hudi-agent commented on code in PR #18384:
URL: https://github.com/apache/hudi/pull/18384#discussion_r3429660667
##########
hudi-client/hudi-client-common/src/test/java/org/apache/hudi/testutils/HoodieWriteableTestTable.java:
##########
@@ -123,7 +123,7 @@ public StoragePath withInserts(String partition, String
fileId, List<HoodieRecor
storage.getConf(),
Double.parseDouble(HoodieStorageConfig.PARQUET_COMPRESSION_RATIO_FRACTION.defaultValue()),
true);
try (HoodieAvroParquetWriter writer = new HoodieAvroParquetWriter(
new StoragePath(Paths.get(basePath, partition,
fileName).toString()), config, currentInstantTime,
- contextSupplier, populateMetaFields)) {
+ contextSupplier, new HoodieTableConfig().getHoodieMetaFieldFlags()))
{
Review Comment:
🤖 With the writer constructor no longer taking the `populateMetaFields`
boolean, the only signal the writer sees is `metaFieldFlags`, and `new
HoodieTableConfig().getHoodieMetaFieldFlags()` is always `ALL_POPULATED`. For
tests that build this utility over a metaClient with `populateMetaFields=false`
and pass a non-null BloomFilter (e.g. `TestHoodieBloomIndex` line 293/637), the
writer's `writeAvro` will now hit `isAnyPopulated()==true` and call
`writeSupport.add(key)`, populating the bloom filter — which the old
`populateMetaFields=false` arg used to suppress. Could this be switched to
`getMetaClient().getTableConfig().getHoodieMetaFieldFlags()` (and the same for
the ORC path at line 149) so the writer's view matches the outer
`populateMetaFields` field on line 73?
<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]
