally
// used for getSchema and in this case we read the schema from the table, this is unlikely to be
// a problem.
PCollection<KV<ShardedKey<DestinationT>, ElementT>> shardedWaitingElements =
result
.get(elementsWaitingForSchemaTag)
// TODO: Consider using GroupIntoBatchs.withShardingKey to get auto sharding here
// instead of fixed sharding.
.apply("assignShard", ParDo.of(new AssignShardFn<>(numShards)))
.setCoder(
KvCoder.of(ShardedKey.Coder.of(destinationCoder), NullableCoder.of(elementCoder)));
PCollectionList<KV<ShardedKey<DestinationT>, ElementT>> waitingElementsList =
PCollectionList.of(shardedWaitingElements).and(tablesPatched);
PCollectionTuple retryResult =
waitingElementsList
.apply("Buffered flatten", Flatten.pCollections())
.apply(
"bufferElements",
ParDo.of(new SchemaUpdateHoldingFn<>(elementCoder, convertMessagesDoFn))
.withOutputTags(
successfulWritesTag,
TupleTagList.of(ImmutableList.of(failedWritesTag, BAD_RECORD_TAG))));
retryResult.get(successfulWritesTag).setCoder(successCoder);
retryResult.get(failedWritesTag).setCoder(errorCoder);
retryResult.get(BAD_RECORD_TAG).setCoder(BadRecord.getCoder(input.getPipeline()));

PCollectionList<KV<ShardedKey<DestinationT>, ElementT>> waitingElementsList =
PCollectionList.of(shardedWaitingElements).and(tablesPatched);
PCollectionTuple retryResult =
waitingElementsList
.apply("Buffered flatten", Flatten.pCollections())
.apply(
"bufferElements",
ParDo.of(new SchemaUpdateHoldingFn<>(elementCoder, convertMessagesDoFn))
.withOutputTags(
successfulWritesTag,
TupleTagList.of(ImmutableList.of(failedWritesTag, BAD_RECORD_TAG))));
retryResult.get(successfulWritesTag).setCoder(successCoder);
retryResult.get(failedWritesTag).setCoder(errorCoder);
retryResult.get(BAD_RECORD_TAG).setCoder(BadRecord.getCoder(input.getPipeline()));

// Flatten successes and failures from both the regular transform and the retry transform.
PCollection<KV<DestinationT, StorageApiWritePayload>> allSuccesses =
PCollectionList.of(result.get(successfulWritesTag))
.and(retryResult.get(successfulWritesTag))
.apply("flattenSuccesses", Flatten.pCollections());
PCollection<BigQueryStorageApiInsertError> allFailures =
PCollectionList.of(result.get(failedWritesTag))
.and(retryResult.get(failedWritesTag))
.apply("flattenFailures", Flatten.pCollections());
PCollection<BadRecord> allBadRecords =
PCollectionList.of(result.get(BAD_RECORD_TAG))
.and(retryResult.get(BAD_RECORD_TAG))
.apply("flattenBadRecords", Flatten.pCollections());
return PCollectionTuple.of(successfulWritesTag, allSuccesses)
.and(failedWritesTag, allFailures)
.and(BAD_RECORD_TAG, allBadRecords);
// Flatten successes and failures from both the regular transform and the retry transform.
PCollection<KV<DestinationT, StorageApiWritePayload>> allSuccesses =
PCollectionList.of(result.get(successfulWritesTag))
.and(retryResult.get(successfulWritesTag))
.apply("flattenSuccesses", Flatten.pCollections());
PCollection<BigQueryStorageApiInsertError> allFailures =
PCollectionList.of(result.get(failedWritesTag))
.and(retryResult.get(failedWritesTag))
.apply("flattenFailures", Flatten.pCollections());
PCollection<BadRecord> allBadRecords =
PCollectionList.of(result.get(BAD_RECORD_TAG))
.and(retryResult.get(BAD_RECORD_TAG))
.apply("flattenBadRecords", Flatten.pCollections());
return PCollectionTuple.of(successfulWritesTag, allSuccesses)
.and(failedWritesTag, allFailures)
.and(BAD_RECORD_TAG, allBadRecords);
}
Comment on lines +153 to +260

@gemini-code-assist gemini-code-assist Bot Apr 29, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

To improve readability and reduce nesting, you can use a guard clause. Instead of wrapping the rest of the method in an else block, you can return early and then have the main logic at the top level of the method.

    final int numShards =
        input
            .getPipeline()
            .getOptions()
            .as(BigQueryOptions.class)
            .getSchemaUpgradeBufferingShards();

    // Throttle the stream to the patch-table function so that only a single update per table per
    // two seconds gets processed (to match quotas). The combiner merges incremental schemas, so
    // we
    // won't miss any updates.
    PCollection<KV<ShardedKey<DestinationT>, ElementT>> tablesPatched =
        result
            .get(patchTableSchemaTag)
            .apply(
                "rewindow",
                Window.<KV<DestinationT, TableSchema>>configure()
                    .triggering(
                        Repeatedly.forever(
                            AfterProcessingTime.pastFirstElementInPane()
                                .plusDelayOf(Duration.standardSeconds(2))))
                    .discardingFiredPanes())
            .apply("merge schemas", Combine.fewKeys(new MergeSchemaCombineFn()))
            .setCoder(KvCoder.of(destinationCoder, ProtoCoder.of(TableSchema.class)))
            .apply(
                "Patch table schema",
                ParDo.of(
                    new PatchTableSchemaDoFn<>(operationName, bqServices, dynamicDestinations)))
            .setCoder(KvCoder.of(destinationCoder, NullableCoder.of(elementCoder)))
            // We need to make sure that all shards of the buffering transform are notified.
            .apply(
                "fanout to all shards",
                FlatMapElements.via(
                    new SimpleFunction<
                        KV<DestinationT, ElementT>,
                        Iterable<KV<ShardedKey<DestinationT>, ElementT>>>() {
                      @Override
                      public Iterable<KV<ShardedKey<DestinationT>, ElementT>> apply(
                          KV<DestinationT, ElementT> elem) {
                        return IntStream.range(0, numShards)
                            .mapToObj(
                                i ->
                                    KV.of(
                                        StorageApiConvertMessages.AssignShardFn.getShardedKey(
                                            elem.getKey(), i, numShards),
                                        elem.getValue()))
                            .collect(Collectors.toList());
                      }
                    }))
            .setCoder(
                KvCoder.of(ShardedKey.Coder.of(destinationCoder), NullableCoder.of(elementCoder)))
            .apply(
                Window.<KV<ShardedKey<DestinationT>, ElementT>>configure()
                    .triggering(DefaultTrigger.of()));

    // Any elements that are waiting for a schema update are sent to this stateful DoFn to be
    // buffered.
    // Note: we currently do not provide the DynamicDestinations object access to the side input
    // in
    // this path.
    // This is because side inputs are not currently available from timer callbacks. Since side
    // inputs are generally
    // used for getSchema and in this case we read the schema from the table, this is unlikely to
    // be
    // a problem.
    PCollection<KV<ShardedKey<DestinationT>, ElementT>> shardedWaitingElements =
        result
            .get(elementsWaitingForSchemaTag)
            // TODO: Consider using GroupIntoBatchs.withShardingKey to get auto sharding here
            // instead of fixed sharding.
            .apply("assignShard", ParDo.of(new AssignShardFn<>(numShards)))
            .setCoder(
                KvCoder.of(
                    ShardedKey.Coder.of(destinationCoder), NullableCoder.of(elementCoder)));

    PCollectionList<KV<ShardedKey<DestinationT>, ElementT>> waitingElementsList =
        PCollectionList.of(shardedWaitingElements).and(tablesPatched);
    PCollectionTuple retryResult =
        waitingElementsList
            .apply("Buffered flatten", Flatten.pCollections())
            .apply(
                "bufferElements",
                ParDo.of(new SchemaUpdateHoldingFn<>(elementCoder, convertMessagesDoFn))
                    .withOutputTags(
                        successfulWritesTag,
                        TupleTagList.of(ImmutableList.of(failedWritesTag, BAD_RECORD_TAG))));
    retryResult.get(successfulWritesTag).setCoder(successCoder);
    retryResult.get(failedWritesTag).setCoder(errorCoder);
    retryResult.get(BAD_RECORD_TAG).setCoder(BadRecord.getCoder(input.getPipeline()));

    // Flatten successes and failures from both the regular transform and the retry transform.
    PCollection<KV<DestinationT, StorageApiWritePayload>> allSuccesses =
        PCollectionList.of(result.get(successfulWritesTag))
            .and(retryResult.get(successfulWritesTag))
            .apply("flattenSuccesses", Flatten.pCollections());
    PCollection<BigQueryStorageApiInsertError> allFailures =
        PCollectionList.of(result.get(failedWritesTag))
            .and(retryResult.get(failedWritesTag))
            .apply("flattenFailures", Flatten.pCollections());
    PCollection<BadRecord> allBadRecords =
        PCollectionList.of(result.get(BAD_RECORD_TAG))
            .and(retryResult.get(BAD_RECORD_TAG))
            .apply("flattenBadRecords", Flatten.pCollections());
    return PCollectionTuple.of(successfulWritesTag, allSuccesses)
        .and(failedWritesTag, allFailures)
        .and(BAD_RECORD_TAG, allBadRecords);

}

static class AssignShardFn<K, V> extends DoFn<KV<K, V>, KV<ShardedKey<K>, V>> {
Expand Down
Loading