hudi-agent commented on code in PR #18065:
URL: https://github.com/apache/hudi/pull/18065#discussion_r3371870053
##########
hudi-hadoop-common/src/main/java/org/apache/hudi/avro/HoodieAvroWriteSupport.java:
##########
@@ -20,38 +20,307 @@
package org.apache.hudi.avro;
import org.apache.hudi.common.bloom.BloomFilter;
+import org.apache.hudi.common.config.HoodieConfig;
+import org.apache.hudi.common.config.HoodieStorageConfig;
import org.apache.hudi.common.schema.HoodieSchema;
+import org.apache.hudi.common.schema.HoodieSchemaField;
+import org.apache.hudi.common.schema.HoodieSchemaType;
+import org.apache.hudi.common.schema.HoodieSchemaUtils;
import org.apache.hudi.common.util.CollectionUtils;
import org.apache.hudi.common.util.Option;
+import org.apache.hudi.common.util.ReflectionUtils;
import org.apache.hudi.common.util.StringUtils;
+import org.apache.hudi.exception.HoodieException;
+import org.apache.avro.Schema;
+import org.apache.avro.generic.GenericData;
+import org.apache.avro.generic.GenericRecord;
+import org.apache.avro.generic.IndexedRecord;
import org.apache.parquet.avro.AvroWriteSupport;
import org.apache.parquet.hadoop.api.WriteSupport;
import org.apache.parquet.schema.MessageType;
+import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
+import java.util.LinkedHashMap;
+import java.util.List;
import java.util.Map;
import java.util.Properties;
+import java.util.regex.Matcher;
+import java.util.regex.Pattern;
+
+import static
org.apache.hudi.common.config.HoodieStorageConfig.PARQUET_VARIANT_FORCE_SHREDDING_SCHEMA_FOR_TEST;
+import static
org.apache.hudi.common.config.HoodieStorageConfig.PARQUET_VARIANT_SHREDDING_PROVIDER_CLASS;
+import static
org.apache.hudi.common.config.HoodieStorageConfig.PARQUET_VARIANT_WRITE_SHREDDING_ENABLED;
/**
- * Wrap AvroWriterSupport for plugging in the bloom filter.
+ * Wrap AvroWriterSupport for plugging in the bloom filter and variant
shredding support.
+ *
+ * <p>When variant columns are configured for shredding (via {@link
HoodieSchema.Variant#isShredded()}),
+ * this class transforms variant records at write time to populate {@code
typed_value} columns
+ * by parsing variant binary data using a {@link VariantShreddingProvider}
loaded via reflection.</p>
*/
public class HoodieAvroWriteSupport<T> extends AvroWriteSupport<T> {
private final Option<HoodieBloomFilterWriteSupport<String>>
bloomFilterWriteSupportOpt;
private final Map<String, String> footerMetadata = new HashMap<>();
protected final Properties properties;
+ /**
+ * Whether variant write shredding is enabled via config.
+ */
+ private final boolean variantWriteShreddingEnabled;
+
+ /**
+ * The effective (possibly shredded) HoodieSchema used for writing.
+ */
+ private final HoodieSchema effectiveHoodieSchema;
+
+ /**
+ * The effective Avro schema (derived from effectiveHoodieSchema).
+ */
+ private final Schema effectiveAvroSchema;
+
+ /**
+ * Indices of top-level variant fields that need shredding transformation.
+ * Empty array if no shredding is needed.
+ */
+ private final int[] shreddedVariantFieldIndices;
Review Comment:
🤖 nit: the three parallel arrays `shreddedVariantFieldIndices` /
`shreddedVariantAvroSchemas` / `shreddedVariantHoodieSchemas` (plus the linear
`findVariantIndex` lookup in the hot `write()` path) would read more cleanly as
a single `Map<Integer, ShreddedVariantField>` keyed by effective field index,
where `ShreddedVariantField` bundles the Avro and Hoodie schemas. That removes
the parallel-array invariant readers have to keep in their head and turns the
per-field lookup into a single `get` call.
<sub><i>- AI-generated; verify before applying. React 👍/👎 to flag
quality.</i></sub>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
