Skip to content

feat(sync): Map BLOB type to struct in Hive and BigQuery sync #18482

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

Open
voonhous wants to merge 2 commits into
base: master
Choose a base branch
from
Open

feat(sync): Map BLOB type to struct in Hive and BigQuery sync #18482

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
39e24c8
feat(sync): Map BLOB type to struct in Hive and BigQuery sync
voonhous Apr 8, 2026
4581c7f
Add test for nullable and nested blobs
voonhous Apr 13, 2026
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
1 change: 1 addition & 0 deletions hudi-gcp/src/main/java/org/apache/hudi/gcp/bigquery/BigQuerySchemaResolver.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -147,6 +147,7 @@ private Field getField(HoodieSchema fieldSchema, String name, boolean nullable)
standardSQLTypeName = StandardSQLTypeName.NUMERIC;
break;
case RECORD:
case BLOB:
return Field.newBuilder(name, StandardSQLTypeName.STRUCT,
FieldList.of(getFields(fieldSchema))).setMode(fieldMode).build();
case ARRAY:
Expand Down
56 changes: 56 additions & 0 deletions hudi-gcp/src/test/java/org/apache/hudi/gcp/bigquery/TestBigQuerySchemaResolver.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -281,6 +281,62 @@ void getTableSchema_withoutPartitionFields() throws Exception {
Assertions.assertEquals(PRIMITIVE_TYPES_BQ_SCHEMA, resolver.getTableSchema(mockMetaClient, Collections.emptyList()));
}

@Test
void convertSchema_blobField() {
Copy link
Copy Markdown
Contributor

@bvaradar bvaradar Apr 9, 2026

Choose a reason for hiding this comment

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

minor: can we use camel case here

Copy link
Copy Markdown
Member Author

@voonhous voonhous Apr 13, 2026 •
edited
Loading

Choose a reason for hiding this comment

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

I see other tests in this style using a mix of camel and snake case, so let's keep it as is.

HoodieSchema input = HoodieSchema.createRecord("testRecord", null, null, false, Arrays.asList(
HoodieSchemaField.of("id", HoodieSchema.create(HoodieSchemaType.INT)),
HoodieSchemaField.of("blob_data", HoodieSchema.createBlob())
));

Field expectedBlobField = Field.newBuilder("blob_data", StandardSQLTypeName.STRUCT,
Field.newBuilder("type", StandardSQLTypeName.STRING).setMode(Field.Mode.REQUIRED).build(),
Field.newBuilder("data", StandardSQLTypeName.BYTES).setMode(Field.Mode.NULLABLE).build(),
Field.newBuilder("reference", StandardSQLTypeName.STRUCT,
Field.newBuilder("external_path", StandardSQLTypeName.STRING).setMode(Field.Mode.REQUIRED).build(),
Field.newBuilder("offset", StandardSQLTypeName.INT64).setMode(Field.Mode.NULLABLE).build(),
Field.newBuilder("length", StandardSQLTypeName.INT64).setMode(Field.Mode.NULLABLE).build(),
Field.newBuilder("managed", StandardSQLTypeName.BOOL).setMode(Field.Mode.REQUIRED).build())
.setMode(Field.Mode.NULLABLE).build())
.setMode(Field.Mode.REQUIRED).build();

Schema result = SCHEMA_RESOLVER.convertSchema(input);
Assertions.assertEquals(2, result.getFields().size());
Assertions.assertEquals(expectedBlobField, result.getFields().get(1));
}

@Test
void convertSchema_nullableBlobField() {
HoodieSchema input = HoodieSchema.createRecord("testRecord", null, null, false, Arrays.asList(
HoodieSchemaField.of("id", HoodieSchema.create(HoodieSchemaType.INT)),
HoodieSchemaField.of("blob_data", HoodieSchema.createNullable(HoodieSchema.createBlob()))
));

Schema result = SCHEMA_RESOLVER.convertSchema(input);
Field blobField = result.getFields().get(1);
Assertions.assertEquals(Field.Mode.NULLABLE, blobField.getMode());
Assertions.assertEquals(StandardSQLTypeName.STRUCT, blobField.getType().getStandardType());
}

@Test
void convertSchema_nestedBlobField() {
HoodieSchema inner = HoodieSchema.createRecord("media", null, null, false, Arrays.asList(
HoodieSchemaField.of("title", HoodieSchema.create(HoodieSchemaType.STRING)),
HoodieSchemaField.of("content", HoodieSchema.createBlob())
));
HoodieSchema input = HoodieSchema.createRecord("testRecord", null, null, false, Arrays.asList(
HoodieSchemaField.of("id", HoodieSchema.create(HoodieSchemaType.INT)),
HoodieSchemaField.of("media", inner)
));

Schema result = SCHEMA_RESOLVER.convertSchema(input);
Field mediaField = result.getFields().get(1);
Assertions.assertEquals(StandardSQLTypeName.STRUCT, mediaField.getType().getStandardType());
// Verify the nested "content" sub-field is a STRUCT (BLOB)
Field contentField = mediaField.getSubFields().stream()
.filter(f -> f.getName().equals("content")).findFirst().get();
Assertions.assertEquals(StandardSQLTypeName.STRUCT, contentField.getType().getStandardType());
}

@Test
void convertSchema_vectorField() {
HoodieSchema input = HoodieSchema.createRecord("testRecord", null, null, false, Arrays.asList(
Expand Down
1 change: 1 addition & 0 deletions hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/util/HiveSchemaUtil.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -249,6 +249,7 @@ private static String convertField(final HoodieSchema fieldSchema, boolean suppo
HoodieSchema valueType = nonNullType.getValueType();
return createHiveMap(convertField(keyType, supportTimestamp, doFormat), convertField(valueType, supportTimestamp, doFormat), doFormat);
case RECORD:
case BLOB:
return createHiveStruct(nonNullType.getFields(), supportTimestamp, doFormat);
default:
throw new UnsupportedOperationException("Cannot convert type " + nonNullType.getType());
Expand Down
14 changes: 13 additions & 1 deletion hudi-sync/hudi-hive-sync/src/test/java/org/apache/hudi/hive/util/TestHiveSchemaUtil.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,7 +149,13 @@ void testSchemaTypeConversion() throws IOException {
HoodieSchemaField.of("time_micros_field", HoodieSchema.createTimeMicros()),
HoodieSchemaField.of("decimal_field", HoodieSchema.createDecimal(10, 2)),
HoodieSchemaField.of("uuid_field", HoodieSchema.create(HoodieSchemaType.UUID)),
HoodieSchemaField.of("vector_field", HoodieSchema.createVector(128))
HoodieSchemaField.of("vector_field", HoodieSchema.createVector(128)),
HoodieSchemaField.of("blob_field", HoodieSchema.createBlob()),
HoodieSchemaField.of("nullable_blob_field", HoodieSchema.createNullable(HoodieSchema.createBlob())),
HoodieSchemaField.of("nested_blob_field", HoodieSchema.createRecord("media", null, null, false, Arrays.asList(
HoodieSchemaField.of("title", HoodieSchema.create(HoodieSchemaType.STRING)),
HoodieSchemaField.of("content", HoodieSchema.createBlob())
)))
)
);

Expand All @@ -174,6 +180,12 @@ void testSchemaTypeConversion() throws IOException {
expected.put("`decimal_field`", "DECIMAL(10 , 2)");
expected.put("`uuid_field`", "binary");
expected.put("`vector_field`", "binary");
expected.put("`blob_field`",
"STRUCT< `type` : string, `data` : binary, `reference` : STRUCT< `external_path` : string, `offset` : bigint, `length` : bigint, `managed` : boolean>>");
expected.put("`nullable_blob_field`",
"STRUCT< `type` : string, `data` : binary, `reference` : STRUCT< `external_path` : string, `offset` : bigint, `length` : bigint, `managed` : boolean>>");
expected.put("`nested_blob_field`",
"STRUCT< `title` : string, `content` : STRUCT< `type` : string, `data` : binary, `reference` : STRUCT< `external_path` : string, `offset` : bigint, `length` : bigint, `managed` : boolean>>>");
assertEquals(expected, actual);
}
}
Loading