Skip to content

feat(bigquery): integrate Arrow query response processing and stream pagination #13944

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

Draft
jinseopkim0 wants to merge 1 commit into
base: feat-bigquery-arrow-deserializer
Choose a base branch
from
Draft

feat(bigquery): integrate Arrow query response processing and stream pagination #13944

Show file tree
Hide file tree
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
253 changes: 235 additions & 18 deletions ...-bigquery/google-cloud-bigquery/src/main/java/com/google/cloud/bigquery/BigQueryImpl.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,9 @@

import com.google.api.core.BetaApi;
import com.google.api.core.InternalApi;
import com.google.api.gax.core.FixedCredentialsProvider;
import com.google.api.gax.paging.Page;
import com.google.api.gax.rpc.ServerStream;
import com.google.api.services.bigquery.model.ErrorProto;
import com.google.api.services.bigquery.model.GetQueryResultsResponse;
import com.google.api.services.bigquery.model.ProjectList;
Expand All @@ -43,6 +45,10 @@
import com.google.cloud.bigquery.InsertAllRequest.RowToInsert;
import com.google.cloud.bigquery.spi.v2.BigQueryRpc;
import com.google.cloud.bigquery.spi.v2.HttpBigQueryRpc;
import com.google.cloud.bigquery.storage.v1.BigQueryReadClient;
import com.google.cloud.bigquery.storage.v1.BigQueryReadSettings;
import com.google.cloud.bigquery.storage.v1.ReadRowsRequest;
import com.google.cloud.bigquery.storage.v1.ReadRowsResponse;
import com.google.common.annotations.VisibleForTesting;
import com.google.common.base.Function;
import com.google.common.base.Strings;
Expand All @@ -57,6 +63,7 @@
import io.opentelemetry.context.Scope;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.concurrent.Callable;
Expand Down Expand Up @@ -264,6 +271,143 @@ public Page<FieldValueList> getNextPage() {
}
}

private static class ArrowQueryPageFetcher implements NextPageFetcher<FieldValueList> {
private static final long serialVersionUID = 1L;

private final JobId jobId;
private final Schema schema;
private final org.apache.arrow.vector.types.pojo.Schema arrowSchemaPojo;
private final BigQueryOptions serviceOptions;
private final long maxResults;

private transient BigQueryReadClient bqReadClient;
private transient ServerStream<ReadRowsResponse> stream;
private transient Iterator<ReadRowsResponse> streamIterator;
private long totalRowsReturned = 0L;
private boolean streamClosed = false;

ArrowQueryPageFetcher(
JobId jobId,
Schema schema,
org.apache.arrow.vector.types.pojo.Schema arrowSchemaPojo,
BigQueryOptions serviceOptions,
long initialRowOffset,
Long maxResults) {
this.jobId = jobId;
this.schema = schema;
this.arrowSchemaPojo = arrowSchemaPojo;
this.serviceOptions = serviceOptions;
this.totalRowsReturned = initialRowOffset;
this.maxResults = maxResults != null ? maxResults : Long.MAX_VALUE;
}
Comment on lines +277 to +302

@gemini-code-assist gemini-code-assist Bot Jul 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.

high

NextPageFetcher is a Serializable interface, meaning implementations like ArrowQueryPageFetcher must be fully serializable. However, org.apache.arrow.vector.types.pojo.Schema does not implement Serializable, which will cause a NotSerializableException if the page or fetcher is serialized.

To fix this, we can store the schema as a JSON string (arrowSchemaJson) using Arrow's built-in Schema.toJson() and Schema.fromJSON(String) methods, and mark the arrowSchemaPojo field as transient so it is lazily deserialized when needed.

    private final JobId jobId;
    private final Schema schema;
    private final String arrowSchemaJson;
    private final BigQueryOptions serviceOptions;
    private final long maxResults;

    private transient org.apache.arrow.vector.types.pojo.Schema arrowSchemaPojo;
    private transient BigQueryReadClient bqReadClient;
    private transient ServerStream<ReadRowsResponse> stream;
    private transient Iterator<ReadRowsResponse> streamIterator;
    private long totalRowsReturned = 0L;
    private boolean streamClosed = false;

    ArrowQueryPageFetcher(
        JobId jobId,
        Schema schema,
        org.apache.arrow.vector.types.pojo.Schema arrowSchemaPojo,
        BigQueryOptions serviceOptions,
        long initialRowOffset,
        Long maxResults) {
      this.jobId = jobId;
      this.schema = schema;
      this.arrowSchemaJson = arrowSchemaPojo != null ? arrowSchemaPojo.toJson() : null;
      this.arrowSchemaPojo = arrowSchemaPojo;
      this.serviceOptions = serviceOptions;
      this.totalRowsReturned = initialRowOffset;
      this.maxResults = maxResults != null ? maxResults : Long.MAX_VALUE;
    }
References
  1. Prefer lazy initialization over eager initialization for resource-intensive objects (such as CharsetEncoder) if they are not guaranteed to be used in all execution paths, to avoid unnecessary performance and memory overhead.


@Override
public Page<FieldValueList> getNextPage() {
if (streamClosed || totalRowsReturned >= maxResults) {
closeClient();
return null;
}

long pageSize = 100000L;
List<FieldValueList> rowBatch = new ArrayList<>();

try {
if (bqReadClient == null) {
BigQueryReadSettings settings =
BigQueryReadSettings.newBuilder()
.setCredentialsProvider(
FixedCredentialsProvider.create(serviceOptions.getCredentials()))
.build();
bqReadClient = BigQueryReadClient.create(settings);
}

if (streamIterator == null) {
String streamName =
String.format(
"projects/%s/locations/%s/jobs/%s/streams/_default",
jobId.getProject() != null ? jobId.getProject() : serviceOptions.getProjectId(),
jobId.getLocation() != null ? jobId.getLocation() : serviceOptions.getLocation(),
jobId.getJob());

ReadRowsRequest readRowsRequest =
ReadRowsRequest.newBuilder()
.setReadStream(streamName)
.setOffset(totalRowsReturned)
.build();

stream = bqReadClient.readRowsCallable().call(readRowsRequest);
streamIterator = stream.iterator();
}

try (org.apache.arrow.memory.BufferAllocator allocator =
new org.apache.arrow.memory.RootAllocator(Long.MAX_VALUE)) {
List<org.apache.arrow.vector.FieldVector> vectors = new ArrayList<>();
for (org.apache.arrow.vector.types.pojo.Field field : arrowSchemaPojo.getFields()) {
vectors.add((org.apache.arrow.vector.FieldVector) field.createVector(allocator));
}
try (org.apache.arrow.vector.VectorSchemaRoot root =
new org.apache.arrow.vector.VectorSchemaRoot(vectors)) {
org.apache.arrow.vector.VectorLoader loader =
new org.apache.arrow.vector.VectorLoader(root);

while (rowBatch.size() < pageSize && streamIterator.hasNext()) {
ReadRowsResponse response = streamIterator.next();
if (response.hasArrowRecordBatch()) {
com.google.cloud.bigquery.storage.v1.ArrowRecordBatch batch =
response.getArrowRecordBatch();
org.apache.arrow.vector.ipc.message.ArrowRecordBatch deserializedBatch =
org.apache.arrow.vector.ipc.message.MessageSerializer.deserializeRecordBatch(
new org.apache.arrow.vector.ipc.ReadChannel(
new org.apache.arrow.vector.util.ByteArrayReadableSeekableByteChannel(
batch.getSerializedRecordBatch().toByteArray())),
allocator);
loader.load(deserializedBatch);
deserializedBatch.close();
int batchRowCount = root.getRowCount();
for (int i = 0; i < batchRowCount; i++) {
rowBatch.add(ArrowDeserializer.arrowRootToFieldValueList(root, i, schema));
}
root.clear();
}
}
}
}
Comment on lines +342 to +374

@gemini-code-assist gemini-code-assist Bot Jul 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.

high

There are two critical resource management issues in this block:

  1. Vector Allocation Leak: If an exception occurs while creating vectors in the loop, any previously allocated FieldVector instances in the vectors list will leak because they are not closed. Using VectorSchemaRoot.create(arrowSchemaPojo, allocator) is the standard, safe way to instantiate a VectorSchemaRoot and automatically handles cleanup of all vectors if allocation fails.
  2. ArrowRecordBatch Leak: org.apache.arrow.vector.ipc.message.ArrowRecordBatch is a closeable resource. If loader.load(deserializedBatch) throws an exception, deserializedBatch.close() is bypassed, causing a memory leak. Wrapping it in a try-with-resources block ensures it is always closed safely.

Additionally, we integrate the lazy deserialization of arrowSchemaPojo from arrowSchemaJson to support serialization.

        if (arrowSchemaPojo == null && arrowSchemaJson != null) {
          arrowSchemaPojo = org.apache.arrow.vector.types.pojo.Schema.fromJSON(arrowSchemaJson);
        }

        try (org.apache.arrow.memory.BufferAllocator allocator =
            new org.apache.arrow.memory.RootAllocator(Long.MAX_VALUE);
            org.apache.arrow.vector.VectorSchemaRoot root =
                org.apache.arrow.vector.VectorSchemaRoot.create(arrowSchemaPojo, allocator)) {
          org.apache.arrow.vector.VectorLoader loader =
              new org.apache.arrow.vector.VectorLoader(root);

          while (rowBatch.size() < pageSize && streamIterator.hasNext()) {
            ReadRowsResponse response = streamIterator.next();
            if (response.hasArrowRecordBatch()) {
              com.google.cloud.bigquery.storage.v1.ArrowRecordBatch batch =
                  response.getArrowRecordBatch();
              try (org.apache.arrow.vector.ipc.message.ArrowRecordBatch deserializedBatch =
                  org.apache.arrow.vector.ipc.message.MessageSerializer.deserializeRecordBatch(
                      new org.apache.arrow.vector.ipc.ReadChannel(
                          new org.apache.arrow.vector.util.ByteArrayReadableSeekableByteChannel(
                              batch.getSerializedRecordBatch().toByteArray())),
                      allocator)) {
                loader.load(deserializedBatch);
                int batchRowCount = root.getRowCount();
                for (int i = 0; i < batchRowCount; i++) {
                  rowBatch.add(ArrowDeserializer.arrowRootToFieldValueList(root, i, schema));
                }
                root.clear();
              }
            }
          }
        }
References
  1. Prefer lazy initialization over eager initialization for resource-intensive objects (such as CharsetEncoder) if they are not guaranteed to be used in all execution paths, to avoid unnecessary performance and memory overhead.


if (rowBatch.isEmpty()) {
streamClosed = true;
closeClient();
return null;
}

totalRowsReturned += rowBatch.size();

String nextPageToken = null;
if (streamIterator.hasNext() && totalRowsReturned < maxResults) {
nextPageToken = String.valueOf(totalRowsReturned);
} else {
streamClosed = true;
closeClient();
}

return new PageImpl<>(this, nextPageToken, rowBatch);

} catch (Exception e) {
streamClosed = true;
closeClient();
throw new BigQueryException(0, "Failed to read Arrow rows from storage stream", e);
}
}

private void closeClient() {
if (bqReadClient != null) {
bqReadClient.close();
bqReadClient = null;
}
streamIterator = null;
stream = null;
}
}

private final HttpBigQueryRpc bigQueryRpc;

private static final BigQueryRetryConfig EMPTY_RETRY_CONFIG =
Expand Down Expand Up @@ -2073,8 +2217,28 @@ public com.google.api.services.bigquery.model.QueryResponse call()

long numRows;
Schema schema;
if (results.getJobComplete() && results.getSchema() != null) {
schema = Schema.fromPb(results.getSchema());
boolean isArrow = false;
org.apache.arrow.vector.types.pojo.Schema arrowSchemaPojo = null;

if (results.getJobComplete()) {
if (results.getSchema() != null) {
schema = Schema.fromPb(results.getSchema());
} else if (results.getArrowSchema() != null) {
isArrow = true;
try {
arrowSchemaPojo =
org.apache.arrow.vector.ipc.message.MessageSerializer.deserializeSchema(
new org.apache.arrow.vector.ipc.ReadChannel(
new org.apache.arrow.vector.util.ByteArrayReadableSeekableByteChannel(
results.getArrowSchema().decodeSerializedSchema())));
schema = ArrowDeserializer.arrowSchemaToBigQuerySchema(arrowSchemaPojo);
} catch (IOException e) {
throw new BigQueryException(0, "Failed to deserialize Arrow schema from response", e);
}
} else {
schema = null;
}

if (results.getNumDmlAffectedRows() == null && results.getTotalRows() == null) {
numRows = 0L;
} else if (results.getNumDmlAffectedRows() != null) {
Expand All @@ -2094,42 +2258,91 @@ public com.google.api.services.bigquery.model.QueryResponse call()
if (results.getPageToken() != null) {
JobId jobId = JobId.fromPb(results.getJobReference());
String cursor = results.getPageToken();

Iterable<FieldValueList> firstPageRows;
NextPageFetcher<FieldValueList> pageFetcher;

if (isArrow) {
if (results.getArrowRecordBatch() != null) {
try {
firstPageRows =
ArrowDeserializer.deserializeRecordBatch(
results.getArrowRecordBatch().decodeSerializedRecordBatch(),
schema,
arrowSchemaPojo);
} catch (IOException e) {
throw new BigQueryException(0, "Failed to deserialize Arrow record batch", e);
}
} else {
firstPageRows = ImmutableList.of();
}
long initialRowOffset =
firstPageRows instanceof List ? ((List<?>) firstPageRows).size() : 0L;
Comment on lines +2279 to +2280

@gemini-code-assist gemini-code-assist Bot Jul 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

Checking if firstPageRows is an instance of List is restrictive. If ArrowDeserializer.deserializeRecordBatch returns a different type of collection (e.g., a custom Collection or Set), the check will fail and initialRowOffset will default to 0L. Checking for java.util.Collection is safer and more robust.

Suggested change
long initialRowOffset =
firstPageRows instanceof List ? ((List<?>) firstPageRows).size() : 0L;
long initialRowOffset =
firstPageRows instanceof java.util.Collection ? ((java.util.Collection<?>) firstPageRows).size() : 0L;

pageFetcher =
new ArrowQueryPageFetcher(
jobId,
schema,
arrowSchemaPojo,
getOptions(),
initialRowOffset,
null); // Or use maxResults from configuration if available
} else {
firstPageRows =
transformTableData(
results.getRows(), schema, getOptions().getDataFormatOptions().useInt64Timestamp());
pageFetcher = new QueryPageFetcher(jobId, schema, getOptions(), cursor, optionMap(options));
}

return TableResult.newBuilder()
.setSchema(schema)
.setTotalRows(numRows)
.setPageNoSchema(
new PageImpl<>(
// fetch next pages of results
new QueryPageFetcher(jobId, schema, getOptions(), cursor, optionMap(options)),
cursor,
transformTableData(
results.getRows(),
schema,
getOptions().getDataFormatOptions().useInt64Timestamp())))
.setPageNoSchema(new PageImpl<>(pageFetcher, cursor, firstPageRows))
.setJobId(jobId)
.setQueryId(results.getQueryId())
.setJobCreationReason(JobCreationReason.fromPb(results.getJobCreationReason()))
.setRowsInPage(results.getRows() != null ? (long) results.getRows().size() : 0L)
.setRowsInPage(
firstPageRows instanceof List ? (long) ((List<?>) firstPageRows).size() : 0L)
Comment on lines +2303 to +2304

@gemini-code-assist gemini-code-assist Bot Jul 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

Checking if firstPageRows is an instance of List is restrictive. If ArrowDeserializer.deserializeRecordBatch returns a different type of collection (e.g., a custom Collection or Set), the check will fail and rowsInPage will default to 0L. Checking for java.util.Collection is safer and more robust.

Suggested change
.setRowsInPage(
firstPageRows instanceof List ? (long) ((List<?>) firstPageRows).size() : 0L)
.setRowsInPage(
firstPageRows instanceof java.util.Collection ? (long) ((java.util.Collection<?>) firstPageRows).size() : 0L)

.build();
}
// only 1 page of result
Iterable<FieldValueList> firstPageRows;
if (isArrow) {
if (results.getArrowRecordBatch() != null) {
try {
firstPageRows =
ArrowDeserializer.deserializeRecordBatch(
results.getArrowRecordBatch().decodeSerializedRecordBatch(),
schema,
arrowSchemaPojo);
} catch (IOException e) {
throw new BigQueryException(0, "Failed to deserialize Arrow record batch", e);
}
} else {
firstPageRows = ImmutableList.of();
}
} else {
firstPageRows =
transformTableData(
results.getRows(), schema, getOptions().getDataFormatOptions().useInt64Timestamp());
}

return TableResult.newBuilder()
.setSchema(schema)
.setTotalRows(numRows)
.setPageNoSchema(
new PageImpl<>(
new TableDataPageFetcher(null, schema, getOptions(), null, optionMap(options)),
isArrow
? null
: new TableDataPageFetcher(
null, schema, getOptions(), null, optionMap(options)),
null,
transformTableData(
results.getRows(),
schema,
getOptions().getDataFormatOptions().useInt64Timestamp())))
firstPageRows))
// Return the JobID of the successful job
.setJobId(
results.getJobReference() != null ? JobId.fromPb(results.getJobReference()) : null)
.setQueryId(results.getQueryId())
.setJobCreationReason(JobCreationReason.fromPb(results.getJobCreationReason()))
.setRowsInPage(results.getRows() != null ? (long) results.getRows().size() : 0L)
.setRowsInPage(firstPageRows instanceof List ? (long) ((List<?>) firstPageRows).size() : 0L)

@gemini-code-assist gemini-code-assist Bot Jul 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

Checking if firstPageRows is an instance of List is restrictive. If ArrowDeserializer.deserializeRecordBatch returns a different type of collection (e.g., a custom Collection or Set), the check will fail and rowsInPage will default to 0L. Checking for java.util.Collection is safer and more robust.

Suggested change
.setRowsInPage(firstPageRows instanceof List ? (long) ((List<?>) firstPageRows).size() : 0L)
.setRowsInPage(firstPageRows instanceof java.util.Collection ? (long) ((java.util.Collection<?>) firstPageRows).size() : 0L)

.build();
}

Expand Down Expand Up @@ -2203,6 +2416,10 @@ && getOptions().getOpenTelemetryTracer() != null) {

return queryRpc(projectId, content, options);
}
if (configuration.getQueryResultsFormat() == QueryResultsFormat.ARROW) {
throw new IllegalArgumentException(
"Arrow results format is only supported for fast query path execution (e.g. no destination table, no custom clustering, etc.).");
}
return create(JobInfo.of(jobId, configuration), options);
} finally {
if (querySpan != null) {
Expand Down
10 changes: 10 additions & 0 deletions ...query/google-cloud-bigquery/src/main/java/com/google/cloud/bigquery/QueryRequestInfo.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,8 @@ final class QueryRequestInfo {
private final DataFormatOptions formatOptions;
private final String reservation;
private final Long jobTimeoutMs;
private final QueryResultsFormat queryResultsFormat;
private final ArrowSerializationOptions arrowSerializationOptions;

QueryRequestInfo(
QueryJobConfiguration config, com.google.cloud.bigquery.DataFormatOptions dataFormatOptions) {
Expand All @@ -66,6 +68,8 @@ final class QueryRequestInfo {
this.formatOptions = dataFormatOptions.toPb();
this.reservation = config.getReservation();
this.jobTimeoutMs = config.getJobTimeoutMs();
this.queryResultsFormat = config.getQueryResultsFormat();
this.arrowSerializationOptions = config.getArrowSerializationOptions();
}

/**
Expand Down Expand Up @@ -142,6 +146,12 @@ QueryRequest toPb() {
if (jobTimeoutMs != null) {
request.setJobTimeoutMs(jobTimeoutMs);
}
if (queryResultsFormat != null) {
request.setQueryResultsFormat(queryResultsFormat.toString());
}
if (arrowSerializationOptions != null) {
request.setArrowSerializationOptions(arrowSerializationOptions.toPb());
}
return request;
}

Expand Down
Loading