[TRACKING] Open PR backlog triage: verdicts for open PRs (close / convert-to-issue / keep)

[voonhous]

Tracking issue for a full triage of the open PR backlog. All 555 open PRs as of 2026-07-16, from #2607 (2021-02-26) to #19305 (2026-07-16), were reviewed against master at e098da7 (2026-07-15) to decide what can be closed, what should be converted into an issue, and what is still viable.

The full per-PR checklists, with a one-line reason for every verdict, are in the first three comments below. Each entry has a checkbox -- tick it as the PR gets addressed (GitHub shows an x-of-y task progress bar per comment):

• Comment 1: CLOSE (166)
• Comment 2: CONVERT to issue, then close (166)
• Comment 3: KEEP (223)

These verdicts are recommendations, not decisions. If you are the author or a reviewer of a PR listed under CLOSE or CONVERT and believe it should stay open, please comment here or on the PR.

Scope and method

• Covers every open apache/hudi PR as of 2026-07-16, from [HUDI-1643] Hudi observability - framework to report stats from execu… #2607 (2021-02-26) to
  [HUDI-9666] Auto-deduce ComplexKeyGenerator single-field record-key encoding on branch-0.x #19305 (2026-07-16).
• Every PR was assessed against master at e098da7 (2026-07-15): whether the
  touched files still exist, whether an equivalent fix or feature already landed
  (the landing PR/commit is named in the reason where found), and whether the
  subsystem has since been rewritten (1.0 timeline/FileGroupReader rewrites,
  hive-sync rewrite, removed engine version poms, etc.).
• Branch policy used: release-0.x/1.0.x and branch-0.x are frozen -- if the fix
  also applies to master the verdict is CONVERT, otherwise CLOSE (active
  committer backports were kept case-by-case). release-1.2.0 is the active
  upcoming release branch and was judged on activity. asf-site PRs were judged
  on whether the docs change is still accurate for the current site.
  feature-branch-rfc100-unstructured-data is treated as active.
• Recent PRs (roughly May-Jul 2026) in normal review flow default to KEEP;
  [DNM] tags were read in context (stacking/merge-gate markers vs abandoned
  experiments).

Verdicts

• CLOSE (166): obsolete -- already landed or superseded on master, code removed or rewritten, dead target branch, or abandoned after blocking review feedback.
• CONVERT (166): the underlying bug/feature is still relevant on current master, but the PR is too stale to rescue. File a GitHub issue capturing the scope, then close the PR.
• KEEP (223): still viable -- applies (or nearly applies) to master, or has active author/committer engagement; worth reviewing or rebasing as-is.

Summary by year created

Year	CLOSE	CONVERT	KEEP	Total
2021	4	1	0	5
2022	25	39	1	65
2023	22	42	6	70
2024	31	24	5	60
2025	54	46	31	131
2026	30	14	180	224
Total	166	166	223	555

Borderline calls flagged during triage

These verdicts were judgment calls; a maintainer close to the area should double-check
before acting.

From the oldest 400:

• [HUDI-2281] Add metadata client APIs to fetch list of data files and … #3417: the only 2021-vintage PR that still merges cleanly, but zero review traction since 2021 (bot bumps only); left as CONVERT rather than KEEP.
• [HUDI-2733] Add support for Thrift sync #4665 (thrift-mode hive sync): CLOSE because HoodieHiveClient was removed in the hive-sync rewrite; arguable CONVERT if thrift mode is still wanted.
• [HUDI-4194] upgrade confluent version to 7.1.1 #5758 vs fix: increase confluent version #18445 (confluent bump): master is still on confluent 5.5.0, so the upgrade is genuinely outstanding; the 2022 PR is CONVERT, the 2026 one (fix: increase confluent version #18445) is KEEP.
• [HUDI-4578] Reduce the scope and duration of holding checkpoint lock … #6337, [HUDI-4629] Create hive table from existing hoodie table failed when the table schema is not defined #6409, [HUDI-4647] Keep the hive sync settings in spark sql consistent #6448, [HUDI-4824]RANGE BUCKET index, base logic and test. #6858: could not fully confirm landed-vs-still-broken; marked CONVERT so scope is not lost, verify before filing issues.
• [HUDI-5548] spark sql show | update hudi's table properties #7680 (show/alter table properties): 3 approvals, but alter-props partially landed via AlterTableCommand; CONVERT covers only the remaining "show properties" scope.
• [HUDI-5560][RFC-42] Make Consistent hash index Bucket Resizing more available… #7748 (RFC-42 bucket-resizing amendment): doc-only change on a merged RFC; CONVERT, but CLOSE is defensible.
• [HUDI-6194] prevent flink writer getting the wrong instant to write #8673 (Flink wrong-instant race): the coordinator was reworked with pending-instant recommit logic, so the race may already be fixed; verify before converting.
• [HUDI-6528] Fix premature RDD unpersist during index lookup #9188 (index perf fix): draft, but committer-updated 2025-12, hence KEEP.
• [HUDI-6760] Add SelfDescribingInputFormatInterface for hive FileInput… #9554: unclear whether the spark-schema-evolution-then-hive-read bug survives the refactored hadoop-mr path; CONVERT conservatively.
• Incoming batch schema is not compatible with the table's one #9980 #10308, [HUDI-7176] Add unit tests for log file only file slices #10396, [HUDI-5026] Fix ORC bootstrap read support in Spark #10406: niche / stacked-on-unmerged / semi-legacy-bootstrap; all CONVERT but each is a defensible CLOSE.
• [HUDI-7634] Rename HoodieStorage APIs #11047: approved cosmetic rename, 2 years stale; CLOSE, though a maintainer could rebase it.
• [HUDI-6510] Support GHCI on Java 17 #11573 (Java 17 CI): CONVERT -- bot.yml is still JDK 11.
• [HUDI-8024] Test secondary index updates and rollback #11706 (SI rollback/compaction test draft): CLOSE unless that coverage gap is confirmed still open.
• [HUDI-8163] Refactor UnMergedLogHandler with iterators #12608 (UnMerged log refactor): CLOSE as an abandoned pure-refactor with 16 unaddressed review rounds; CONVERT if the refactor is still desired.
• [HUDI-8553] Support writing record positions to log blocks from Spark SQL UPDATE and DELETE statements #12732 (record positions from Spark SQL): core position-write infra landed via [HUDI-6795] Implement writing record_positions to log blocks for updates and deletes #9581; the remaining SQL-DML gap may be partly covered already.
• [HUDI-8945] Rename merger classes to be consistent with merge modes #12747 (merger rename): KEEP because yihua rebased it in 2026, but CI is failing and rename refactors rot fast.
• [HUDI-9000] Adding timestamp ordering validation for flink 0.x branch #12824 (branch-0.x timestamp ordering): approved for 0.x; CLOSE assumes no further 0.x release is planned.
• [HUDI-9187][HUDI-6249] Update usage of read/write in FileSystemView code paths #12982: reached an approval before being abandoned on a since-refactored FSView subsystem; CLOSE.
• [HUDI-9442] Improve key lookup in metadata table #13348 (MDT key-lookup perf): CLOSE as superseded-by-own-author rework, but the perf goal itself is arguably still open.
• [HUDI-9560][RFC-97] Deprecate Hudi Payload Class Usage #13499 (RFC-97 payload deprecation doc): KEEP, though the deprecation work is landing on master without it.
• [HUDI-9610] Support to get property from catalog for copy_to_temp_view call #13582, [HUDI-9734] Improve Hudi tooling to support file groups and specific partitions procedures #13745: small approved-or-applied changes stalled most of a year; KEEP vs CONVERT is a coin flip.
• Add functional compat tests for Incremental Source #13954, test: add functional payload test cases for v6 (OverwriteWithLatestAvroPayload, DefaultHoodieRecordPayload) #14041, test: Add coverage for with OverwriteWithLatestAvroPayload with ordering fields set #14051 (test-coverage PRs): CONVERT to preserve coverage intent; given the project's test-bloat concerns, plain CLOSE is defensible.
• Ensure MOR table works, with lance base files and avro logs file #14154, Validate that basic compaction and clustering works with Lance  #14156, Add lance schema evolution support for add column #14160, Validate vector and blob content support #14193 (Lance RFC-100 stack): all CONFLICTING and mostly idle since late 2025; kept because feature-branch-rfc100-unstructured-data is active and they depend on the unmerged reader PR Support COW bulk-insert, insert, upsert, delete works with spark datasource and lance #14137.
• feat: Support schema on write for flink #14248 (Flink schema-on-write): large and CONFLICTING, dormant since Dec 2025, but active committer author; KEEP.
• chore(ci): Update Java 11 image for Azure CI #17692: trivial CI image bump by a PMC member; called CLOSE as stale, could still be wanted.
• MDT Test framework without writing data files #17796 (XL MDT test/benchmark framework): heavy review investment but stalled ~5 months; KEEP, could be CONVERT if dropped.
• feat(blob): Blob cleaner updates #18159 (blob cleaner): KEEP, but the RFC-100 plan was restructured by merged feat(blob): update approach to remove reliance on column groups, break down plan #18013 and cleaner work continues on a feature branch, so it may effectively be superseded.
• chore(deps): bump io.airlift:aircompressor from 0.27 to 2.0.3 #18247 (aircompressor 0.27 -> 2.0.3 CVE bump): KEEP since the CVE is unaddressed, but the major-version jump likely needs manual API migration.
• (WIP) feat: Add Claude Code Skills for Hudi #18338 (in-repo Claude Code skills): mergeable and active; whether .claude/ belongs in-repo is a maintainer call.
• [HUDI-7946][RFC-90] Add support for cancellable clustering table service plans #18346 (RFC-90 re-post): zero engagement; CONVERT borders on CLOSE.
• feat: Provide new TransactionManager functionality and move instant time management to Transactionmanager #18350, feat: Support selective meta field population #18384: recent committer PRs buried under 90-100 file conflicts; KEEP, but CONVERT is reasonable.

From the newest 155:

• feat(core): implement file-level sorting for upsert (#18357) #18504 (RFC-103 LSM phase 1): 3 months stale and conflicting, but blocked on a design question the author posed to the maintainers -- ball is in their court; KEEP.
• Log structured schema-change diffs in DeltaStreamer #18655 (SchemaChangeLogger): closest CONVERT-vs-KEEP call in the tail; stale unreviewed draft, scope still relevant.
• feat(rollback): add RollbackOrphanDetector utility for archive orphan guard (#18783) #18795 (RollbackOrphanDetector): KEEP, but maintainers could not reproduce the underlying issue, so it may end up rejected.
• test(spark): repro for reconcile.schema=true blocking documented type promotion #18806, test(spark): codify MDT col_stats behavior on ADD COLUMN schema evolution #18807 vs test(spark): repro for bytes→string promotion crashing data-skipping read #18810 (test-only repro PRs): split on engagement -- the first two have committer review (KEEP), test(spark): repro for bytes→string promotion crashing data-skipping read #18810 has none plus red CI (CONVERT).
• chore(ci): modernize Codecov config for the coverage initiative #19138, [DNM][1.2] Add versioned binary list state for flink writer #19228, fix(DNM): adjust the native log format metadata as separate entries in footer #19247, [DNM, stacked] fix(storage-format): fix multi-block previous-block offset and add byte-level HFile writer tests #19083 ([DNM]-tagged): DNM read as stacking/merge-gate markers on active work, not abandonment; all KEEP. [DNM][0.14] Add versioned binary list state for flink writer #19229 by contrast was DNM onto frozen release-0.14.1 with zero reviews (CLOSE).
• fix(meta-sync): Normalize meta sync base path to avoid consecutive/trailing slashes in table LOCATION #19248, [HUDI-9666] Auto-deduce ComplexKeyGenerator single-field record-key encoding on branch-0.x #19305 (branch-0.x): kept despite the freeze rule -- branch-0.x still merged fixes in Apr 2026 and both are active committer backports; a master port of fix(meta-sync): Normalize meta sync base path to avoid consecutive/trailing slashes in table LOCATION #19248's fix may also be needed.
• docs(blog): explainer wave 2: streaming data lake, CDC, lakehouse comparisons #19270-docs(blog): streaming, write-heavy, read-heavy and benchmark comparisons #19273: stacked asf-site blog series with identical create timestamps; all four kept as active drafts rather than treating later ones as superseding earlier.

[voonhous]

CLOSE (166)

Tick the box once the PR has been closed (or the verdict overturned -- note it in a reply).

• #2607 [HUDI-1643] Hudi observability - framework to report stats from execu... (nbalajee, 2021-02-26, metrics): DistributedMetricsReporter/HoodieDistributedMetrics never landed; PR conflicting + changes-requested, abandoned since 2021
• #3515 [HUDI-2341] Adding blog on immutable data lakes (nsivabalan, 2021-08-20, asf-site, docs): asf-site blog never merged, changes-requested and abandoned since 2021
• #3549 [HUDI-2369] Blog on bulk_insert sort modes (nsivabalan, 2021-08-27, asf-site, docs): asf-site blog never merged, changes-requested and abandoned since 2021
• #4481 [HUDI-3075] [WIP] - Docs for Debezium DeltaStreamer Source (kywe665, 2021-12-31, asf-site, docs): WIP asf-site debezium docs abandoned since 2021; hoodie_deltastreamer.md doc long since restructured
• #4665 [HUDI-2733] Add support for Thrift sync (stym06, 2022-01-21, hive-sync): HoodieHiveClient removed in hive-sync rewrite; thrift sync mode niche, changes-requested, abandoned
• #4718 [HUDI-3345][RFC-36] Hudi metastore server (minihippo, 2022-01-29, metaserver): RFC-36 realized as hudi-metaserver module (same author); RFC doc PR never merged and now superseded
• #5057 [HUDI-3651] optimize the hoodie hive client and ddl executor code wit... (JerryYue-M, 2022-03-17, hive-sync): refactor targets HoodieHiveClient which was removed in hive-sync rewrite; cleanup now moot
• #5269 [HUDI-3636] Create new write clients for async table services in DeltaStreamer and Spark streaming sink (yihua, 2022-04-09, deltastreamer): changes-requested, untouched since 2022; async table-service and StreamSync architecture heavily rewritten since
• #5554 [HUDI-4082][HUDI-FLINK]enhance the readability of hudi operator names... (cuibo01, 2022-05-11, flink): cosmetic Flink operator-name change; low value, 4yr stale, pipeline (Pipelines.java) has evolved since
• #5831 [WIP] [HUDI-4220] Created an utility to convert JSON to Avro (vingov, 2022-06-10, utilities): WIP JsonAvroConverter never landed; niche utility, abandoned since 2022
• #5926 [HUDI-3475] Initialize hudi table management module (yuzhaojing, 2022-06-21, table-service-manager): HoodieTableServiceManagerClient stub landed but standalone TSM server module never merged; effort stalled, absent from master
• #6045 [WIP][HUDI-4345] Enable partition pruning and projection for incremental query on COW table (yihua, 2022-07-05, spark): WIP; CopyOnWriteIncrementalRelation removed in 1.x incremental-query rewrite, diff no longer applies
• #6087 [HUDI-4364]: changes for integrating column stats index into presto-h... (pratyakshsharma, 2022-07-12, query-engine): draft; presto connector replaced by hudi-trino-plugin, ColumnDomain classes removed, col-stats index landed separately
• #6108 WIP:optimize the bucket bootstrap (JerryYue-M, 2022-07-14, flink): Abandoned WIP since 2022, no description and zero review; BucketStreamWriteFunction still on master but optimization scope undefined
• #6253 [HUDI-4511] Create write client every compact in HoodieFlinkCompactor (yuzhaojing, 2022-07-30, flink): HoodieFlinkCompactor rewritten (createWriteClientV2 + explicit close + txn manager); 2022 per-compact-client diff no longer applies
• #6345 [HUDI-4552][RFC-58] Integrate column stats index with all query engines (pratyakshsharma, 2022-08-09, rfc): RFC-58 doc never merged (absent on master); column stats index landed via MDT col_stats + data skipping
• #6382 [HUDI-4612][RFC-59] RFC-59 Materials: Multiple event_time Fields Latest Verification in a Single Table (XinyaoTian, 2022-08-12, rfc): RFC-59 proposal unadopted (absent on master), abandoned since 2022; multiple-event_time feature never pursued
• #6576 [HUDI-4678][RFC-61] RFC for Snapshot view management (fengjian428, 2022-09-02, rfc): RFC-61 snapshot view management doc never merged (absent on master); not pursued, abandoned since 2022
• #6600 [RFC-62] Diagnostic Reporter (zhangyue19921010, 2022-09-05, rfc): RFC-62 Diagnostic Reporter doc unadopted (absent on master), abandoned since 2022
• #6737 [HUDI-4373] Flink Consistent hashing bucket index write path code (YuweiXiao, 2022-09-22, flink): Flink consistent-hashing bucket write path already on master (ConsistentBucketStreamWriteFunction, FlinkConsistentBucketClusteringPlanStrategy); landed via later PR
• #6902 [HUDI-3529][RFC-67] Hudi bundle standards (xushiyan, 2022-10-10, rfc): RFC-67 bundle standards doc never merged (absent on master); bundling standardized organically, abandoned since 2022
• #6905 [HUDI-5001] column name sanitization for row source (vamshigv, 2022-10-10, deltastreamer): Column-name sanitization landed (SANITIZE_SCHEMA_FIELD_NAMES config + SanitizationUtils on master)
• #6926 [HUDI-3676] Enhance tests for trigger clean every Nth commit (pratyakshsharma, 2022-10-11, table-service): Coverage already on master; TestCleanPlanExecutor exercises withMaxCommitsBeforeCleaning
• #6963 [WIP][HUDI-5040] flink multi writer for bucket index (XuQianJin-Stars, 2022-10-16, flink): Abandoned WIP; touches timeline/log-format/OCC core rewritten since 2022, diff meaningless
• #7080 [HUDI-4142][RFC-64] New APIs to facilitate faster Query Engine integrations (alexeykudinkin, 2022-10-28, rfc): RFC-64 doc never merged (no rfc-64 on master); the faster query-engine reader APIs shipped as HoodieFileGroupReader in 1.0. Superseded by shipped implementation.
• #7203 [HUDI-5208] Enable Spark to support OPERATION_METADATA_FIELD (dongkelun, 2022-11-15, spark-sql): Enable Spark _hoodie_operation metadata field; danny noted Spark already resolves the field. Already supported on master.
• #7359 [HUDI-3304] Allow selective partial update (bschell, 2022-12-01, writer): Selective partial update; author-abandoned, continuation opened as [HUDI-3304] Add support for selective partial update #9979, and partial-update rewritten on master (PartialUpdateMode/PartialUpdateHandler). Superseded.
• #7415 [HUDI-5355] support spark 3.2+ hudi table query with tvf (scxwhite, 2022-12-09, spark-sql): Spark 3.2+ TVF query; TVF support already landed on master (TableValuedFunctions.scala plus multiple Hoodie*TableValuedFunction). Superseded, and PR is CONFLICTING.
• #7456 [HUDI-4917][FOLLOW_UP]Optimize codes logic to not break the old class meaning (boneanxs, 2022-12-14, index): Follow-up code cleanup on bloom-index read handles; no review traction, abandoned since 2022. Low-value refactor, not worth rescuing.
• #7740 [HUDI-5464] Suffix instant time for metadata partition init deltacommit (xushiyan, 2023-01-24, metadata-table): Suffix instant time for MDT partition-init deltacommit; danny: suffix removed generally in 1.x so not needed. Obsolete for the 1.x timeline.
• #7929 [HUDI-5754] Add new sources to deltastreamer docs (codope, 2023-02-13, asf-site, docs): targets renamed file (hoodie_deltastreamer.md now hoodie_streaming_ingestion.md); Proto/Pulsar/GCS sources shipped in 0.13 and long documented
• #7966 [HUDI-5806] {W-I-P] hudi-cli option to find nearest matching commits (pramodbiligiri, 2023-02-15, cli): draft W-I-P, Impact/Risk left as TODO, abandoned since 2024; low-value CLI convenience flag never finished
• #8024 [MINOR] Improve RollbackToInstantTimeProcedure (XuQianJin-Stars, 2023-02-23, spark-sql): RollbackToInstantTimeProcedure on master already takes required instant_time param; ability to roll back a specified commit effectively exists
• #8478 [HUDI-6086] Improve HiveSchemaUtil#generateCreateDDL With StringBuilder (slfan1989, 2023-04-17, hive-sync): pure readability refactor, no functional change; abandoned after 12 review rounds with danny0405 since 2024; not worth reviving
• #8513 Adding multiple stream writing support over 0.12.1 (nsivabalan, 2023-04-20, release-0.12.1, spark): targets frozen release-0.12.1 branch; multi-stream/structured-streaming sink already on master; no master relevance
• #8598 [HUDI-6143] use the startoffset of each logfile when the downstream tasks read the logfile (ThinkerLei, 2023-04-28, flink): log record reader rewritten in 1.x; AbstractHoodieLogRecordReader removed from master, diff no longer applies; no reviews, abandoned 2024
• #8719 [HUDI-6215] Making snappy as default compression codec for parquet (nsivabalan, 2023-05-16, core-config): contentious default flip; parquet default deliberately still gzip on master; one-line config PR, no reviews, abandoned 2024
• #8724 [HUDI-6220] Add HUDI code version to commit files and hoodie.properties. (prashantwason, 2023-05-16, write-client): Version-in-commit-metadata landed via feat: Add HUDI version and engine properties to commit metadata #18183 (cde86ff); HoodieVersion.java already on master
• #8842 [HUDI-6283] Improve ReflectionUtils APIs (xushiyan, 2023-05-29, common): ReflectionUtils moved to hudi-io and already caches classes via ConcurrentHashMap; caffeine/rename change superseded
• #8911 [Hudi-8882] Compatible with versions below hive 2.3 to read hudi rt table (thomasg19930417, 2023-06-09, hive): Adds hive <2.3 read shims; Hudi 1.x requires hive 2.3+, contentious (33 review comments, no approval) and abandoned 2024-02
• #9074 [MINOR] Upload coverage report to codecov. (prashantwason, 2023-06-28, ci): Codecov already wired: .codecov.yml present and .github/workflows/bot.yml references codecov; MINOR PR abandoned 2024-02
• #9111 Add CodeQL workflow (jorgectf, 2023-07-03, ci): codeql.yml never added to master; generic GitHub Security Lab bot PR abandoned 2024-02, not adopted
• #9226 [HUDI-6352] take actual commit time (StateTransitionTime) into consideration for clustering archival (hbgstc123, 2023-07-19, table-service): Archive-before-clean for clustering now guarded by rewritten ClusteringUtils.getEarliestInstantToRetainForClustering; method the PR patched is gone
• #9235 [RFC-72] Redesign Hudi Spark Integration (jonvex, 2023-07-19, rfc): RFC-72 spark redesign shipped in 1.0 (new file group reader); rfc-72.md never merged, PR abandoned 2024-02
• #9485 [HUDI-6730] Enable hoodie configuration using the --conf option with the "spark." prefix (wecharyu, 2023-08-19, spark-sql): spark.hoodie.* conf prefix already supported on master (HoodieSqlCommonUtils filters spark.hoodie., ProvidesHoodieConfig normalizes)
• #9564 [HUDI-6712] Add Parquet file metadata loader (linliu-code, 2023-08-28, storage): ParquetFileMetadataLoader/keyed-lookup foundation never landed or used (absent on master); speculative, CHANGES_REQUESTED, abandoned 2024-02
• #9762 [HUDI-6888] [WIP] Optimize scanInternalV2 in AbstractHoodieLogRecordReader (codope, 2023-09-21, log-reader): WIP abandoned; AbstractHoodieLogRecordReader and scanInternalV2 removed from master in 1.0 FileGroupReader rewrite
• #9767 [MINOR][DOC] Update index type desc (leosanqing, 2023-09-22, asf-site, docs): stale asf-site indexing.md tweak; author noted overlap already merged in 2023, indexing docs since rewritten for 1.0
• #9899 [HUDI-6967] Add clearJobStatus api in HoodieEngineContext (wecharyu, 2023-10-22, spark-client): already on master: clearJobStatus exists in Spark/Flink/Java EngineContext and is called after setJobStatus in BaseSparkCommitActionExecutor/CleanPlan/BloomIndex
• #9979 [HUDI-3304] Add support for selective partial update (CTTY, 2023-11-03, partial-update): superseded by 1.0 partial-update merge-mode work ([HUDI-3304] Support partial update payload #4676, [HUDI-8401][RFC-97] Add partial update support with merge mode to assist with migration from payload to merge mode #13498, [HUDI-8835] Support partial update in COMMIT_TIME_ORDERING merge mode #13263); payload-based approach outdated
• #10408 [HUDI-5026] Fix ORC bootstrap read support in Spark (on release 0.13.1) (yihua, 2023-12-26, release-0.13.1, bootstrap-orc): targets frozen release-0.13.1; master backport is draft [HUDI-5026] Fix ORC bootstrap read support in Spark #10406, no separate value on release branch
• #10845 [HUDI-7497] Add a global timeline mingled with active and archived instants (danny0405, 2024-03-11, timeline): experimental unified active+archived timeline never merged; superseded by 1.0 LSM archived-timeline rework
• #10969 [HUDI-8236] Upgrade maven-surefire-plugin and maven-failsafe-plugin to 3.5.0 (sullis, 2024-04-07, build): obsolete: master pom already on maven-surefire/failsafe 3.5.4, past the requested 3.5.0
• #11047 [HUDI-7634] Rename HoodieStorage APIs (yihua, 2024-04-17, storage-io): Cosmetic rename (listDirectEntries -> listDirectory); approved 2024 but never merged, name still in wide use on master, would conflict.
• #11069 [HUDI-7649] during inline clustering, update the hudi schema retrieval method (empcl, 2024-04-22, clustering): MultipleSparkJobExecutionStrategy schema handling rewritten to HoodieSchema/performClustering(schema) on master; one-line diff no longer applies.
• #11107 update fork count (the-other-tim-brown, 2024-04-26, build): Vague draft touching removed spark2/spark3.0.x/3.1.x poms; those modules deleted in Hudi 1.0; abandoned since 2024-05.
• #11374 [HUDI-7823] Simplify dependency management on exclusions (yihua, 2024-06-01, build): Pure exclusions refactor superseded by pom evolution; touches removed flink1.14-1.18 modules, exclusions reorganized on master.
• #11383 Deltastreamer changes to add support for reading kafka commit offsets from Kafka (KishanFairmatic, 2024-06-03, release-0.13.0, deltastreamer): Targets frozen release-0.13.0; DeltaSync/DeltaStreamer renamed to HoodieStreamer on master; niche Kafka-offset flag never adapted to master.
• #11435 [HUDI-3304][branch-0.x] Add support for selective partial update (CTTY, 2024-06-11, branch-0.x, writer): Targets frozen branch-0.x; master version tracked by [HUDI-3304] Add support for selective partial update #9979 (referenced in PR body).
• #11572 [HUDI-2955] Support Hadoop3 (CTTY, 2024-07-05, build-hadoop): Hadoop3 now supported on master via hadoop3 profiles (hadoop.version 3.4.0/3.4.2/3.4.3); XL PR superseded.
• #11593 [HUDI-7967][RFC][WIP] Robust spark writes rfc (nsivabalan, 2024-07-08, rfc): WIP draft RFC-79 (robust spark writes) never merged; rfc-79 absent on master; 2025-08 bump was bulk, no real progress since 2024.
• #11605 [HUDI-7974] Create empty clean commit at a cadence and make it configurable (suryaprasanna, 2024-07-10, clean): Feature landed on master: INTERVAL_TO_CREATE_EMPTY_CLEAN_HOURS / getIntervalToCreateEmptyCleanHours present in HoodieWriteConfig.
• #11606 [HUDI-7975] Provide an API to create empty commit (suryaprasanna, 2024-07-10, writer): CONFLICTING and abandoned after yihua changes-requested (2024-09); empty-commit support already present on master (allowEmptyCommit).
• #11663 Reduce index overhead (the-other-tim-brown, 2024-07-22, index): Draft with no description, 18-file index-handle refactor abandoned 2024-09; would conflict widely, author moved on.
• #11706 [HUDI-8024] Test secondary index updates and rollback (codope, 2024-07-30, secondary-index): Test-only draft (TestSecondaryIndexPruning) abandoned 2024-09; secondary index rewritten (v2) with its own test suites since.
• #11740 [HUDI-8292]: avoid commit promotion type not supported by aws tools (parisni, 2024-08-08, hive-sync): Stale/untested (author never confirmed it works; changes-requested unaddressed since Oct 2024); underlying Glue schema-evolution bug already tracked by issue [SUPPORT] Hudi should rollback commit when metasync fails #11701.
• #11870 [HUDI-6791][WIP] Add CDC support from HoodieFileGroupReader (linliu-code, 2024-09-03, cdc): WIP signature stub (2 files) abandoned same day; HoodieNewFileGroupReader/CDCFileGroupReaderIterator not on master; FileGroupReader CDC evolved separately.
• #11874 Multi table streamer (smtwilio, 2024-09-03, streamer): Vague draft abandoned 2024-10; HoodieMultiTableStreamer already exists on master, no described change.
• #11957 [HUDI-8213] Exclude jackson-databind from hudi-spark-bundle to fix CVE-2017-17485 (senthh, 2024-09-18, release-0.14.2, build-security): Targets frozen release-0.14.2; htrace/jackson-databind not in master hudi-spark-bundle (HBase deps being removed via own HFile readers [HUDI-7170] Implement HFile reader independent of HBase #10241/[HUDI-7218] Integrate new HFile reader with file reader factory #10330).
• #12067 [HUDI-8313] Adding a new sync tool to integrate with lakeview (pkgajulapalli, 2024-10-06, sync): Adds vendor-specific hudi-lakeview-sync (Onehouse Lakeview, proprietary); out of scope for upstream Hudi, no traction.
• #12131 [HUDI-8354] WIP: Have cloud event source use QueryContext (CTTY, 2024-10-21, utilities): WIP abandoned since 2024-10; touches HoodieDefaultTimeline which was removed in the 1.0 timeline rewrite, diff no longer applies
• #12195 [HUDI-8141] RFC-83: Incremental Query with Completion Time (CTTY, 2024-11-04, docs): RFC-83 slot reassigned to Incremental Table Service ([HUDI-8780][RFC-83] Incremental Table Service #12514 landed); completion-time incremental read already shipped in Hudi 1.0
• #12196 Add integration test framework (yihua, 2024-11-04, ci): Superseded by same-author larger Add integration test framework #12265; hudi-integration-test module never landed on master
• #12211 [HUDI-8485] Fix JSON Reader tests (csivaguru, 2024-11-06, utilities): Minor JSON-reader test fix, zero reviews, abandoned 2024-11; SourceFormatAdapter reworked since via fix: Use target schema for non-FileBased/SchemaRegistry providers in SourceFormatAdapter #17946
• #12235 [DOCS] Update sql_ddl.md for Flink SQL table creation with partitioning fix (YesIfwRONG, 2024-11-12, asf-site, docs): Fix targets frozen versioned docs for ancient 0.15.0 release only; version-specific ancient-release docs per triage rule
• #12275 [DNM] Error injection on Hudi 1.0 RC1 (yihua, 2024-11-18, release-1.0.0, testing): DNM draft, CONFLICTING, on frozen release-1.0.0; RC1 error-injection validation long completed
• #12404 [HUDI-8631] Fix the bug where the Flink table config hoodie.populate.meta.fields is not effective and optimize write performance (usberkeley, 2024-12-03, flink): populate.meta.fields fix landed via [HUDI-8631] Support of hoodie.populate.meta.fields for Flink append mode #12516 (HUDI-8631); reviewer redirected optimization work to existing HUDI-8308
• #12421 [MINOR]fix mor write log still has partition column when set drop partition (houyuting, 2024-12-04, writer): Abandoned draft since 2024-12, single-file minor unverified fix, no reviews
• #12429 [draft] Added code for DataFrame API for Spark-hudi datasource writer path (jalpeshborad, 2024-12-05, spark): Abandoned exploratory draft since 2024-12; contains stray .javaJB files, no reviews, not a mergeable approach
• #12460 [WIP][HUDI-3625][RFC-60] OSSStorageStrategy POC (zhangyue19921010, 2024-12-11, storage): WIP RFC-60 OSS storage-strategy POC, exploratory only, abandoned since 2025-02; design tracked via RFC-60 ([HUDI-3625][RFC-60] Update RFC-60 With Federated Storage Layout #12407)
• #12464 Update Presto Hudi documentation to move to own site instead of presto site (werowe, 2024-12-11, asf-site, docs): Misconfigured PR replaying whole master diff onto asf-site (100 files); bhasudha flagged it, superseded by clean add presto native documentation in the hudi site #12468
• #12521 [MINOR] Avoid unnecessary storage interactions for heartbeat (TheR1sing3un, 2024-12-19, writer): danny0405 pushed back on heartbeat always-trigger semantics; concern unaddressed, abandoned since 2024-12
• #12608 [HUDI-8163] Refactor UnMergedLogHandler with iterators (vamshipasunuru1, 2025-01-09, log-reader): Pure iterator refactor abandoned after 16 review rounds; scanners still function, no bug/feature gap to preserve
• #12622 [HUDI-8746] Support multi base file formats through fg reader (linliu-code, 2025-01-12, spark-reader): Superseded by [HUDI-8746] Add ORC support to FileGroupReader paths and migrate Spark multi-format reading to FileGroupReader #13632 (same HUDI-8746) which landed ORC/multi-format via FileGroupReader on master
• #12824 [HUDI-9000] Adding timestamp ordering validation for flink 0.x branch (nsivabalan, 2025-02-10, branch-0.x, flink): base branch-0.x (frozen); backports timestamp-ordering validation master already has; now CONFLICTING
• #12841 [MINOR] HoodieBackedTableMetadataWriter and HoodieTableMetadataUtil code cleanup (codope, 2025-02-14, metadata-table): MINOR cleanup touching 35 files, now heavily conflicting; changes-requested plus CI failures, abandoned since Feb 2025
• #12982 [HUDI-9187][HUDI-6249] Update usage of read/write in FileSystemView code paths (the-other-tim-brown, 2025-03-14, file-system-view): FSView read/write refactor; got 1 approval but abandoned 15mo, FSView heavily refactored since (LSM FG reader, Lombok, double-iterating fix)
• #12983 [HUDI-9177] Add Indexer abstraction and refactor metadata table initialization logic (yihua, 2025-03-15, metadata-table): Superseded by author's reworked feat(index): Add Indexer abstraction and refactor metadata table init… #18348 (feat(index): Add Indexer abstraction and refactor MDT init) landed on master
• #13009 [HUDI-8409] Add test to ensure merge mode during upgrade and downgrade (linliu-code, 2025-03-20, upgrade-downgrade): HUDI-8409 fix plus tests landed via [HUDI-8409] Fixing merge mode config during upgrade and downgrade from version 7 to 8 and back  #13046; this test-only PR abandoned after 2x changes-requested
• #13015 [HUDI-9208] Add file system view storage configs to MDT (yihua, 2025-03-21, metadata-table): Minor MDT config PR abandoned by active committer after 2 comments; MDT writer path reworked by indexer refactor (feat(index): Add Indexer abstraction and refactor metadata table init… #18348/feat(index): Refactor metadata table update logic based on index abstraction #18372)
• #13021 hudi0.13.1 (hanleiit, 2025-03-24, misc): Junk PR: 100 unrelated files (bot.yml, azure-pipelines, docker demo configs) from a stale 0.13.1 fork; no coherent change
• #13034 [WIP] [HUDI-9219] Fixing .close() calls for ClosableIterator (vinothchandar, 2025-03-26, core-io): Superseded by [HUDI-9219] Add callbacks for CloseableIterators and eager close data readers #13178 [HUDI-9219] Add callbacks for CloseableIterators and eager close data readers, landed on master
• #13071 [HUDI-9132] Avoid empty string row key for delete and update operations (linliu-code, 2025-04-02, spark-sql): Already landed on master via [HUDI-9132] Avoid empty string row key for delete and update operations #12929 (commit 35ab3bc, identical HUDI-9132 title)
• #13076 [MINOR][DRAFT] test partial updates (linliu-code, 2025-04-03, test): Draft test-scaffolding branch (2 test files), no reviews, abandoned since Apr 2025
• #13082 [HUDI-6249] Move thread-safety to map impls for FSView (the-other-tim-brown, 2025-04-03, file-system-view): Draft follow-up to abandoned 12982 FSView thread-safety effort; superseded by later FSView refactors, no reviews, 15mo stale
• #13105 [DRAFT][NMV] Keep merge mode test branch (linliu-code, 2025-04-08, test): Explicit scratch/no-merge-value branch (49 files, examples/integration-test), never intended to merge
• #13226 [DNM][Stacked on 12983] Refactor MDT update logic with Indexer abstraction (yihua, 2025-04-27, metadata-table): Superseded by author's reworked feat(index): Refactor metadata table update logic based on index abstraction #18372 (Refactor MDT update logic based on index abstraction) landed on master
• #13272 [HUDI-9391] Unify column stats collection for metadata table across engines (yihua, 2025-05-08, metadata-table): Superseded by engine-agnostic ColumnStatsIndexer landed with indexer abstraction (feat(index): Add Indexer abstraction and refactor metadata table init… #18348); draft abandoned
• #13276 [HUDI-9397] Move index enabling condition to MetadataPartitionType (yihua, 2025-05-08, metadata-table): Landed: MetadataPartitionType.isMetadataPartitionEnabled(config,tableConfig) now on master via the indexer refactor
• #13280 [HUDI-9398] Generalize index definition update in the table metadata (yihua, 2025-05-09, metadata-table): Draft absorbed into the landed indexer abstraction refactor (feat(index): Add Indexer abstraction and refactor metadata table init… #18348/feat(index): Refactor metadata table update logic based on index abstraction #18372); abandoned, no reviews
• #13293 [HUDI-8746] Introduce SparkFileReader for general base file format (linliu-code, 2025-05-12, spark-reader): Superseded by [HUDI-8746] Add ORC support to FileGroupReader paths and migrate Spark multi-format reading to FileGroupReader #13632 (HUDI-8746 Add ORC support to FGR and migrate Spark multi-format reading to FileGroupReader)
• #13348 [HUDI-9442] Improve key lookup in metadata table (yihua, 2025-05-23, metadata-table): Abandoned MDT key-lookup WIP draft; author reworked and landed MDT via indexer abstraction; LookUpKeyCollection never merged
• #13349 [HUDI-9444] WIP Improve Key Lookup for HFile/Metadata (yihua, 2025-05-23, metadata-table): WIP fork of same key-lookup effort as 13348, same day; abandoned, superseded by landed MDT indexer refactor
• #13386 [HUDI-9361] Use merger to replace spark ExpressionPayload (linliu-code, 2025-06-02, spark-sql): Exploratory draft (SparkExpressionMerger not on master), no reviews; superseded by the landed HUDI-8401 payload>merge-mode direction
• #13389 [HUDI-8401] Add merger classes for CDC payloads. (linliu-code, 2025-06-03, cdc-merger): Merger classes never landed; HUDI-8401 payload>merge-mode migration landed via a different structure ([HUDI-8401][RFC-97] Add partial update support with merge mode to assist with migration from payload to merge mode #13498/[HUDI-8401] Fix 8 to 9 Upgrade handler and 9 to 8 downgrade handler for payload deprecation #13519/[HUDI-8401] Upgrade current table version to 9 #13533), actively tracked
• #13412 [HUDI-9395][DNM] Remove FG reader flag Usage (linliu-code, 2025-06-09, reader): DNM/draft experiment branch (23 files); FILE_GROUP_READER_ENABLED still on master, flag-removal cleanup never proposed for merge
• #13421 [HUDI-8401] Add merger class LegacyMySqlDebeziumAvroMerger (linliu-code, 2025-06-12, cdc-merger): LegacyMySqlDebeziumAvroMerger never landed; subset of HUDI-8401 migration superseded by the landed merge-mode approach
• #13427 Added BlockCache to HFile reader. (pacificleo, 2025-06-12, release-0.14.2, storage-hfile): Targets frozen release-0.14.2; master replaced HBase HFile reader with native HoodieNativeAvroHFileReader, HBase CacheConfig/BlockCache approach N/A
• #13442 [DRAFT] Dockerfile include maven (the-other-tim-brown, 2025-06-16, ci): Draft abandoned since 2025-06; minor bundle-validation Dockerfile tooling, never taken past WIP
• #13443 [HUDI-9395] Validation schema evolution (linliu-code, 2025-06-17, tests-schema-evolution): Draft test-only, stale since 2025-06; overlaps broader schema-on-read fg-reader test effort in [HUDI-9657] Test schema on read in fg reader #13628
• #13456 [MINOR] The build is interrupted due to jars conflict on hudi 0.15.0 and spark3.5 (wangxiaopeng0329, 2025-06-18, branch-0.x, build): Targets branch-0.x maintenance freeze; 0.15.0/spark3.5-specific jar conflict, not relevant to 1.x master
• #13505 [HUDI-9563] Add tags for AWS Glue database and remove it for tables (vinishjail97, 2025-07-01, aws-glue-sync): Superseded by GLUE_SYNC_RESOURCE_TAGS on master which tags both Glue databases and tables; draft stale since 2025-07
• #13540 [HUDI-8401] Support KeepValues partial merge mode (linliu-code, 2025-07-09, reader-partial-update): HUDI-8401 partial update merge mode landed via [HUDI-8401][RFC-97] Add partial update support with merge mode to assist with migration from payload to merge mode #13498 and series; PartialUpdateMode + KEEP_VALUES now on master
• #13623 [HUDI-8401] Unifying Partial Update modes and Merge Into Partial Update Encoding (nsivabalan, 2025-07-25, reader-partial-update): HUDI-8401 partial update unification landed via [HUDI-8401][RFC-97] Add partial update support with merge mode to assist with migration from payload to merge mode #13498/[HUDI-8401] Fix 8 to 9 Upgrade handler and 9 to 8 downgrade handler for payload deprecation #13519 series; PartialUpdateMode on master, draft superseded
• #13675 [MINOR] Follow up refactor of index lookup (Davis-Zhang-Onehouse, 2025-08-04, metadata-table-index): Parent [HUDI-9656] Index look up path clean up #13647 index-lookup rename landed (RawKey classes on master); stale conflicting refactor-only follow-up
• #13681 [HUDI-9638][DNM] Table creation for v9 based on COW Handle (linliu-code, 2025-08-05, write-table-version): Explicit DNM experimental draft ('to see if many more tests pass after rebase'), not meant to merge
• #13692 add test to ensure that avro payload is used (jonvex, 2025-08-07, tests): Draft test-only, no description, stale since 2025-08; single TestHoodieDeltaStreamer tweak never developed
• #13695 [MINOR] code refactor for better hoodie data clean up protocol (Davis-Zhang-Onehouse, 2025-08-07, metadata-table-data): HoodieDataUtils abstraction landed on master via [HUDI-9778] Remove HoodieRecord wrapper from MDT reads #13796 (HUDI-9778); conflicting refactor superseded
• #13708 add missing spark evolutions in (jonvex, 2025-08-11, spark-schema-evolution): Draft, trivial 1-file change, no description, stale since 2025-08; never developed past WIP
• #13735 [HUDI-9701] Refactor upgrade downgrade logic (linliu-code, 2025-08-19, upgrade-downgrade): needsUpgrade/needsDowngrade split already present in master UpgradeDowngrade.java; refactor landed independently
• #13751 [DNM][HUDI-9419] Fix upgrade action for Flink writer (cshuo, 2025-08-22, release-1.0.2, flink-upgrade): Master fix landed via [HUDI-9419] Fix upgrade action for Flink writer #13310 (tryUpgrade in StreamWriteOperatorCoordinator); DNM backport to frozen release-1.0.2
• #13832 [DRAFT] incremental query testing (rahil-c, 2025-09-03, spark-tests): Bare [DRAFT] scaffold, single TestCOWDataSource change, abandoned since 2025-09; no substance to review
• #13901 Fix classloader for commit operation in flink coordinator (cshuo, 2025-09-16, release-1.0.2, flink): Classloader fix already on master (ExplicitClassloaderThreadFactory present); PR targets frozen release-1.0.2
• #14003 feat: Adding completion time column for Incremental Queries V2 (vamshikrishnakyatham, 2025-09-27, spark-sql): Duplicate V2 draft of feat: Addition of virtual _hoodie_commit_completion_time column which is used for Incremental Queries #14037; feature already tracked by open issue Addition of virtual _hoodie_commit_completion_time column which is used for Incremental Queries #14036; stale draft since 2025-10
• #14037 feat: Addition of virtual _hoodie_commit_completion_time column which is used for Incremental Queries (vamshikrishnakyatham, 2025-10-02, spark-sql): Virtual completion-time column not landed; feature tracked by open issue Addition of virtual _hoodie_commit_completion_time column which is used for Incremental Queries #14036; PR stale since 2025-10, author moved to feat: Adding completion time column for Incremental Queries V2 #14003
• #14063 [DO NOT MERGE] refactor(spark): Remove spark datasource relations (jonvex, 2025-10-08, spark): [DO NOT MERGE] experimental XL refactor stacked on refactor(spark): Rework tests that disable FileGroup Reader in Spark #14061; explicitly not for merge
• #14103 perf: Improve metadata record merging and compaction with lazy deserialization (yihua, 2025-10-16, metadata-table): Metadata merge perf (lazy deser) draft; CONFLICTING, idle since 2025-11; work tracked by referenced issue Improve metadata record merging and compaction with lazy Avro serde #14370
• #14260 Metadata sync unified view (vamsikarnika, 2025-11-13, release-0.14.0, metadata-table): Draft targeting frozen release-0.14.0; feature (not a master bug) unrescuable on master; close per release-branch freeze rule
• #14377 feat: Add config to disable bloom filter for HFile (yihua, 2025-11-27, storage): HFile bloom-filter toggle already landed as hoodie.hfile.bloom.filter.enabled via feat(mdt): Support writing bloom filter for hfile log block #18936; draft superseded
• #14378 Add compressor implementation of zstd compression codec (yihua, 2025-11-27, io): Same-day draft never pursued; HoodieAirliftZstdCompressor absent from master; scope tracked by issue 14369
• #17447 feat: Implement xxhash for generating bloom filter (yihua, 2025-12-01, common): Same-day draft xxhash spike abandoned; XxHash.java not on master; scope tracked by issue 17446
• #17449 fix: Make HFile block size configurable for metadata table (yihua, 2025-12-01, metadata-table): Same-day draft to make MDT HFile block size configurable, never pursued; scope tracked by issue 17448
• #17478 minor: Enabling streaming writes to metadata code paths only when eligible metadata partitions (nsivabalan, 2025-12-04, metadata-table): Minor MDT streaming-write eligibility optimization; same-day draft abandoned by nsivabalan, never pursued
• #17527 chore: Add a script to combine multiple staging repos with jars from different JDKs (yihua, 2025-12-08, release): Same-day draft standalone multi-JDK staging-repo combine script; release tooling never pursued
• #17692 chore(ci): Update Java 11 image for Azure CI (yihua, 2025-12-23, ci): trivial Java 11 Azure CI image bump, abandoned 5 months, time-sensitive image likely stale now
• #17693 feat: MDT Test framework without writing data files (PavithranRick, 2025-12-23, metadata-table): draft superseded by MDT Test framework without writing data files #17796 (same MDT test-framework effort by same team, which got the reviews)
• #17775 feat(spark): add config to list partition paths from HMS in file index (prashantwason, 2026-01-04, spark/file-index): maintainer nsivabalan asked to close it out; also leaked .claude/settings.local.json
• #17786 feat: Integrate Trino 477 Docker image into Hudi docker demo environment (deepakpanda93, 2026-01-05, docker): draft abandoned 2 days after creation, no reviews, docker-demo Trino image bump goes stale fast
• #17832 docs: Add lance max file size (KiteSoar, 2026-01-12, asf-site, docs): hoodie.lance.max.file.size already present in auto-generated asf-site configurations.md (regenerated 2026-05-28)
• #17854 [HUDI-8212] Add extra config of billing project ID for BigQuery sync ... (linliu-code, 2026-01-13, branch-0.x, gcp/bigquery): branch-0.x backport; BIGQUERY_SYNC_BILLING_PROJECT_ID config already on master, not needed
• #17868 [HUDI-9088] Fix unnecessary scanning of target table in MERGE INTO on Spark (linliu-code, 2026-01-14, release-0.14.2, spark-sql/merge-into): already fixed on master via [HUDI-9088] Fix unnecessary scanning of target table in MERGE INTO on Spark #12934 (HUDI-9088); stale frozen-release-0.14.2 backport
• #17874 [HUDI-9666] Fix the record key encoding with a single record key field and add a guard for complex key generator (lokeshj1703, 2026-01-14, release-0.14.2, keygen): already landed on master via [HUDI-9666] Fix the record key encoding with a single record key field and add a guard for complex key generator #13650 (HUDI-9666); stale frozen-release-0.14.2 backport
• #17906 feat: add storage LP implementation for azure (alexr17, 2026-01-16, azure/lock): Azure storage lock provider already on master (AzureStorageLockClient present, predates refactor refactor(client)!: reorganize hudi-client-common internal package layout #19192)
• #17977 fix: create release branch release-1.0.3 (linliu-code, 2026-01-22, release-1.0.2, release): throwaway draft release-branch bootstrap PR (create release-1.0.3), abandoned same day, obsolete
• #18048 [do not review] Draft for blocking archival on latest ingestion write (kbuci, 2026-01-29, archival): explicit [do not review] experimental draft, abandoned same day it was opened
• #18096 [DNM] Branch 0.x with logic types fix debug (lokeshj1703, 2026-02-05, branch-0.x, core): DNM debug draft targeting frozen branch-0.x; no master relevance
• #18102 [WIP] perf: Skip reading file slice only containing an initial log file (cshuo, 2026-02-06, metadata-table): Abandoned WIP (created=updated 2026-02-06, no description, 11 lines); never iterated, easy to redo if the MDT read optimization is wanted
• #18145 fix(metadata-table): Disable archival and clean on MDT if they are disabled on main table (nada-attia, 2026-02-09, metadata-table): Abandoned after nsivabalan questioned the premise (MDT table services already guarded by data-table archival by design); CI failing, no follow-up since 2026-02-13
• #18169 fix: [HUDI-9159] S3 implementation of StorageLock for StorageBasedLockProvider (PavithranRick, 2026-02-10, branch-0.x, locking): Feature already landed on master (S3StorageLockClient.java and TestS3StorageBasedLockProvider present); draft on frozen branch-0.x superseded
• #18170 (DNM) fix: Upgrade Spark to 3.2.4 and fix parquet schema conversion (yihua, 2026-02-10, release-0.12.3, spark): DNM draft targeting ancient frozen release-0.12.3; no master relevance
• #18171 [HUDI-9666] Fix the record key encoding with a single record key field and add a guard for complex key generator (PavithranRick, 2026-02-11, branch-0.x, key-generator): Draft on frozen branch-0.x, abandoned since 2026-02-11 by inactive author; single-key-field encoding fix is 0.x-specific (key-gen rewritten in 1.x)
• #18194 fix: clone record data in batch reader of Flink source (HuangZhenQiu, 2026-02-13, flink): FLIP-27 source reader rewritten; target files DefaultHoodieBatchReader/RecordCloner no longer exist on master (now HoodieSourceSplitReader), diff no longer applies; abandoned same-day
• #18201 (DNM) chore: Improve CI time (yihua, 2026-02-14, ci): DNM draft experiment, no description, last touched 2026-02-17
• #18259 feat(blob): RFC-100: Update plan for cleaner (the-other-tim-brown, 2026-02-26, rfc): Small rfc-100.md cleaner-plan edit superseded by later merged RFC-100 restructuring (feat(blob): update approach to remove reliance on column groups, break down plan #18013 break-down-plan, feat(blob): RFC-100: Clarify inline vs out-of-line blob read behavior #18728 clarify read behavior)
• #18261 test(spark): add TimestampBasedKeyGenerator partitioning coverage (yohaku999, 2026-02-28, spark-sql): Test-only PR now CONFLICTING, stalled since 2026-03; coverage gap already tracked by open issue Add more tests to TestCOWDataSource for TimestampbasedKeyGen #14753, external author inactive
• #18262 fix cdc compare result (qq726658006, 2026-02-28, cdc): Already on master: HoodieCDCFileSplit.compareTo has the LOG_FILE beforeFileSlice log-count tie-breaker (landed via base PR fix(flink): Use blocking instant generation when CDC is enabled #18206)
• #18318 [Do not review] Apply max commits to clean config (kbuci, 2026-03-12, clean): Explicit '[Do not review]' draft scratch PR, conflicting, untouched since 2026-03-18
• #18351 (WIP) benchmark: [RFC-98] COW table read performance comparison (geserdugarov, 2026-03-19, benchmark): WIP RFC-98 benchmark scaffold, mostly copies of feat(dsv2): [RFC-98] DSv2 read for COW #18277, not intended to merge
• #18373 Trino/Spark benchmarking tool (vamsikarnika, 2026-03-24, benchmark): No reviews, untouched since 2026-03-26, conflicting, carries unrelated core metadata diffs (branch noise)
• #18389 [DNM] [WIP] Logical ts vectorized (lokeshj1703, 2026-03-26, branch-0.x, spark): [DNM][WIP] draft on branch-0.x (maintenance freeze), conflicting, untouched since Mar
• #18455 [ENG-38911] Test hudi rs with hudi functional tests (Davis-Zhang-Onehouse, 2026-04-03, ci): Internal Gluten/rust test scaffold, RCA md at repo root, hardcoded /home/ubuntu paths, conflicting; not for master
• #18622 chore: [DNM] Release 1.2.0 RC1 bundle validation (yihua, 2026-04-28, release-1.2.0, release): Explicit [DNM] RC1 bundle-validation harness; release-1.2.0 already shows published version 1.2.0 (commit 05-22), validation obsolete
• #18737 Flink dataloss fixes (suryaprasanna, 2026-05-14, release-0.14.2, flink): Backport to frozen legacy release-0.14.2; fixes already on master, CI failing
• #18757 chore(flink): cherrypick append write code path fixes (HuangZhenQiu, 2026-05-16, release-0.14.2, flink): Cherrypick of 2023 PRs ([HUDI-5506] StreamWriteOperatorCoordinator may not recommit with partial uncommitted write metadata event #7611/[HUDI-6588] Fix duplicate fileId on TM failover and recovery #9357/[HUDI-6944] Fix Flink bootstrap concurrency issue causes job to keep rolling back #9867 etc) onto frozen legacy release-0.14.2; on master already
• #18768 feat: [DNM] Async instant generation fixes over 0.14.1 (cshuo, 2026-05-18, release-0.14.1, flink): Explicitly [DNM], targets frozen legacy release-0.14.1 backport of async-instant fixes
• #19229 [DNM][0.14] Add versioned binary list state for flink writer (cshuo, 2026-07-09, release-0.14.1, flink): DNM backport of versioned binary list state to frozen release-0.14.1 (0.14.x long EOL); no reviews, experimental.

[voonhous]

CONVERT to issue, then close (166)

Tick the box once the scope has been captured as an issue and the PR closed (or the verdict overturned -- note it in a reply).

• #3417 [HUDI-2281] Add metadata client APIs to fetch list of data files and ... (satishkotha, 2021-08-05, metadata-table): HoodieSnapshot/IncrementalMetadataClient APIs never landed; plausible convenience API but PR untouched since 2021 (bot bumps only)
• #4944 [HUDI-3554]: Support Custom Naming of RO and RT tables for Hudi MOR table (fengjian428, 2022-03-03, hive-sync): custom RO/RT naming still plausible; HiveSyncTool/Config exist but hudi-flink path moved, PR stale since 2022
• #5071 [HUDI-1881]: draft implementation for trigger based on data availability (pratyakshsharma, 2022-03-19, deltastreamer): data-availability trigger for MultiTableDeltaStreamer; draft abandoned 2022 but parallel-ingest scope still valid
• #5239 [WIP][HUDI-3787] Implement hudi redis index (XuQianJin-Stars, 2022-04-06, index): Redis index WIP never merged; no redis index on master, feature still plausible but PR abandoned
• #5359 [HUDI-3916] New Key Generator Option: NanoidKeyGenerator (zhangyue19921010, 2022-04-19, key-gen): NanoidKeyGenerator not on master; feature still plausible but PR stale (bot bumps only since 2022)
• #5406 [HUDI-3954] Don't keep the last commit before the earliest commit to retain (dongkelun, 2022-05-09, table-service-clean): cleaner retention behavior; CleanPlanner heavily refactored since 2022 so diff won't apply, but behavior worth re-evaluating
• #5534 [HUDI-4056] Refine partition matching in when drop / truncate partition. (jinxing64, 2022-05-09, spark-sql): partial-spec drop/truncate partition matching still relevant; commands exist but PR stale since 2022
• #5678 [HUDI-4153]StreamReadOperator supports async opeing of foramt. (cuibo01, 2022-05-25, flink): async format opening in StreamReadOperator; operator still exists, perf enhancement still relevant but PR stale
• #5758 [HUDI-4194] upgrade confluent version to 7.1.1 (victorxiang30, 2022-06-06, dependencies): master still on confluent 5.5.0; upgrade still wanted but target 7.1.1 now outdated, PR stale, retarget needed
• #5760 [HUDI-4196]support index alignment (cuibo01, 2022-06-06, flink): Flink bootstrap index alignment; no IndexAlignment on master, feature plausible but PR stale since 2022
• #5978 [HUDI-4284] Implement bloom lookup tree as red-black tree (yabola, 2022-06-26, index): bloom KeyRangeLookupTree still exists, no rbtree landed; balancing perf enhancement still relevant but PR stale
• #5991 [HUDI-4329] Add separate control for Flink compaction operation sync/async mode (chenshzh, 2022-06-28, flink): CompactFunction rewritten to CompactOperator+handlers so diff moot; sync/async compaction config capability worth re-evaluating
• #6134 WIP:refactor hoodie stream source based flip-27 and support watermark (JerryYue-M, 2022-07-18, flink): FLIP-27 source + watermark modernization; no flip27 package on master, Flink still on legacy SourceFunction; WIP too stale to rescue
• #6273 [HUDI-4521] Ensure flink coordinator do re-commit during restart for changing write.tasks number (i.e., write parallelism) (YuweiXiao, 2022-08-02, flink): Re-commit on write.tasks parallelism change never landed (WriteMetadataEvent has no parallelism field); bug still relevant, review stalled 2022
• #6337 [HUDI-4578] Reduce the scope and duration of holding checkpoint lock (XuQianJin-Stars, 2022-08-09, flink): StreamReadMonitoringFunction still holds coarse checkpointLock; lock-scope reduction not applied, PR stale since 2022
• #6409 [HUDI-4629] Create hive table from existing hoodie table failed when the table schema is not defined (dongkelun, 2022-08-16, spark-sql): Create table over existing hoodie table without schema; HoodieCatalogTable evolved, fix not confirmed; dongkelun PR stale since 2022
• #6419 [HUDI-2057][HUDI-5499] CTAS Generate An External Table When Create Managed Table (dongkelun, 2022-08-17, spark-sql): CTAS managed-vs-external bug (recurring per PR body); abandoned after xushiyan CHANGES_REQUESTED, still worth an issue
• #6435 [HUDI-4639] Flink-sql batch query support query delete_record (flashJd, 2022-08-18, flink): Flink batch query of delete_record not on master; feature stale since 2022, author gone
• #6446 [HUDI-4645] Flink batch incremental changelog query (flashJd, 2022-08-19, flink): Flink batch incremental changelog query not on master; touches core file-group APIs, PR stale since 2022
• #6448 [HUDI-4647] Keep the hive sync settings in spark sql consistent (dongkelun, 2022-08-19, spark-sql): Hive-sync config consistency across Spark SQL paths still valid; MergeIntoHoodieTableCommand since split per Spark version so diff won't apply
• #6493 [HUDI-4715] Needed To ReSync Hive In StreamWriteOperatorCoordinator's initInstant method (JerryYue-M, 2022-08-25, flink): Flink coordinator hive re-sync on initInstant/recovery not landed; niche bug, PR stale since 2022
• #6510 [HUDI-4724] Add function of skip the _rt suffix for read snapshot (linfey90, 2022-08-26, hive-sync): skip _rt suffix for snapshot table not on master (only skip_ro_suffix exists); symmetric feature, PR stale since 2022
• #6581 [HUDI-4773] Adding New filter mode to clustering to filter for recently touched files (nsivabalan, 2022-09-04, clustering): Recently-touched-files clustering filter mode not on master (only partition filter modes); optimization still valid, PR stale
• #6858 [HUDI-4824] RANGE BUCKET index, base logic and test. (wqwl611, 2022-10-03, index): RANGE bucket index not on master; targets auto-increment PK CDC, partially overlaps consistent-hashing bucket; PR stale since 2022
• #6974 [WIP][HUDI-5043] Add Hudi CLI command to roll back inflight replacecommit for clustering (yihua, 2022-10-18, cli): CLI rollback of inflight clustering replacecommit still missing on master; WIP by committer (blocker/triaged), worth an issue
• #7000 [HUDI-5060] Make all clean policies support incremental mode to find partition paths (dongkelun, 2022-10-19, table-service): Extend incremental partition-path discovery to all clean policies; CleanPlanner only partially supports it; PR stale since 2022
• #7019 [HUDI-5068] Support Flink cow batch create fs input splits asynchronously (chenshzh, 2022-10-21, flink): Flink COW async fs input-split building; CopyOnWriteInputFormat still on master but PR got only CI-rerun activity since 2022, no real review. Capture perf idea as issue.
• #7035 [HUDI-5075] Adding support to rollback residual clustering after disabling clustering (nsivabalan, 2022-10-22, table-service): Rollback residual clustering after disabling it; vinoth requested avoiding new configs, never addressed. Table-service concern still valid, capture as issue.
• #7041 [HUDI-5053] Create clean complete commit when there is none to clean in order to leverage incremental cleaning (hussein-awala, 2022-10-23, table-service): Draft since 2022 emitting empty clean commit to keep incremental cleaning; changes-requested unaddressed. Incremental cleaning still exists, capture optimization as issue.
• #7075 [HUDI-5100][flink]Support writing tasks independently in the flink batch mode (waywtdcc, 2022-10-27, flink): Flink batch-mode independent write tasks; danny wanted it for 0.14.0 but never landed, CkpMetadata still exists. Still-open resource-allocation gap, capture as issue.
• #7099 [HUDI-5095] Flink: Stores a special watermark(flag) to identify the c... (JerryYue-M, 2022-10-31, flink): Flink event-time watermark flag; 1 approval but changes-requested unaddressed since 2024. Sink files still exist, feature still relevant, capture as issue.
• #7144 [HUDI-5164] improve delete files for drop table and truncate table (XuQianJin-Stars, 2022-11-05, spark-sql): Improve file deletion on drop/truncate table; 2 approvals but UT failure unaddressed since 2024, author inactive. Commands still exist, capture as issue.
• #7146 [HUDI-5165][WIP] Adding sorting option with insert/upsert operation (nsivabalan, 2022-11-06, spark): WIP sorting option for insert/upsert; CONFLICTING, multiple community users still request it. Needs rewrite, capture as issue.
• #7183 [HUDI-5194] Fix schema files cleaning by FileBasedInternalSchemaStorageManager (xiarixiaoyao, 2022-11-11, schema): Fix schema-file cleaning by FileBasedInternalSchemaStorageManager; bvaradar asked rebase on [HUDI-4588][HUDI-4472] Addressing schema handling issues in the write path #6358, never done. Manager still on master, bug may persist, capture as issue.
• #7189 [HUDI-5159]Add support write success file to finished partition in flink streaming write (KevinyhZou, 2022-11-12, flink): Flink success-file per finished partition; touches removed flink1.13/1.14 adapters. danny wants pluggable generation logic. Needs rewrite, capture as issue.
• #7255 [HUDI-5250] use the estimate record size when estimation threshold is l... (honeyaya, 2022-11-21, table-service): Use estimated record size when below threshold; 2 approvals but missing negative-threshold test, author gone. UpsertPartitioner/config still exist, capture small fix as issue.
• #7379 [HUDI-5323] Support virtual keys in Bloom Index and always write bloom filters to parquet files (yihua, 2022-12-05, index): Virtual keys in bloom index + always-write bloom filters; 59-file 2022 diff, index/reader path rewritten for MDT in 1.x. Diff meaningless now, capture concept as issue.
• #7469 [HUDI-5386] Cleaning conflicts when write concurrency mode is OCC (LinMingQiang, 2022-12-15, table-service): Cleaning conflicts under OCC (concurrent rollback lacks lock); discussion stalled 2024 with no ticket-lock design agreed. Problem still valid, capture as issue.
• #7519 [HUDI-5422] Control KEEP_LATEST_VERSIONS clean replaced files immediately or delete after a while (zhangyue19921010, 2022-12-20, table-service): Control immediate vs delayed cleanup of replaced files under KEEP_LATEST_VERSIONS; no review, stale since 2024. CleanPlanner still exists, capture feature as issue.
• #7567 [docs] add update data section in how-to writing data guide (brandonJY, 2022-12-28, asf-site, docs): asf-site doc adding an update-data section to writing_data guide; unmerged since 2022, content likely stale vs current SQL. Capture doc gap as issue.
• #7598 [HUDI-5495] add some property to table config (XuQianJin-Stars, 2023-01-04, table-config): Add properties to table config; vague scope, 1 approval, stale since 2024, HoodieTableConfig heavily changed since. Capture as issue if still wanted.
• #7680 [HUDI-5548] spark sql show | update hudi's table properties (XuQianJin-Stars, 2023-01-16, spark-sql): SQL show/alter hudi table properties; 3 approvals but failing test never fixed, author gone. Alter-props partly landed via AlterTableCommand. Capture remainder as issue.
• #7718 [HUDI-5591] HoodieSparkSqlWriter#getHiveTableNames needs to consider ... (LinMingQiang, 2023-01-20, spark-sql): Fix HoodieSparkSqlWriter#getHiveTableNames; method still on master (line 961), small fix stalled since 2024. Capture as issue.
• #7748 [HUDI-5560][RFC-42] Make Consistent hash index Bucket Resizing more available... (fengjian428, 2023-01-25, rfc): RFC-42 amendment to make consistent-hash bucket resizing more available; base RFC already on master, doc-only enhancement stale since 2023. Capture enhancement as issue.
• #7811 [HUDI-5518] Support canal-json for HoodieDeltaStreamer (kaori-seasons, 2023-01-31, deltastreamer): canal-json KafkaSourcePostProcessor never landed (no canal in hudi-utilities on master); no reviews, abandoned 3y; capture as feature
• #7842 [HUDI-5695]Fix the wrong use of order field in PartialUpdateAvroPayload class. (scxwhite, 2023-02-03, payload): PartialUpdateAvroPayload still on master and still merges by orderingVal; ordering-field bug may persist; PR stalled after danny0405 feedback
• #7847 [HUDI-5697] Revisiting refreshing of Hudi relations after write operations on the tables (alexeykudinkin, 2023-02-03, spark-sql): over-eager Catalog.refreshTable after DML still a concern; PR adds now-absent HoodieCatalogUtils.scala, stalled 2023, author inactive
• #7859 [HUDI-2175] Adding support for dynamic schemas (nsivabalan, 2023-02-06, schema): dynamic schema reconcile strategy not landed (SchemaReconcileUtils absent on master); CONFLICTING, no reviews, abandoned 2024
• #7904 [HUDI-5735] Flink-hudi write time format data UTC time zone problem (alexander1005, 2023-02-09, flink): flink timestamp UTC read/write mismatch may still exist; converters still on master; PR broken (DateUtils lacks .java), no reviews, abandoned
• #7984 hash partition in spark datasoure (lvhu-goodluck, 2023-02-16, spark-datasource): hash partitioning not implemented on master (no _hoodie_hash_part); invasive 13-file PR, no reviews, abandoned 2024; capture as feature
• #8056 [MINOR] add integrity check of parquet file (XuQianJin-Stars, 2023-02-27, storage): parquet write/read integrity check for HoodieRowDataParquetWriter still a valid enhancement; files exist, PR stalled after danny0405 comment 2024
• #8057 [HUDI-5831] Address flakiness of early conflict detection tests (xicm, 2023-02-27, tests): MarkerBasedEarlyConflictDetectionRunnable and multi-writer test still on master; race in testHoodieClientBasicMultiWriter flakiness may persist
• #8208 [HUDI-5903]: Introduce Glue sync configs (pratyakshsharma, 2023-03-16, catalog-sync): Glue region/connection configs still absent from GlueCatalogSyncClientConfig on master; CHANGES_REQUESTED by bvaradar never addressed
• #8219 [HUDI-5949] Check the write operation configured by user for better troubleshooting (wecharyu, 2023-03-17, spark-sql): insert/upsert-without-precombine config validation still useful; HoodieSparkSqlWriter/ProvidesHoodieConfig exist; stalled after danny0405 review
• #8424 [HUDI-6065] Fix deltastreamer shutdown when post write termination strategy enabled (LiJie20190102, 2023-04-11, deltastreamer): HoodieDeltaStreamer still on master; table-service shutdown hang on post-write-termination may persist; single-file fix stalled after codope comment
• #8562 [DOCS] Update HMS lock provider behavior (codope, 2023-04-24, asf-site, docs): concurrency_control.md still live; HMS-lock crash-behavior clarification still useful but blocked by vinothchandar CHANGES_REQUESTED never addressed
• #8570 [HUDI-6136] Adding idempotent writes to Hudi (nsivabalan, 2023-04-25, spark-datasource): idempotent writes not landed (no writer.identifier/batch.identifier configs on master); feature still valuable; stalled 2024 after bvaradar comment
• #8657 [HUDI-6150] Support bucketing for each hive client (parisni, 2023-05-06, hive-sync): hive-bucketing propagation not landed (HiveBucketingSpec absent on master); author unsure on hash semantics; capture as feature, no formal review
• #8673 [HUDI-6194] prevent flink writer getting the wrong instant to write (hbgstc123, 2023-05-09, flink): flink bootstrap-event race writing to wrong instant is a data-consistency bug; StreamWriteOperatorCoordinator reworked, needs verify+capture; stalled 2025
• #8680 [HUDI-6192] Make HoodieFlinkCompactor and HoodieFlinkClusteringJob service mode as long runnning streaming job (SteNicholas, 2023-05-10, flink): long-running streaming service mode for flink compactor/clustering not landed (ServiceSourceFunction absent on master); stalled after danny0405 review 2024
• #8753 [HUDI-5095] Flink: Stores a special watermark(flag) to identify the current progress of writing data (SteNicholas, 2023-05-18, flink): Flink write-progress watermark in commit metadata never landed; PR stale since 2023, no reviews, only bulk label bump
• #8851 [HUDI-6281] Comprehensive schema evolution supports column change with a default value (SteNicholas, 2023-05-31, schema-evolution): Touches removed spark3.0.x/3.1.x modules and old InternalSchema pkg; feature relevant but diff unusable, CHANGES_REQUESTED unaddressed
• #8891 [Hudi-6318] The skip merge config for incremental-read ensures consistency in both stream and batch scenarios (zhuanshenbsj1, 2023-06-06, flink): Flink incremental-read skip-merge batch/stream consistency; 2-file PR stale since 2024-03, only comments no approval
• #8967 [HUDI-6378] allow to delete twice for an empty table (lvyanquan, 2023-06-14, write-client): Idempotent delete on empty-table (empty schema) edge case; BaseHoodieWriteClient rewritten to HoodieSchema, diff stale
• #9056 [HUDI-6456] [DOC] Add parquet blooms documentation (parisni, 2023-06-26, asf-site, docs): parquet_bloom docs not on asf-site; feature exists so doc still wanted, but PR stale with unaddressed review comments
• #9167 [HUDI-6519] The default value of read.streaming.enabled is determined by execution.runtime-mode (SteNicholas, 2023-07-10, flink): Flink config default-by-runtime-mode behavior change; debated (8 comments) with no consensus, abandoned 2024-09
• #9196 [WIP] Support rollback of instants with timestamp less than logcompaction (suryaprasanna, 2023-07-14, table-service): Rollback of instants < logcompaction; 2-yr WIP shell whose diff touches only test files, not reviewable as-is
• #9241 [HUDI-6573] Flink supports disable log append in mor table (zhuanshenbsj1, 2023-07-20, write-client): Disable-log-append config for MOR never landed (no such config on master); niche perf option, stale since 2024-02
• #9389 [HUDI-6667] ClientIds should generate next id automatically with random uuid instead of incremental id (SteNicholas, 2023-08-08, flink): Flink ClientIds still uses incremental id on master; concurrent-batch client-id conflict fix not landed, abandoned 2024-02
• #9544 [HUDI-6757] Fix compaction execution terminated in async threads in flink bounded (flashJd, 2023-08-26, flink): Flink bounded async-compaction terminated on close; CompactOperator.close() still closes executor so bug likely persists, PR stale no approval
• #9554 [HUDI-6760] Add SelfDescribingInputFormatInterface for hive FileInputFormat (Zouxxyy, 2023-08-28, hive): Spark schema-evolution then hive-rt read fails; SelfDescribing approach not on master, hadoop-mr refactored, diff stale (14 comments no approval)
• #9760 [HUDI-6887] Add test for Record Index and MIT queries (lokeshj1703, 2023-09-21, spark-sql): RLI+MIT data-skipping test coverage; approved but 2-yr stale and TestMergeIntoTable relocated to dml/others, needs rewrite
• #9794 [HUDI-6895] Change default timeline timezone from local to UTC (codope, 2023-09-27, timeline): UTC default still desirable but PR left WIP with TODO and needs migration plan; last discussion 2025-03 unresolved
• #9878 [HUDI-6953] Optimizing hudi sink operators generation (zhuanshenbsj1, 2023-10-18, flink): Flink COW insert-cluster/compaction/clean operator split not landed (only test [HUDI-6953] Adding test for composite keys with bulk insert row writer #10214 for 6953); pipeline evolved, needs rework
• #9889 [HUDI-6960] Support read partition values from path when schema evolution enabled (wecharyu, 2023-10-19, spark-reader): fix not landed; BaseFileOnlyRelation shouldExtractPartitionValuesFromPartitionPath/toHadoopFsRelation still on master, bug likely persists, PR stale
• #9905 [HUDI-6970] Stream read allows skipping archived commits (zhuanshenbsj1, 2023-10-23, flink): Flink stream-read OOM from loading archived commits still relevant; PR CONFLICTING, 2026 update is label churn, content stale since 2023
• #9922 [HUDI-6989] Stop handling more data if task is aborted & clean partial files if possible in task side (boneanxs, 2023-10-26, spark-writer): task-interrupt/partial-file cleanup improvement still relevant (HoodieWriteHandle/TaskContextSupplier present); PR stale since 2024, needs rebase
• #10207 [HUDI-7158] Expose event time config to record event time in commit file (zhangyue19921010, 2023-11-29, core): decoupling event-time from precombine still valid, but event-time subsystem reworked in 1.0 (feat(common): per-partition event-time rollup; decouple watermark tracking from EVENT_TIME_ORDERING #18778 rollup/watermark); PR stale, CI failing
• #10308 Incoming batch schema is not compatible with the table's one Incoming batch schema is not compatible with the table's one #9980 (njalan, 2023-12-12, hive-sync): schema-incompat from duplicate table names across hive schemas may still occur; niche fix, CI failing, no maintainer traction, PR stale
• #10326 [HUDI-6129] Support rate limit for Spark streaming source (boneanxs, 2023-12-14, spark-streaming): rate-limit for Spark streaming source (maxFilesPerTrigger/maxRowsPerTrigger) still wanted; re-impl of [HUDI-6129] Support rate limit for Spark streaming source #8796 left as early-stage draft, never finished
• #10396 [HUDI-7176] Add unit tests for log file only file slices (linliu-code, 2023-12-21, tests): log-only file-slice test coverage still useful; PR stacked on unmerged [HUDI-7240] Clean delete logic #10398, CI failing, stale since 2024
• #10406 [HUDI-5026] Fix ORC bootstrap read support in Spark (yihua, 2023-12-26, bootstrap-orc): ORC bootstrap read gap still real, but draft abandoned >2yr with CI failing; capture scope as issue
• #10462 [HUDI-7251] [WIP] Support partial schema for CDC workload (linliu-code, 2024-01-09, cdc): partial-schema for CDC append path still relevant but PR is WIP abandoned since 2024-05, CI failing
• #10526 [MINOR] Fix flaky testMultiWriterWithAsyncTableServicesWithConflict test (Zouxxyy, 2024-01-18, tests): test still present in TestHoodieClientMultiWriter and still flaky; PR CONFLICTING, review feedback unaddressed since 2024
• #10578 [HUDI-7404] Bloom execution improvements (the-other-tim-brown, 2024-01-28, bloom-index): bloom-filter perf improvements still relevant; PR left with open design questions, stale since 2024
• #10648 [MINOR] Support replacecommit rollback as part of rollbackFailedWrites (suryaprasanna, 2024-02-11, rollback): replacecommit rollback in rollbackFailedWrites still relevant; PR had empty desc, unaddressed review, stale since 2024
• #10727 [HUDI-7436] Fix the conditions for determining whether the records need to be rewritten (ThinkerLei, 2024-02-22, schema-evolution): rewrite-condition fix not landed (HoodieMergeHelper still on master); small bugfix, author stale since 2024, needs rebase
• #10846 [HUDI-7498] Fix schema for HoodieTimestampAwareParquetInputFormat (codope, 2024-03-11, hive-input-format): field-ordering fix not landed; HoodieTimestampAwareParquetInputFormat still on master; draft abandoned since 2024
• #11008 [MINOR] Make ordering deterministic in small file selection (the-other-tim-brown, 2024-04-12, spark-writer): deterministic small-file selection still useful for test stability; not landed, PR stale since 2024, likely needs rebase
• #11364 [HUDI-7814] Exclude unused transitive dependencies that introduce vulnerabilities (yihua, 2024-05-29, build-security): Transitive-dep vuln exclusions (log4j 1.x etc.) still worthwhile, but diff is CONFLICTING and touches removed flink1.14-1.18 poms; capture as dep-hygiene issue.
• #11537 [HUDI-7939] Validate file slices upto last completed commit in HoodieTableMetadataValidator (lokeshj1703, 2024-06-28, metadata-table): HoodieMetadataTableValidator still on master; validator correctness fix (compare views at last completed instant) worthwhile; draft abandoned 2024-09.
• #11573 [HUDI-6510] Support GHCI on Java 17 (CTTY, 2024-07-05, ci): Java 17 CI not landed (bot.yml still JDK 11); Java 17 CI coverage still wanted, but XL test-touching PR is stale/conflicting.
• #11607 [HUDI-3625][RFC-60] Add StorageStrategy to HoodieStorage (CTTY, 2024-07-10, storage-io): RFC-60 doc merged but StorageStrategy impl (this 90-file PR) never landed; too large/stale to rebase; capture RFC-60 implementation as issue.
• #11630 [HUDI-7986] Fix Duplicate handling behavior when Precombine value is not set (csivaguru, 2024-07-13, writer): Dedup-when-precombine-unset bug fix; PR is CONFLICTING and abandoned 2024-09; bug may still exist on master, capture as issue.
• #12014 [HUDI-7484] Enable url encoded partitioning if hive style partition enabled (codope, 2024-09-26, spark-sql): Auto-coupling url-encoding to hive-style partitioning not landed (configs still independent); real partition-dir usability issue, PR stalled in review 2024-10.
• #12076 [HUDI-8330] Fix default value of bootstrap table config (yihua, 2024-10-09, bootstrap): Bootstrap config default-value fix, small but unreviewed and abandoned 2024-10; bootstrap config code still on master, capture the default-value bug as issue.
• #12102 [HUDI-8367] Fix log file marker creation in file group initialization in metadata table writer (yihua, 2024-10-15, metadata-table): MDT log-marker creation fix stalled on CI since 2024-11; MDT writer since refactored by index abstraction feat(index): Add Indexer abstraction and refactor metadata table init… #18348/feat(index): Refactor metadata table update logic based on index abstraction #18372, capture bug as issue
• #12151 [HUDI-8263] Add validation for functional index in HoodieMetadataTableValidator (lokeshj1703, 2024-10-23, metadata-table): MDT validator still lacks expression/functional index validation; approach stale (FunctionalIndexSupport renamed ExpressionIndexSupport in 1.0)
• #12158 feat: updated blogs design (pintusoliya, 2024-10-24, asf-site, docs): Blog restyle touches 100 mdx files, open since 2024-10 with no progress; certain to conflict with blogs added since, capture redesign as issue
• #12232 [HUDI-8502] Fix the potential data loss issue in PartialUpdateAvroPayload#preCombine (usberkeley, 2024-11-11, writer): Real preCombine data-loss concern, reviewer-engaged (danny0405) but stalled 2024-12; PartialUpdateAvroPayload still on master, capture as issue
• #12265 Add integration test framework (yihua, 2024-11-15, ci): Docker-based integration-test framework for hudi-streamer never merged; abandoned 2024-11 with no reviews, capture scope as issue
• #12368 Added RCA and steps for resolving Skipping already committed batch (vidanimegh, 2024-11-28, asf-site, docs): Useful streaming-error RCA FAQ but target faq_writing_tables.md no longer in website/docs (docs restructured); capture content as issue
• #12407 [HUDI-3625][RFC-60] Update RFC-60 With Federated Storage Layout (zhangyue19921010, 2024-12-03, docs): RFC-60 federated storage layout design; engaged (19 reviews) but dormant since 2025-08 with POC [WIP][HUDI-3625][RFC-60] OSSStorageStrategy POC #12460 also stalled, track as issue
• #12490 [HUDI-8632] Fix bootstrap support with file group reader in compaction and clustering in Spark (yihua, 2024-12-14, bootstrap): Bootstrap + file-group-reader compaction/clustering support still a gap; stale since 2025-04, FGR merge-handle class reworked (native log format)
• #12536 [HUDI-8790] Fix most blocker/high sonarqube reported reliability issues (vinothchandar, 2024-12-24, reliability): Grab-bag of blocker/high sonarqube reliability fixes across 11 files, stale since 2025-03; issues still worth addressing, capture as issue
• #12546 [HUDI-8797] Fix the potential filesystem connection leak (Zouxxyy, 2024-12-27, core-io): Filesystem connection-leak fix is valuable but stuck as draft on unstable test, abandoned since 2025-01; capture leak as issue
• #12580 [HUDI-8830] use maxLatestCommitTime to compute splitLatestCommitDelay (LXin96, 2025-01-06, flink): Flink split-delay metric refinement; files still exist but stalled on danny0405 questioning the metric's meaning, unresolved
• #12582 [MINOR] Add file slice in delta write stats (codope, 2025-01-06, metadata-table): Populate file slice in delta write stats to optimize SI update; HoodieAppendHandle still exists, stale since Jan 2025, discussion inconclusive
• #12599 [HUDI-8858]Support ShowRecordLocationProcedure (houyuting, 2025-01-08, spark-sql): Procedure never landed (file absent on master); reviewers asked for a generalized index-agnostic procedure, author inactive
• #12642 [HUDI-8869] Refresh file index and reload timeline in HoodieBaseRelation (yuzhaojing, 2025-01-15, spark): Timeline-freshness fix in HoodieBaseRelation (file still exists) blocked by unresolved mutable-relation design debate; stale
• #12656 [HUDI-8685] Re-enabling flaky/buggy test (TestHoodieBackedTableMetadata.testCleaningArchivingAndCompaction) (nsivabalan, 2025-01-16, metadata-table): testCleaningArchivingAndCompaction still @disabled on master; re-enable still wanted but PR stale >1yr
• #12669 [HUDI-8072] Enable log file only test for Hive MOR tables (linliu-code, 2025-01-17, hive): Log-file-only Hive MOR test enablement; test files still exist, approved-then-changes-requested then abandoned Jan 2025
• #12732 [HUDI-8553] Support writing record positions to log blocks from Spark SQL UPDATE and DELETE statements (lokeshj1703, 2025-01-29, spark-sql): Core position-write infra landed ([HUDI-6795] Implement writing record_positions to log blocks for updates and deletes #9581); Spark-SQL UPDATE/DELETE enablement gap remains, PR stale since mid-2025
• #12774 [HUDI-7730] HoodieStorage.openSeekable should not have wrapStream param (ktblsva, 2025-02-05, storage): wrapStream param still present in HoodieStorage.openSeekable on master; stalled on reviewer pushback, author inactive
• #12795 [HUDI-8966] RFC-88: New Schema/DataType/Expression Abstractions (cshuo, 2025-02-06, rfc): RFC-88 doc never merged (rfc/rfc-88 absent); engine-neutral abstraction scope still relevant but design debate (decouple Flink perf) unresolved
• #12846 [DNM][HUDI-XXXX] Avoid caching base path in HoodieROTablePathFilter (xushiyan, 2025-02-16, branch-0.x, hive): base branch-0.x; base-path caching bug (metaClientCache) still present in HoodieROTablePathFilter on master, marked DNM
• #12861 [HUDI-9062] Make meta sync tools use global Hudi configs (xushiyan, 2025-02-20, branch-0.x, meta-sync): base branch-0.x; author confirms no master PR exists; global-config wiring for meta-sync tools still relevant on master
• #12871 [MINOR] Add test to prune partitions for TimestampBasedKeyGenerator (vinishjail97, 2025-02-21, spark): Partition-prune test for TimestampBasedKeyGenerator; HoodieFileIndex/prune files still exist, no reviews, stale >1yr
• #12878 [HUDI-8868] Close cachedAllInputFileSlices in BaseHoodieTableFileIndex (vinishjail97, 2025-02-24, file-index): cachedAllInputFileSlices spillable-map leak still present; class moved to core/read pkg so diff wont apply, bug relevant
• #12887 [HUDI-8907] Added RFC-86 draft document (jalpeshborad, 2025-02-27, rfc): RFC-86 DataFrame write-path draft never merged (rfc/rfc-86 absent); feature scope still valuable, draft abandoned
• #12938 [HUDI-9135] Bloom index data skew bucket index partitioner (nsivabalan, 2025-03-07, spark-index): Bloom-index data-skew bucketized partitioner fix not landed; BucketizedBloomCheckPartitioner exists, perf issue relevant, author stale
• #12939 [HUDI-9136] Fixing data skew with Bloom index sort partitioner (nsivabalan, 2025-03-07, spark-index): Bloom-index sort-partitioner data-skew fix not landed; SparkHoodieBloomIndexHelper exists, perf issue relevant, author stale
• #12941 [HUDI-9142] Read timestamp-millis logical type column as long correctly in Spark (yihua, 2025-03-08, spark-reader): Draft abandoned >1yr; timestamp-millis-as-long read correctness bug not fixed (HoodieParquetFileFormatHelper exists), scope valuable
• #12971 [HUDI-9169] Fix data loss in insert/delete/ insert-again cases (TheR1sing3un, 2025-03-13, spark-reader): Blocker data-loss (insert/delete/insert-again); record buffer rewritten into buffer/ pkg so diff obsolete, verify+refix on new code
• #13288 [HUDI-9172] Fix timestamp millis being read as micros for log files (jonvex, 2025-05-12, spark-reader): Real read-correctness bug (millis decoded as micros in log files); only a narrow related fix (fix: Converting ordering value with timestamp millis type properly fo… #13998) landed; draft 12mo stale
• #13342 [HUDI-9482] Fix the issue where the file slice was wrongly filtered out (TheR1sing3un, 2025-05-22, file-system-view): Real correctness bug (file slice wrongly filtered) with active review but community PR stalled 12mo; FSView/CompletionTimeQueryView refactored since
• #13422 [DRAFT] Avoid timeline server thrashing with async table services (the-other-tim-brown, 2025-06-12, table-service): Real stability issue (timeline-server thrashing under async table services) with substantial review discussion; draft never landed, 13mo stale
• #13628 [HUDI-9657] Test schema on read in fg reader (jonvex, 2025-07-26, tests-schema-on-read): SchemaOnReadEvolutionTestUtils not on master; schema-on-read fg-reader coverage gap real but PR stale since 2025-08
• #13713 [HUDI-9550] Index related doc update for hudi 1.1 (Davis-Zhang-Onehouse, 2025-08-12, asf-site, docs): Index docs for 1.1 still wanted but PR conflicting on asf-site with stray .gitignore edit; recapture as clean docs task
• #13807 [HUDI-9758] Allow reflection utils to use the ThreadContextClassLoader (linliu-code, 2025-08-29, common-reflection): Thread-context classloader config not on master (ReflectionUtils lacks it); useful capability but draft stale since 2025-08
• #13816 [HUDI-9775] Support for show_metadata_table_history with appropriate matching data table info (vamshikrishnakyatham, 2025-08-30, spark-sql): ShowMetadataTableHistoryProcedure not on master; new diagnostic procedure with 0 reviews, idle since 2025-09; capture feature as issue
• #13831 [HUDI-9714] Disallow changing merge related properties for HoodieStreamer (yjc408, 2025-09-03, hoodie-streamer): Fixes merge-property switching in HoodieStreamer, bug likely still on master; PR now CONFLICTING, review stalled since 2025-09
• #13889 [MINOR] Add tests for auto keygen for immutable and mutable workflow (linliu-code, 2025-09-13, spark-tests): [DRAFT] auto-keygen test additions, idle since 2025-11; coverage useful but draft stale -- capture as issue
• #13954 Add functional compat tests for Incremental Source (rahil-c, 2025-09-21, hoodie-streamer-tests): Incremental-source compat tests; mergeable but stalled after 20 reviews since 2025-10; capture coverage as issue
• #14041 test: add functional payload test cases for v6 (OverwriteWithLatestAvroPayload, DefaultHoodieRecordPayload) (rahil-c, 2025-10-02, spark-tests): v6 payload upgrade/downgrade functional tests; mergeable but idle since 2025-11; capture coverage as issue
• #14051 test: Add coverage for with OverwriteWithLatestAvroPayload with ordering fields set (rahil-c, 2025-10-03, spark-tests): Follow-up test to test: add functional payload test cases for v6 (OverwriteWithLatestAvroPayload, DefaultHoodieRecordPayload) #14041 for ordering fields; now CONFLICTING, idle since 2025-11; capture coverage as issue
• #14100 docs: create regression warnings for timestamp issue (jonvex, 2025-10-15, asf-site, docs): Draft adding timestamp-regression warnings to 0.14-1.0.2 release notes; useful content, draft stale since 2025-12; capture as issue
• #14101 feat: Add script to build and push Docker images (deepakpanda93, 2025-10-16, docker): Docker build/push helper scripts; now CONFLICTING, idle since 2025-11; capture tooling request as issue
• #14145 Fix CVE-2025-30065 (vamsikarnika, 2025-10-23, security): CVE-2025-30065 (parquet-avro); master still on parquet 1.13.1 so unaddressed; draft touches only gcp-bundle -- capture project-wide bump as issue
• #14173 refactor: Consolidate file writer side interfaces for new file format integrations (rahil-c, 2025-10-28, file-io): Exploratory refactor to consolidate writer interfaces for new file formats; not landed, idle since 2025-10; capture as issue
• #14176 fix: Remove table config hoodie.table.create.schema (yihua, 2025-10-29, table-config): hoodie.table.create.schema still on master (CREATE_SCHEMA present); removal draft stalled since 2025-10; capture cleanup as issue
• #14299 feat: update dependencies and to make trino-hudi-plugin compatible with trino 478. (RatulDawar, 2025-11-18, trino): Trino 478 dependency bump now outdated (Trino well past 478); plugin-compat scope still wanted, capture as issue
• #14303 feat: [HUDI-9775] Support for show_metadata_table_history with appropriate matching data table info (PavithranRick, 2025-11-20, spark-sql): New show_metadata_table_history procedure not on master; useful tooling but PavithranRick gone since Feb; capture as issue
• #14318 feat: [HUDI-9780] Concurrency Control test for Show Commit, Compaction and Clustering procedures (PavithranRick, 2025-11-21, spark-sql): Test-only concurrency-control coverage for procedures; no reviewer traction, author gone; capture as test-coverage issue
• #17460 feat: [HUDI-9765] Support for plans along with partition metadata in all the spark action procedures (PavithranRick, 2025-12-02, spark-sql): Large (28-file) spark-procedure plans+partition-metadata feature; not landed, PavithranRick gone since Feb; capture as issue
• #17463 feat: [HUDI-9734] Improve Hudi tooling to support file groups and specific partitions procedures (PavithranRick, 2025-12-03, spark-sql): File-group/partition history procedures tooling not on master; PavithranRick gone since Feb; capture as issue
• #17482 fix: HUDI- Skip partition path extraction for non partitioned tables (PavithranRick, 2025-12-04, writer): Skip partition-path extraction for non-partitioned tables fix; not on master, author gone since Feb; capture as issue
• #17489 fix: HUDI-17479 - Skip expression index for direct column access queries (PavithranRick, 2025-12-05, index): Skip expression index for direct column-access queries optimization; not on master, PavithranRick gone; capture as issue
• #17499 fix: ENG-27975 - Expression index (col stats) does not work for TIMESTAMP type (PavithranRick, 2025-12-05, index): Draft fixing expression-index col-stats for TIMESTAMP type; real bug, not fixed on master, author gone; capture as issue
• #17539 fix: [GI-17169] - Rollback inflight indexing does not clean table config inflight MDT partition (PavithranRick, 2025-12-10, metadata-table): Draft (test-only) for rollback-inflight-indexing not cleaning MDT partition config; real bug, no fix, author gone; capture as issue
• #17604 fix: [HUDI-8401] Unifying Partial Update modes and Merge Into Partial Update Encoding (PavithranRick, 2025-12-16, writer): HUDI-8401 partial-update unification; nsivabalan CHANGES_REQUESTED then stalled, merge subsystem churned on master; capture scope as issue
• #17759 fix: Reproduce nested partition columns pruning data validation failure (vinishjail97, 2025-12-31, spark/file-index): repro of real nested-partition-column pruning data-validation bug (since [HUDI-6941] Fix partition pruning for multiple partition fields #9863) still on master; PR is investigation, no fix, 0 reviews
• #17804 fix: Revert Partition stats should be controlled using column stats config (fix: Partition stats should be controlled using column stats config #14165) (PavithranRick, 2026-01-08, metadata-table): reverts fix: Partition stats should be controlled using column stats config #14165 (on master); concern to toggle partition stats independently of col stats still valid; 0 reviews, abandoned
• #18073 fix: GI-18072 - Allow retries on marker file creation requests (PavithranRick, 2026-02-02, markers): Idempotent marker-creation retry on timeout still relevant on master, but PR stalled since 2026-02-23 and author (intern) inactive; capture in issue
• #18152 fix: Support out of order schema evolution fix (linliu-code, 2026-02-09, release-0.14.2, meta-sync): Targets frozen release-0.14.2, CONFLICTING; recreate-table-on-incompatible-schema meta-sync capability may be wanted on master, capture as issue
• #18154 fix: Set credentials for Hive metastore based lock provider (linliu-code, 2026-02-09, release-0.14.2, hive-sync): Targets frozen release-0.14.2, CONFLICTING; HiveMetastoreBasedLockProvider still on master so credential fix may apply there too
• #18157 fix: Support schema evolution with nested data type (linliu-code, 2026-02-09, release-0.14.2, schema-evolution): Targets frozen release-0.14.2, CONFLICTING; underlying bug Array of struct corruption after upsert with sparkRecordMerger when schema evolution on the struct and data written with insert #17454 still OPEN and HoodieInternalRowUtils.scala still on master, so it applies to master
• #18248 Fast bootstrap rfc (vamshipasunuru1, 2026-02-25, rfc): rfc-102 number now assigned to embeddings RFC on master, so this doc collides; fast-bootstrap design (issue Support REGISTER_ONLY bootstrap mode for cold storage partitions #18135) still relevant, re-file under new RFC number
• #18305 [HUDI-7503] logcompaction execution should fail if another active writer is already executing the same plan (kbuci, 2026-03-11, table-service): Companion [HUDI-7503] Compaction execution should fail if another active writer is already executing the same plan #18012 (HUDI-7503 compaction) merged as ddfcc92; logcompaction gap remains but PR conflicting/stale since Apr
• #18339 feat: Fix scheduling issues caused by ForkPoolJoin in AverageRecordSizeEstimator (rahil-c, 2026-03-18, writer): Estimator relocated to execution/estimator and now uses plain stream(); draft conflicting/stale since Apr, diff outdated
• #18346 [HUDI-7946][RFC-90] Add support for cancellable clustering table service plans (kbuci, 2026-03-19, rfc-clustering): RFC-90 doc re-post of deleted [Duplicate please see #18346] Add support for cancellable clustering table service plans #12856; rfc-90 not on master, zero engagement, untouched since creation
• #18419 feat(metadata): Support hfile index block cache for hfile reader (cshuo, 2026-03-30, metadata-hfile): Base block cache already landed via [HUDI-7219] Add caching support for HFileBlocks #13724; reader-metadata cache extension distinct but conflicting/no-review since Apr
• #18458 feat: Add option to skip invalid JSON records in HoodieStreamer (yaojiejia, 2026-04-03, streamer): Skip invalid JSON records (closes Invalid json input record crashes HoodieDeltaStreamer #15942); feature relevant but conflicting/stale since Apr, non-committer author
• #18608 feat(schema) Ensure spark SQL insert into preserves hudi_type for VEC... (rahil-c, 2026-04-27, spark-sql): Real bug (VEC/BLOB lose hudi_type metadata on INSERT INTO, schema-compat fails) but stale ~3mo draft, conflicting; capture bug as issue
• #18655 Log structured schema-change diffs in DeltaStreamer (linliu-code, 2026-04-29, deltastreamer): SchemaChangeLogger enhancement still relevant (StreamSync exists) but stale ~2mo unreviewed draft; capture as issue
• #18810 test(spark): repro for bytes->string promotion crashing data-skipping read (linliu-code, 2026-05-22, data-skipping): Test-only repro of real bytes->string data-skipping ClassCastException; no human review, CI failing; capture bug as issue

[voonhous]

KEEP (223)

Tick the box once the PR has been re-reviewed / rebased / merged (or closed after all -- note it in a reply).

• #7129 [MINOR] Support column type evolution for Hive (fsilent, 2022-11-03, hive): Column type evolution for Hive via SelfDescribingInputFormat; 2 approvals, HoodieParquetInputFormatBase still on master, only pom-dep cleanup + jira pending. Worth rebasing.
• #8338 [HUDI-5996] Verify the consistency of bucket num at job startup (LinMingQiang, 2023-03-31, flink): genuinely active: rangareddy re-review request 2026-07-10 plus danny0405/hudi-agent; only needs rebase (CONFLICTING) on live flink bucket-num guard
• #8526 [HUDI-6116] Optimize log block reading by removing seeks to check corrupted blocks. (prashantwason, 2023-04-20, storage): APPROVED twice by danny0405; optimization not yet landed (HoodieLogFileReader still does eager isBlockCorrupted seek); worth rebasing
• #9007 [HUDI-6405] Fix incremental file sync for clustering and logcompaction (suryaprasanna, 2023-06-17, table-service): Blocker incremental sync fix; IncrementalTimelineSyncFileSystemView still on master, has approvals, updated 2025-12 by committer
• #9035 [HUDI-6416] Completion markers for handling execution engine (spark) retries (nbalajee, 2023-06-22, write-client): Blocker completion-markers for spark stage retries; active committer PR updated 2026-02, all touched files present on master
• #9188 [HUDI-6528] Fix premature RDD unpersist during index lookup (xushiyan, 2023-07-13, spark-index): Critical index-lookup RDD caching perf fix; draft but updated 2025-12 by committer, target files present on master
• #9492 [RFC-70]: Add Reverse streamer RFC (pratyakshsharma, 2023-08-21, rfc): Reverse-streamer design proposal; feature not yet built, active RFC discussion updated 2025-08, cheap to keep for review
• #10876 [HUDI-7512] sort input records for insert operation (bhat-vinay, 2024-03-18, spark-writer): active feature (UpsertSortPartitioner for INSERT, not landed) with substantive maintainer discussion Dec 2025; CONFLICTING, needs rebase
• #11793 [HUDI-8033] RFC-81: Hudi Data Ordering (usberkeley, 2024-08-19, rfc): RFC-81 Data Ordering actively under review; extensive committer discussion with real activity through 2025-11.
• #12270 [HUDI-4468] Simplify TimeTravel logic for Spark (KnightChess, 2024-11-17, spark-sql): Active PR: 19 reviews, updated 2026-02-25 with author pinging reviewer (jonvex); still under live review
• #12393 [HUDI-8605] Fix Trino failure when reading corrupted block at end of log file (usberkeley, 2024-12-02, log-format): HoodieLogFileReader corrupted-block fix still relevant; active review request from yihua (2026-02), applies to current master
• #12468 add presto native documentation in the hudi site (werowe, 2024-12-11, asf-site, docs): Adds new presto_hudi_connector.md (not yet on site); committer bhasudha explicitly owns and will review+merge, still wanted
• #12747 [HUDI-8945] Rename merger classes to be consistent with merge modes (yihua, 2025-01-31, spark-reader): Merger-class rename not landed (new names absent on master); active committer yihua rebased it in 2026, still applicable
• #12794 [HUDI-8970] Change scheduleandexecute in RunCompactionProcedure to combined (voonhous, 2025-02-06, spark-sql): RunCompactionProcedure schedule+execute combine; already APPROVED, only a minor follow-up fix pending, file still exists
• #12952 fix: Allow bootstrap to complete without updating col stats (CTTY, 2025-03-10, bootstrap): Bootstrap-without-colstats fix; BaseCommitActionExecutor exists, author active as of Nov 2025 and states bug affects 1.0.2
• #13147 feat(timeline-ui): [HUDI-9315] Add Hudi Timeline UI (voonhous, 2025-04-15, timeline-ui): Active RFC-94 feature PR, updated 2026-07-16 (today); RFC merged via feat(ui): [HUDI-9315] Add RFC-94: Hudi Timeline UI #13152, implementation under active development
• #13450 fix: Bloom Index can fail with async compaction (the-other-tim-brown, 2025-06-17, index-bloom): Real bug fix closing issue Bloom index inspects base file from incomplete compaction #17059; touched files present on master, updated 2026-02, worth reviewing
• #13499 [HUDI-9560][RFC-97] Deprecate Hudi Payload Class Usage (linliu-code, 2025-06-29, rfc): RFC-97 not on master; documents active HUDI-8401 payload-deprecation direction, RFC doc worth landing
• #13577 [HUDI-9663] Provide new TransactionManager functionality and move instant time management to Transactionmanager (the-other-tim-brown, 2025-07-18, transaction-timeline): Large committer-led refactor, updated 2026-03 with approvals; currently conflicting, needs rebase but still relevant
• #13582 [HUDI-9610] Support to get property from catalog for copy_to_temp_view call (hehuiyuan, 2025-07-20, spark-sql-procedures): Small self-contained CopyToTempViewProcedure enhancement; target files present on master, nearly applies
• #13606 [HUDI-9634] Archival considers retaining the the earliest retain instant in the clean plan (TheR1sing3un, 2025-07-24, archival-clean): Real improvement preserving clean incremental scanning; active review, CleanPlanner/TimelineArchiverV2 present on master
• #13697 [MINOR] Refactor index definition reading in HoodieTableMetaClient (yihua, 2025-08-08, metadata-table): Approved 1-file readability refactor of index-definition reading; low-risk, still applies, worth landing
• #13745 [HUDI-9734] Improve Hudi tooling to support file groups and specific partitions procedures (vamshikrishnakyatham, 2025-08-22, spark-sql-procedures): New show_file_group_history procedure not on master; heavy review engagement plus an approval, worth finishing
• #13793 [HUDI-9765] Support for plans along with partition metadata in all the spark action procedures (vamshikrishnakyatham, 2025-08-28, spark-sql-procedures): Unified show-procedures feature not on master, updated 2025-12 with active review; worth reviewing/rebasing
• #13881 [HUDI-9791][DOCS] Update HFile format for compatibility with HBase (linliu-code, 2025-09-11, docs): hfile_format.md live on master; small accurate doc update for HBase HFile compat, 1 review, applies -- worth merging
• #13888 test: that exception is thrown when malformed union is used (jonvex, 2025-09-12, hudi-common-tests): Tiny unit test for merged fix fix: throw exception instead of return original schema #13885 (throw on malformed union); behavior on master, test not; applies cleanly
• #13925 feat: integrate BUCKET_INDEX for HoodieJavaClient (vinishjail97, 2025-09-18, java-client): BUCKET_INDEX for HoodieJavaClient not on master; real feature, active author, updated 2026-02 -- worth reviewing
• #14154 Ensure MOR table works, with lance base files and avro logs file (rahil-c, 2025-10-24, feature-branch-rfc100-unstructured-data, rfc100-lance): RFC-100 Lance MOR validation on active feature branch; CONFLICTING, needs rebase and depends on reader PR Support COW bulk-insert, insert, upsert, delete works with spark datasource and lance #14137
• #14156 Validate that basic compaction and clustering works with Lance (rahil-c, 2025-10-25, feature-branch-rfc100-unstructured-data, rfc100-lance): RFC-100 Lance compaction/clustering validation on active feature branch; CONFLICTING, needs rebase
• #14160 Add lance schema evolution support for add column (rahil-c, 2025-10-26, feature-branch-rfc100-unstructured-data, rfc100-lance): RFC-100 Lance schema-evolution (add column) on active feature branch; updated 2026-04, CONFLICTING, needs rebase
• #14193 Validate vector and blob content support (rahil-c, 2025-11-02, feature-branch-rfc100-unstructured-data, rfc100-lance): RFC-100 Lance vector/blob validation on active feature branch; CONFLICTING, needs rebase
• #14234 fix: aws glue sync set comment (parisni, 2025-11-07, hive-sync): AWS Glue comment regression fix (from [HUDI-6638] Upgrade AWS Java SDK to V2 #9347); APPROVED and MERGEABLE, updated 2026-02 -- ready to merge
• #14235 [HUDI-5533] Support spark nested columns comments (parisni, 2025-11-07, spark): Real nested-column-comments feature (HUDI-5533); CONFLICTING but parisni active and danny0405/yihua engaged; needs rebase, worth reviewing
• #14248 feat: Support schema on write for flink (cshuo, 2025-11-12, flink): Schema-on-write for Flink by active committer cshuo; CONFLICTING and dormant since Dec but wanted feature (issue 14252), needs rebase
• #14255 docs: RFC-103 - Support Vector Indexing in Apache Hudi (suryaprasanna, 2025-11-13, rfc): Active RFC-103 vector indexing under review; vinothchandar CHANGES_REQUESTED, updated Apr 2026; keep for ongoing RFC discussion
• #14257 [14256] Add source path field in table created from S3/GCS source (kroushan-nit, 2025-11-13, utilities): Small S3/GCS source-path field enhancement; not on master, nsivabalan/yihua reviewed; nearly applies, worth reviewing
• #14286 [HUDI-9159] Update Storage LP rfc with new retry scenario and proposed handling (alexr17, 2025-11-14, rfc): Small doc update to active RFC-91 storage LP retry handling; yihua engaged and author responded; worth landing
• #17464 fix: CREATE TABLE LIKE command (mansipp, 2025-12-03, spark-sql): Small CREATE TABLE LIKE bug fix; CreateHoodieTableLikeCommand still on master, yihua reviewed; nearly applies, worth reviewing
• #17591 fix: Fixing unions for partitioned RLI bootstrap (nsivabalan, 2025-12-14, metadata-table): RLI-bootstrap union fix by committer nsivabalan; cshuo APPROVED, updated Jun 2026; CONFLICTING but active, keep to rebase/merge
• #17675 refactor: simplify the ReflectionUtils#getTopLevelClassesInClasspath (VahidRamezaniDB, 2025-12-22, common): Small ReflectionUtils.getTopLevelClassesInClasspath simplify refactor; method still verbose on master, yihua reviewed; still applies
• #17683 perf: Improving data skipping with column stats when there exist colu... (cshuo, 2025-12-23, data-skipping): perf improvement for column-stats data skipping, linked open issue Improving ColumnStatsProbe when filter include columns without column stats index #17598, 2 reviews; conflicting, needs rebase
• #17724 fix: Make MDT cleaner/rollback/finalize-write parallelism configurable (nsivabalan, 2025-12-26, metadata-table): small config change, rebased and feedback addressed June 2026 by committer; conflicting, needs re-rebase
• #17725 feat: Allow non-registry schema provider in debezium source (CTTY, 2025-12-26, deltastreamer): small 1-file self-contained feature, DebeziumSource still on master, likely still applies
• #17766 feat: Add floor size file limit config to SparkSizeBasedClusteringPlanStrategy (rahil-c, 2026-01-01, clustering): small self-contained clustering config + test on existing files, likely still applies
• #17796 MDT Test framework without writing data files (vamsikarnika, 2026-01-07, metadata-table): active MDT test/benchmark framework, 8 reviews, nsivabalan driving with deliverables; stalled 5mo but heavy investment
• #17806 [HUDI] FIX: Fixed incomplete cleanup of temporary directories generat... (qukaikai123, 2026-01-08, hive-sync/test): small test-util temp-dir cleanup fix in HiveTestUtil, self-contained, likely still applies
• #17863 fix: Add validation for invalid HoodieWriteStat with zero counts (vinishjail97, 2026-01-14, writer/validation): small self-contained write-stat validation + new exception on existing HoodieTable, likely still applies
• #17975 fix: kafka delay metrics reporting when offsets caught up (alexr17, 2026-01-21, streaming/kafka): active PR, updated 2026-07-13, 5 reviews, small KafkaOffsetGen fix + test
• #17979 feat(lance): Support Lance data log blocks (the-other-tim-brown, 2026-01-22, lance): committer feature PR in active Lance area; vinothchandar engaged, moved to draft pending inline-read decision
• #18026 fix: improve incremental streaming thread safety (linliu-code, 2026-01-28, spark/incremental): thread-safety fix on IncrementalRelationV1/V2, committer lokeshj1703 added test, touches current files
• #18040 feat: support watermark emitter from split info (HuangZhenQiu, 2026-01-29, flink): Flink watermark emitter, 24 review comments, linked open issue Hudi Watermark Record Emitter #14424, small focused change
• #18054 [MINOR] Add database name to the metadata table config (Rajeev-01, 2026-01-30, metadata-table): small 1-file MDT database-name config, danny0405/codope engaged, use case explained, likely still applies
• #18061 fix(spark): improve error message when precombine field value is null (prashantwason, 2026-01-30, spark-sql): Committer PR, rebased and updated 2026-07-02; low-risk error-message improvement in HoodieCreateRecordUtils, closes Improve error message when precombine field value is null during payload creation #18060, still applies
• #18121 fix: Adding support for encoding with Timeline Server interaction (nsivabalan, 2026-02-07, timeline-server): Committer PR updated 2026-04-05; small UTF-8 charset override for timeline-server comms, touched files still on master, worth rebasing
• #18159 feat(blob): Blob cleaner updates (the-other-tim-brown, 2026-02-09, blob-cleaner): RFC-100 blob cleaner impl by RFC owner, closes [Milestone 1] Clean managed blob files #18111; cleaner not yet on master, active feature area, worth rebasing (uncertain: plan restructured via feat(blob): update approach to remove reliance on column groups, break down plan #18013, work also on tim/blob-cleaner branch)
• #18182 feat(metadata): Support sub-directory bucketing for MDT partitions (prashantwason, 2026-02-11, metadata-table): Committer feature updated 2026-06-07; bucketing configs not on master (not landed), backward-compatible opt-in, worth reviewing
• #18215 perf(metadata table) Metadata table archival should proceed until earliest commit to retain on data table (kbuci, 2026-02-17, metadata-table): Committer perf PR updated 2026-05-22; getEarliestInstantForMetadataArchival still 2-arg on master (not landed), helpers absent, applies to master
• #18237 fix: Fixing repeated clean attempts using same instant times in mdt when data table has out of order commits (nsivabalan, 2026-02-22, metadata-table): Committer PR updated 2026-06-17; master executeClean in TableVersionSix still lacks the existing-clean-instant guard, fix applies (version6 still supported)
• #18247 chore(deps): bump io.airlift:aircompressor from 0.27 to 2.0.3 (dependabot[bot], 2026-02-25, dependencies): Master still on airlift.version 0.27; dependabot bump to 2.0.3 addresses CVE-2025-67721, needs maintainer eval of 2.x API break, auto-maintained
• #18252 [HUDI-XXXXX] fix: Parallelize cloud object existence checks in S3EventsHoodieIncrSource (vinishjail97, 2026-02-26, deltastreamer): Committer perf PR updated 2026-03-18, CI green; getCloudObjectMetadataPerPartition still on master and EXISTS_CHECK_PARALLELISM not landed, real 5h->fast fix applies
• #18276 docs: [RFC-98] Design doc of DSv2 read support (geserdugarov, 2026-03-04, spark-sql): Only the RFC-98 claim skeleton ([HUDI-4449] Claim of RFC-98, Spark Datasource V2 Read #13609) is on master; full design doc still under active discussion with vinothchandar/yihua, benchmarks in progress ((WIP) benchmark: [RFC-98] COW table read performance comparison #18351)
• #18277 feat(dsv2): [RFC-98] DSv2 read for COW (geserdugarov, 2026-03-04, spark-sql): Active RFC-98 DSv2 COW read impl; updated 2026-07-12, 60 review rounds; core in-progress feature
• #18293 refactor(spark): remove redundant SparkAdapter methods (yihua, 2026-03-08, spark): Refreshed against master 2026-07-02 (covers Spark 3.3-4.2); only rebase-conflicting, active committer cleanup
• #18338 (WIP) feat: Add Claude Code Skills for Hudi (yihua, 2026-03-18, dev-tooling): WIP .claude/skills dev tooling, still MERGEABLE, updated 2026-06-25; active committer work
• #18350 feat: Provide new TransactionManager functionality and move instant time management to Transactionmanager (danny0405, 2026-03-19, core-transaction): Active core refactor by danny0405; cc'd vinothchandar for final review; needs 100-file rebase but core direction
• #18359 feat(blob): Supplementing RFC-100 with blob cleaner design (voonhous, 2026-03-20, rfc-blob-cleaner): Active RFC-100 blob cleaner design docs, MERGEABLE, updated 2026-07-02, under review
• #18374 feat: Support spark nested columns comments (danny0405, 2026-03-24, hive-sync): Nested column comment sync fix by danny0405, still relevant; conflicting since May, committer can rebase
• #18378 fix: Fixing partitioned RLI for nested partitions (nsivabalan, 2026-03-24, metadata-table): RLI nested-partition dot-encoding fix, MERGEABLE and APPROVED, updated 2026-06-23; ready to land
• #18384 feat: Support selective meta field population (prashantwason, 2026-03-25, writer): Selective meta fields (closes Support selective meta field population #18383), recent 2026-06-17; changes-requested + 87-file conflict but linked issue, committer
• #18396 feat(core): Pass the number of updates from workload profiling through to merge handle (yihua, 2026-03-26, writer): Merge-handle update-count infra by yihua; 32-file conflict since May, committer can rebase
• #18401 feat: Update CachedSchemaRegistryClient to have configs passed in constructor which can be used for gcp schema registry (dharmendersheshma, 2026-03-27, utilities): SchemaRegistry gcp auth configs, MERGEABLE, active updated 2026-07-10, under review
• #18425 feat(vector): Support complex types with VECTORS (rahil-c, 2026-03-31, vector): VECTOR nested/complex type support (RFC-99), active draft updated 2026-06-27, conflicting; keep for rebase
• #18445 fix: increase confluent version (linliu-code, 2026-04-01, build-deps): Confluent bump to 7.5.1; master still 5.5.0, yihua actively requested upgrade in Jul; MERGEABLE, still wanted
• #18466 fix(packaging): restore protobuf shade relocation in utilities bundles (officialasishkumar, 2026-04-04, packaging): Protobuf relocation still absent on master (closes ClassCastException: com.google.protobuf.Timestamp cannot be cast to IndexedRecord during merge when using Protobuf-derived Avro schemas #18242), MERGEABLE valid packaging fix
• #18468 feat(flink): Support emit watermark in the end of each split (HuangZhenQiu, 2026-04-04, flink): Per-split watermark (closes Hudi Watermark Record Emitter #14424); author responsive, committers cc'd, wanted feature; conflicting, needs rebase
• #18492 fix: Fix lock expiration metric (linliu-code, 2026-04-11, concurrency): Bug still present on master (metric uses oldExpirationMs L637); MERGEABLE, active updated 2026-07-01
• #18504 feat(core): implement file-level sorting for upsert ([Phase 1] Implement file-level sorting for upsert #18357) (kartikeyaagrawal, 2026-04-15, core): RFC-103 LSM phase-1 feature impl; blocked on open IS_SORTED persistence design Q to yihua, needs rebase (conflicting), not abandoned
• #18524 chore(docker): address PR chore: Add Java 17 Hadoop base image and Spark 4.0.1 docker compose s… #18520 review comments for Spark 4.0.1 stack (voonhous, 2026-04-17, docker): Follow-up to merged chore: Add Java 17 Hadoop base image and Spark 4.0.1 docker compose s… #18520 (Spark 4.0.1 docker stack, merged 2026-04-17); real review-comment cleanup, needs rebase (conflicting)
• #18526 feat(utilities): Add HoodieStashPartitionsTool for backing up and removing partitions (nsivabalan, 2026-04-18, cli): Active PR updated 2026-07-16, mergeable, 16 review rounds, closes [WIP][to be discussed] Add HUDI APIs for "stashing" and "restoring" partitions #17866; new backup/remove-partitions tool worth reviewing
• #18528 chore(deps-dev): bump com.fasterxml.jackson.core:jackson-core from 2.13.5 to 2.18.6 in /hudi-azure (dependabot[bot], 2026-04-18, deps): Routine dependabot jackson-core test-scope bump in hudi-azure, mergeable, recent (05-18); maintainer to merge or bot to supersede
• #18554 Adding hudi-cli-mcp to assist AI assistants to help and manage hudi tables (nsivabalan, 2026-04-23, cli): Active new hudi-cli-mcp module, updated 2026-07-10, mergeable, in normal review flow
• #18572 test: Add coverage for deleteMdtIfNecessaryBeforeRestore API (ashokkumar-allu, 2026-04-23, metadata-table): Test coverage addition, mergeable, updated 2026-05-23 with 10 review rounds; applies to master
• #18574 fix: ProtoConversionUtil UINT64 default fails Avro 1.12 FIXED validator (tiennguyen-onehouse, 2026-04-23, utilities): Real bug fix (UINT64 default vs Avro 1.12 FIXED validator), mergeable, updated 2026-06-30, closes [BUG] ProtoConversionUtil UINT64 default fails Avro 1.12 FIXED validator #18573
• #18597 test(schema): Add lance fileformat test for custom types on MOR (voonhous, 2026-04-26, schema): Lance custom-types MOR test, recent (2026-06-14), part of active custom-types work, needs rebase (conflicting)
• #18611 fix(metadata-table): add config to skip zero-size data files on MDT initialization (nada-attia, 2026-04-27, metadata-table): Active PR updated 2026-07-14, mergeable, 12 review rounds, closes Zero-size data files during MDT bootstrap cause rebootstrap job failures #18612; MDT init skip-zero-size config
• #18637 [MINOR] Resolve hasOperationField from table config to avoid reading a data file (vinishjail97, 2026-04-28, core): Clean minor perf fix (avoid sequential I/O reading data file for hasOperationField), mergeable, applies to master
• #18642 Hudi-trino upgrade to 480 (ad1happy2go, 2026-04-29, trino): Trino plugin upgrade to 480, recent draft (updated 2026-06-09), mergeable, valuable active work
• #18661 docs(rfc): Add RFC-104 to unify schema evolution on schema-on-read (voonhous, 2026-04-30, docs): RFC-104 schema-evolution proposal doc, recent (2026-06-11), mergeable, in review flow
• #18665 feat(blob): Accept partial {type,data} or {type,reference} structs on write (voonhous, 2026-04-30, blob): Active blob-write feature, updated 2026-07-11, mergeable, closes Quality of Life improvements for Blob writes #18635, part of RFC-100 unstructured-data work
• #18666 test: Add VARIANT/VECTOR/BLOB to HoodieTestDataGenerator default schema (yihua, 2026-04-30, tests): Custom-types test-data coverage by top committer, updated 2026-05-25, needs rebase (conflicting)
• #18667 test: Surface unstructured types in TestSavepoint and TestMultiFS (yihua, 2026-04-30, tests): Unstructured-types test coverage by top committer, updated 2026-05-20, needs rebase (conflicting)
• #18682 type:devtask auto-infer DefaultSparkRecordMerger for Lance base file f... (rahil-c, 2026-05-03, lance): Lance record-merger auto-infer, recent draft (2026-06-15), part of active Lance base-file work; needs rebase
• #18683 type:story Add Parquet DESCRIPTOR mode for blob inline reading (rahil-c, 2026-05-03, blob): Parquet DESCRIPTOR blob inline-read mode, recent (2026-06-15), 5 review rounds; needs rebase (conflicting)
• #18685 feat(flink): add metrics for RLIBootstrapOperator (HuangZhenQiu, 2026-05-04, flink): Flink RLIBootstrapOperator metrics, actively reviewed (11 rounds), updated 2026-06-15; needs rebase
• #18687 docs(docker): document build_docker_images.sh flags (201573, 2026-05-04, docs): Docs for docker/build_docker_images.sh (exists on master), closes Documentation update for build_docker_images.sh #18503, author responsive (CI green); needs rebase
• #18690 feat(flink): expose metadata table metrics to Flink operators (HuangZhenQiu, 2026-05-04, flink): Flink MDT metrics feature, updated 2026-05-20, 8 review rounds; needs rebase (conflicting)
• #18692 feat: Add ClusteringGroupWriter SPI for pluggable clustering writes (vinishjail97, 2026-05-06, clustering): Pluggable ClusteringGroupWriter SPI, mergeable, recent (2026-05-21), in review flow
• #18699 [HUDI-18691] Honor IF NOT EXISTS when creating indexes (201573, 2026-05-07, spark-sql): Honor IF NOT EXISTS on CREATE INDEX, mergeable, updated 2026-05-20, closes [BUG] CREATE INDEX IF NOT EXISTS record_index throws "Index already exists" — ignoreIfExists flag is silently dropped #18691; applies to master
• #18715 [HUDI-18496] Respect configured record type when filtering index keys (201573, 2026-05-10, index): Recent, mergeable, CI green, awaiting review; closes HoodieIndexUtils.filterKeysFromFile hardcodes AVRO fallback, ignoring configured record merger type #18496 record-type index-key filtering
• #18718 fix: report renewed storage lock expiration metric (officialasishkumar, 2026-05-11, storage-lock): Recent lock-metric fix, mergeable, active review (wombatu-kun), closes wrong metric from renewLock() function #18493
• #18721 [HUDI-9624] Warn when event-time field is set without watermark tracking (1fanwang, 2026-05-12, write-config): Valid small warn on misconfigured event-time field; conflicting, needs rebase; not superseded
• #18726 feat(spark): Make Hudi's analyzed plans introspectable to lineage tooling (codope, 2026-05-13, spark-sql): Committer PR exposing MERGE plan children for lineage; conflicting, needs rebase
• #18729 docs: add batch vector search demo + notebook for RFC-102 (rahil-c, 2026-05-13, docs): Recent RFC-102 vector-search demo, mergeable, shangxinli reviewed; active feature area
• #18770 perf(spark): fast-path SELECT count() on COW tables via parquet footer row counts (SELECT count(*) on Hudi COW tables reads full file contents where vanilla Spark+Parquet reads only row counts from footers (Hudi 1.1.1, Spark 3.4) #18769) (linliu-code, 2026-05-18, spark): Recent count() footer fast-path perf win, active review; conflicting, needs rebase
• #18776 fix: close the file writers properly in fail cases (danny0405, 2026-05-19, writer-io): Committer resource-leak fix on writer fail paths; nsivabalan reviewed; conflicting, needs rebase
• #18792 fix: Fix col-stats corruption from NaN values and stats-truncation (linliu-code, 2026-05-20, metadata-table): Real col-stats NaN corruption/data-skipping correctness fix ([SUPPORT] Single NaN value corrupts col-stats and silently drops query rows across whole table (regression in 1.1.1) #18754); conflicting, needs rebase
• #18794 fix(spark): honor spark.sql.parquet.outputTimestampType on the write path ([SUPPORT] Spark write path silently ignores spark.sql.parquet.outputTimestampType and hoodie.parquet.outputtimestamptype #18752) (linliu-code, 2026-05-20, spark): Real write-path timestamp-type interop fix ([SUPPORT] Spark write path silently ignores spark.sql.parquet.outputTimestampType and hoodie.parquet.outputtimestamptype #18752); conflicting, needs rebase
• #18795 feat(rollback): add RollbackOrphanDetector utility for archive orphan guard (Rollback orphan parquet files become reader-visible duplicates / corruption after archive prunes the rollback instant #18783) (shangxinli, 2026-05-20, rollback): Active discussion on Rollback orphan parquet files become reader-visible duplicates / corruption after archive prunes the rollback instant #18783; mergeable, CI green; note maintainers could not reproduce the issue
• #18801 chore(deps): bump org.apache.commons:commons-configuration2 from 2.11.0 to 2.15.0 in /packaging/hudi-cli-bundle (dependabot[bot], 2026-05-21, deps): Recent dependabot bump, mergeable, CI green; maintainer merge decision
• #18802 feat(utilities): propagate per-partition event-time watermarks from upstream Hudi sources (shangxinli, 2026-05-21, utilities): Phase 2 of RFC: Commit-Level Partition Min/Max EventTime Metadata in Apache Hudi #17512 watermark propagation, builds on feat(common): per-partition event-time rollup; decouple watermark tracking from EVENT_TIME_ORDERING #18778; conflicting, needs rebase
• #18806 test(spark): repro for reconcile.schema=true blocking documented type promotion (linliu-code, 2026-05-21, schema-evolution): Repro test with active committer engagement (yihua 2026-06-17); mergeable
• #18807 test(spark): codify MDT col_stats behavior on ADD COLUMN schema evolution (linliu-code, 2026-05-21, metadata-table): Repro test with active reviewer engagement (wombatu-kun 2026-06-14); mergeable, CI green
• #18812 feat(flink): Support writing top-level VARIANT columns via Flink SQL and add initial write/table/service spark compat tests (kbuci, 2026-05-22, flink): Active Flink VARIANT write feature (Support writing top level VARIANT columns for flink writer #18870), heavy danny0405 review; conflicting, needs rebase
• #18814 fix(common): normalize hudi table base path in implicit-key lock providers (Davis-Zhang-Onehouse, 2026-05-22, lock-provider): Recent lock base-path normalization fix, mergeable, active review (yihua 2026-06-25)
• #18816 fix(tests): fix hudi-cli tests to use current table version and enable in CI (kavya685, 2026-05-22, ci): Re-enables excluded hudi-cli CI tests on table v1.x (Run hudi-cli tests in Azure CI #16448); danny0405 reviewed; conflicting, needs rebase
• #18817 fix(common): normalize hudi table base path in StorageBasedLockProvider (Davis-Zhang-Onehouse, 2026-05-22, lock-provider): Follow-up to fix(common): normalize hudi table base path in implicit-key lock providers #18814 for StorageBasedLockProvider; mergeable, CI green, recent (2026-06-25)
• #18821 feat(flink): support singleton RocksDBDAO creation in RocksDBIndexBackend (HuangZhenQiu, 2026-05-22, flink): Recent Flink RocksDBDAO sharing feature, mergeable, active review discussion with cshuo
• #18826 feat(metadata): optimize RLI bootstrap by sizing file groups from base file footer row counts (nsivabalan, 2026-05-23, metadata-table): Approved by codope, in land queue (voonhous asked to resolve conflicts 2026-07-15); conflicting, needs rebase
• #18837 feat(trino): Migrate the Trino-Hudi connector into the Hudi repo (RFC-105) (voonhous, 2026-05-25, trino): Highly active RFC-105 Trino connector migration, 61 reviews, mergeable, updated 2026-07-15
• #18843 fix(append): size buffered record, not incoming, in block-flush gate (nsivabalan, 2026-05-26, log-append): Recent write-path correctness fix by committer, in review flow; needs rebase (CONFLICTING) but active
• #18849 Skip source RDD persist when index is RECORD_INDEX (vinishjail97, 2026-05-26, spark-write): Recent perf change, mergeable, awaiting review; not superseded
• #18860 feat(append): dynamically cap log-block buffer at per-task memory (nsivabalan, 2026-05-27, log-append): Recent memory-cap feature by committer; needs rebase (CONFLICTING) but in normal review flow
• #18885 fix: strip Hudi meta fields from write schema (nsivabalan, 2026-05-31, core-write): Recent schema-handling fix, mergeable, awaiting review
• #18887 fix: do not fall back to timeline server markers on transient HDFS failures (vamshipasunuru1, 2026-05-31, markers): Recent correctness fix for marker fallback, mergeable, active (updated Jul 7)
• #18889 [HUDI-18888] fix(streamer): downgrade source-emitted V2 checkpoint to V1 for v6 writes (nsivabalan, 2026-06-01, streamer): Recent checkpoint-compat fix by committer, updated Jul 13; needs rebase (CONFLICTING) but active
• #18893 feat(utilities): Migrate schema providers to HoodieSchema (Pavan-249, 2026-06-01, utilities-schema): In-flight Avro->HoodieSchema migration work; needs rebase (CONFLICTING) but active
• #18901 feat: support Lance metadata stats (puneetdixit200, 2026-06-02, lance): Recent Lance metadata-stats feature, updated Jul 12; needs rebase (CONFLICTING) but active
• #18910 chore(deps): bump org.apache.thrift:libthrift from 0.14.0 to 0.23.0 in /packaging/hudi-integ-test-bundle (dependabot[bot], 2026-06-03, deps): Dependabot libthrift bump, CI SUCCESS, mergeable, no newer superseding bump
• #18914 fix(utilities): Re-inject _corrupt_record in ErrorTableAwareChainedTransformer after each transformer (raghusrhyme, 2026-06-04, utilities): Recent error-table transformer fix, mergeable, updated Jul 10
• #18923 refactor(variant): self-align log-block variant rows, drop buffer-level projection hook (voonhous, 2026-06-05, variant): Active variant refactor, updated Jul 15; needs rebase (CONFLICTING) but in review flow
• #18941 fix(record-index-bootstrap): sort record index keys by UTF-8 bytes to match HFile sorting (nada-attia, 2026-06-08, record-index): Recent RI bootstrap sort-order correctness fix, mergeable, updated Jul 12
• #18942 fix(core): use correct clustering action for 0.x vs 1.x in BaseHoodieTableServiceClient (rahil-c, 2026-06-08, core): Recent clustering-action fix, not yet landed on master; needs rebase (CONFLICTING) but valid
• #18949 fix(streamer): Include start commit in S3/GCS IncrSource incremental query (yihua, 2026-06-10, streamer): Recent data-loss fix by committer; draft but real in-progress work, mergeable
• #18954 docs(datasource): START_COMMIT is exclusive, not inclusive (yihua, 2026-06-10, docs): Recent doc-accuracy fix by committer; draft but valid, not yet landed on master
• #18955 docs(configs): START_COMMIT is exclusive, not inclusive (yihua, 2026-06-10, asf-site, docs): asf-site companion to docs(datasource): START_COMMIT is exclusive, not inclusive #18954, corrects current config behavior, still wanted; recent committer draft
• #18961 feat(variant): auto-infer per-file variant shredding schemas on shredding inference (voonhous, 2026-06-10, variant): Recent variant-shredding feature, updated Jun 25; needs rebase (CONFLICTING) but active
• #18975 [HUDI-18139] fix(spark): make post-sync Spark catalog cache refresh best-effort (ad1happy2go, 2026-06-11, spark-sql): Recent catalog-refresh fix by committer, mergeable, in review flow
• #18983 feat(hive-sync): batch and parallelize HMS partition operations (nsivabalan, 2026-06-11, hive-sync): Base of stacked hive-sync series (feat(hive-sync): batch and parallelize HMS partition operations #18983<-feat(hive-sync): batch and parallelize HiveQL partition operations #18984<-feat(hive-sync): batch and parallelize JDBC partition operations #18986), CI SUCCESS; draft but active committer work
• #18984 feat(hive-sync): batch and parallelize HiveQL partition operations (nsivabalan, 2026-06-12, hive-sync): Middle of stacked hive-sync series, mergeable, updated Jul 14; active
• #18986 feat(hive-sync): batch and parallelize JDBC partition operations (nsivabalan, 2026-06-12, hive-sync): Top of stacked hive-sync series, updated Jul 14; draft pending feat(hive-sync): batch and parallelize HMS partition operations #18983/feat(hive-sync): batch and parallelize HiveQL partition operations #18984 merge but active
• #18988 feat(client): enrich write commit callback message and fire it for table-service commits (codope, 2026-06-12, core-client): Recent callback feature by committer, updated Jul 4; needs rebase (CONFLICTING) but active
• #18990 perf(spark): restore Catalyst fast-paths for SimpleKeyGen/Nonpartitioned bulk insert (nsivabalan, 2026-06-12, spark): Recent bulk-insert perf fix by committer, mergeable, in review flow
• #19005 fix(spark): runtime-merge full file groups for MOR incremental queries (Partial updates do not support incremental query #18943) (ad1happy2go, 2026-06-15, spark): Actively reviewed (CHANGES_REQUESTED), updated today; mergeable MOR incremental fix
• #19018 perf(kafka-connect): avoid per-record TopicPartition allocation in HoodieSinkTask.put (wombatu-kun, 2026-06-16, kafka-connect): Recent kafka-connect perf micro-opt, mergeable, awaiting review
• #19021 perf(io): Avoid per-entry KeyValue allocation in HFileDataBlock.seekTo (wombatu-kun, 2026-06-16, io-hfile): recent perf fix in native HFile seekTo hot path; mergeable, 7 reviewers (changes-requested), normal review flow
• #19023 fix: initialize lastCompletedTxnAndMetadata before recommitInstant in Flink StreamWriteOperatorCoordinator (suryaprasanna, 2026-06-16, flink): recent Flink OCC recommit bug fix; heavy active review (22), conflicting but 2026-06 in normal flow; rebase and continue
• #19029 fix(schema): permit opt-in timestamp-precision evolution (yihua, 2026-06-17, schema-evolution): recent opt-in timestamp-precision schema evolution by committer; active review, conflicting but current; keep for rebase
• #19033 feat(hive-sync): parallelize DROP partitions in HiveQL sync mode (nsivabalan, 2026-06-17, hive-sync): recent hive-sync DROP parallelization by committer; builds on feat(hive-sync): batch and parallelize HiveQL partition operations #18984, closes [IMPROVEMENT] Hive Sync partition operations lack batching and parallelism, causing 4x-9x slowdown for large tables #18331, mergeable, active
• #19041 feat(clean): support clean-by-time retention boundary and archive protection (fhan688, 2026-06-19, clean): recent clean-by-completion-time fix; updated 2026-07-14, 13 reviewers, mergeable; active review flow
• #19042 [MINOR] Make HoodieTable AutoCloseable and close per-cycle tables in BaseHoodieTableServiceClient (Davis-Zhang-Onehouse, 2026-06-19, writer): recent HoodieTable AutoCloseable resource-leak fix; conflicting but 2026-06 and in review; rebase and continue
• #19045 feat(metadata): introduce HoodieMetadataTableLayout SPI with sub-directory bucketing for global RLI (nsivabalan, 2026-06-22, metadata-table): recent MDT layout SPI feature by committer; 27 reviewers heavy engagement, closes [FEATURE] Support Sub-directory Bucketing for Metadata Table Partitions to Overcome File Count Limits #18178; keep for review/rebase
• #19046 docs: rfc-107 Dynamic Partitioned Cache for Flink Hudi Upsert (HuangZhenQiu, 2026-06-22, rfc): recent RFC-107 Flink dynamic partitioned cache doc; active RFC review (7), updated 2026-07-11
• #19047 feat(writer): add COMMIT_TIME_ONLY meta-fields mode for incremental queries on minimal-meta-fields tables (nsivabalan, 2026-06-22, writer): recent COMMIT_TIME_ONLY meta-fields mode by committer; conflicting but updated 2026-07-12, addresses Support selective meta field population #18383; active
• #19056 fix(spark): reject BLOB columns as record key / ordering / partition path fields (ZZZxDong, 2026-06-24, spark-sql): recent fix rejecting BLOB key/order/partition fields; closes BLOB column should be rejected as primaryKey / recordKey #18819, conflicting but 2026-07 and in review; rebase
• #19071 fix(storage-format): emit HBase-readable block-index keys in the native HFile writer (yihua, 2026-06-25, io-hfile): recent native HFile writer HBase-compat fix by committer; mergeable, active review; base of a live stack
• #19077 perf(fs-view): skip bootstrap index init for non-bootstrap tables to cut S3 HEAD calls (wangxianghu, 2026-06-26, fs-view): recent fs-view S3 HEAD-call perf fix; mergeable, 10 reviewers, active review flow
• #19080 [MINOR] Add MDT-partition-shape parameterized HBase-read compat test (shangxinli, 2026-06-26, io-hfile): recent draft MDT-shape HBase-read compat test stacked on fix(storage-format): emit HBase-readable block-index keys in the native HFile writer #19071/[DNM, stacked] fix(storage-format): fix multi-block previous-block offset and add byte-level HFile writer tests #19083; valuable coverage, land after base PRs
• #19081 feat(partition-ttl): Introduce KeepByEventTimeStrategy to expire partitions by event time (wangxianghu, 2026-06-27, partition-ttl): recent draft event-time partition TTL strategy; 18 reviewers heavy engagement, updated 2026-07-03; finish review
• #19082 test: Spark core tests (yihua, 2026-06-27, ci): recent draft CI core-test-subset optimization by committer; cuts ~2100 runner-min/push; active, keep
• #19083 [DNM, stacked] fix(storage-format): fix multi-block previous-block offset and add byte-level HFile writer tests (yihua, 2026-06-27, io-hfile): DNM only means stacked on fix(storage-format): emit HBase-readable block-index keys in the native HFile writer #19071; real recent prev-block-offset fix + writer tests by committer; keep to land in stack
• #19085 fix: reduce MDT compaction heap pressure in BitCaskDiskMap and ExternalSpillableMap (nsivabalan, 2026-06-28, metadata-table): recent MDT compaction heap-pressure OOM fix by committer; mergeable, closes [SUPPORT] Metadata-table compaction OOM on executors: unbounded heap from BitCaskDiskMap.valueMetadataMap, LazyFileIterator sort, and ExternalSpillableMap.keySet() #19084; active
• #19107 [HUDI-19095] Add RFC-104 - Vector Search Index in MDT (chrevanthreddy, 2026-06-29, rfc): recent RFC-104 vector-search-index-in-MDT doc; mergeable, active RFC review under umbrella RFC-109: MDT-backed vector index umbrella tracking issue #19094; keep
• #19108 feat(metrics): emit hoodie.table.version as gauge on client init (shangxinli, 2026-06-29, metrics): recent draft table-version gauge metric; small observability feature, updated 2026-06-30; keep
• #19110 feat(utilities): add Postgres and Mysql Debezium CDC transformers (rahil-c, 2026-06-29, utilities): recent Debezium CDC transformers by committer; updated 2026-07-16 today, 11 reviewers, mergeable; very active
• #19115 fix(streamer): write non-partitioned sample-writes table for record-size estimation (yihua, 2026-06-30, streamer): recent sample-writes record-size-estimation fix by committer; already APPROVED, mergeable; keep to merge
• #19117 feat(flink): Add pendingCommitInstantCount metric for write coordinator (cshuo, 2026-06-30, flink): recent Flink pendingCommitInstantCount metric; closes Add coordinator metric for pending commit instants in Flink streaming write #19116, conflicting but 2026-06 and in review; rebase
• #19120 test: stabilize flaky tests in TestJavaHoodieBackedMetadata and TestMultipleHoodieJavaWriteClient (lokeshj1703, 2026-06-30, tests): recent flaky-test stabilization by committer; updated 2026-07-13, conflicting but active; rebase and continue
• #19123 fix: support reading tables partitioned on a nested column (vinishjail97, 2026-06-30, spark-reader): recent nested-partition-column read fix by committer; mergeable, closes Reading a table partitioned on a nested column fails with "Illegal character in: <field>" #19122, 7 reviewers; active
• #19126 fix: Handle map/array-nested leaf columns in column stats collection during MOR log-append (vinishjail97, 2026-07-01, metadata-table): recent col-stats map/array nested-leaf fix by committer; closes Column stats on MOR table version 9 crashes on map/array-nested leaf columns during log-append #19125, conflicting (AvroRecordContext moved pkg); rebase
• #19128 feat(config): express small-file threshold as a fraction of max file size (nsivabalan, 2026-07-01, config): Recent (Jul 2026), mergeable, active review flow with changes-requested feedback being iterated on
• #19132 refactor(spark): prune the forked extended SQL parser to Hudi-only statements (yihua, 2026-07-02, spark-sql): Recent parser cleanup (drops ~13k lines of forked grammar), mergeable, under review by author yihua
• #19138 chore(ci): modernize Codecov config for the coverage initiative [DNM] (yihua, 2026-07-02, ci): DNM is procedural (temp CI-trigger comment to be reverted); real codecov.yml modernization intended to merge, active
• #19161 test(spark): add stored procedure coverage (yihua, 2026-07-03, spark-tests): Recent test-coverage add for stored procedures, mergeable, actively reviewed (6 review rounds)
• #19162 test(spark): add catalog and V2 table DDL coverage (yihua, 2026-07-03, spark-tests): Recent catalog/V2 DDL test coverage, mergeable, actively reviewed; distinct scope from sibling coverage PRs
• #19163 test(spark): add write and commit path coverage (yihua, 2026-07-03, spark-tests): Recent write/commit-path test coverage, mergeable, actively reviewed (6 rounds)
• #19164 test(spark): cover low-coverage small classes in spark datasource (yihua, 2026-07-03, spark-tests): Recent datasource small-class test coverage, mergeable, heavily reviewed (12 rounds), updated Jul 13
• #19165 test(spark): add file-group reader read path coverage (yihua, 2026-07-03, spark-tests): Recent file-group reader read-path test coverage, mergeable, under active review
• #19169 fix(spark): correct data-skipping translation for LIKE 'prefix%' (StartsWith) (yihua, 2026-07-03, spark): Recent real bug fix (closes [SUPPORT] Data-skipping silently drops rows for LIKE 'prefix%' (StartsWith) — regression in 1.1.x #18791), small mergeable diff for StartsWith data-skipping translation
• #19173 chore: gate pull requests on 80% patch coverage (yihua, 2026-07-03, ci): Recent CI policy proposal (patch-coverage gate workflow) tied to active coverage initiative, mergeable
• #19202 fix(flink): remove the Source V2 read teardown race via materialized bounded minibatches (wombatu-kun, 2026-07-06, flink): Very active flink race fix, already APPROVED, updated Jul 16 (today), mergeable
• #19204 test: add unit test coverage for secondary index, expression, and bloom filter classes in hudi-common (rahil-c, 2026-07-06, common-tests): Recent hudi-common test coverage; conflicting so needs rebase but scope current and under review
• #19205 feat(writer): add hoodie.meta.fields.mode for selective meta-field population on CoW tables (nsivabalan, 2026-07-06, writer): Recent feature (closes Support selective meta field population #18383), mergeable draft under active review by author nsivabalan
• #19206 feat(cli): add table set-meta-fields-mode command (nsivabalan, 2026-07-06, cli): Recent CLI companion to feat(writer): add hoodie.meta.fields.mode for selective meta-field population on CoW tables #19205 meta-fields-mode feature, mergeable, actively developed
• #19211 [Draft WIP] feat(flink): Support reading and materializing OOL BLOB columns (kbuci, 2026-07-07, flink): Active WIP for RFC-100 unstructured-data feature (closes [Flink][Blob phase 1] Allow FlinkSQL reads to materialize OOL blobs #19032), recent and clearly valuable scope
• #19214 refactor(common)!: relocate table, reader, util and transaction domains to org.apache.hudi.core (vinothchandar, 2026-07-07, refactor): Large in-flight package relocation by project lead vinothchandar, updated Jul 14; conflicting but active
• #19217 test(trino): add a Trino E2E testcontainers pipeline for the RFC-105 connector (voonhous, 2026-07-07, trino): Recent Trino E2E testcontainers pipeline, mergeable draft actively developed, updated Jul 15
• #19219 test(spark): add unit coverage for Spark row, schema and sort utilities (yihua, 2026-07-07, spark-tests): Recent Spark row/schema/sort util test coverage, mergeable draft in active coverage initiative
• #19220 test(common): add unit coverage for config and utility classes (yihua, 2026-07-07, common-tests): Recent config/util test coverage, mergeable draft in active coverage initiative
• #19221 test(common): add unit coverage for metrics reporters and schema utilities (yihua, 2026-07-07, common-tests): Recent metrics-reporter/schema-util test coverage, mergeable draft in active coverage initiative
• #19222 test(client): add unit coverage for client utilities and services (yihua, 2026-07-07, client-tests): Recent client util/service test coverage, mergeable draft in active coverage initiative
• #19223 test(common): add log-format reader and scanner coverage (yihua, 2026-07-07, common-tests): Recent log-format reader/scanner test coverage, mergeable draft in active coverage initiative
• #19226 fix(common): Fix MOR compaction rollback planning and cleanup (fhan688, 2026-07-09, compaction): Active MOR compaction rollback fix, mergeable, heavily reviewed (14 rounds), updated Jul 15
• #19227 feat(utilities): add HoodieShadowPipeline for creating and syncing shadow tables (suryaprasanna, 2026-07-09, utilities): Recent new feature (shadow-table pipeline), mergeable, awaiting first review
• #19228 [DNM][1.2] Add versioned binary list state for flink writer (cshuo, 2026-07-09, release-1.2.0, flink): Active change on live release-1.2.0 branch, CI passing, updated Jul 13; DNM appears procedural (release sequencing)
• #19230 feat(utilities): revamp TableSizeStats with detectors, skew metrics, JSON output; extract SmallFileDetector (nsivabalan, 2026-07-09, utilities): Recent (Jul-09) TableSizeStats revamp feature by committer nsivabalan; mergeable, in normal review flow.
• #19239 fix(meta-sync): advance last commit time synced when it trails the active timeline midpoint (yihua, 2026-07-09, meta-sync): Recent (Jul-09) draft meta-sync last-commit-time fix by yihua; small, active investigation, mergeable.
• #19243 fix(metadata-table): add config to skip zero-size data files on MDT initialization (lokeshj1703, 2026-07-10, metadata-table): Recent (Jul-10) MDT zero-size-file skip fix by lokeshj1703; 2 review rounds, needs rebase (conflicting).
• #19247 fix(DNM): adjust the native log format metadata as separate entries in footer (danny0405, 2026-07-10, core): Recent (Jul-10) native-log footer metadata fix by danny0405 under active review (5 rounds, updated Jul-15); DNM is merge-gate, keep for review.
• #19248 fix(meta-sync): Normalize meta sync base path to avoid consecutive/trailing slashes in table LOCATION (ad1happy2go, 2026-07-10, branch-0.x, meta-sync): Recent (Jul-10) meta-sync path-normalization fix, 7 active review rounds; targets branch-0.x maintenance, bug likely on master too so a master port may be needed.
• #19252 fix(core): cover missed VORTEX dispatch in reader factory and log format selection (yihua, 2026-07-11, core): Recent (Jul-11) VORTEX dispatch fix by yihua; 2 review rounds, needs rebase (conflicting).
• #19253 test(core): add file format dispatch exhaustiveness coverage (yihua, 2026-07-12, core): Recent (Jul-12) draft file-format dispatch coverage tests by yihua; part of active VORTEX/file-format work stack.
• #19254 refactor(io): extract shared Arrow writer base for Lance and Vortex (yihua, 2026-07-12, io): Recent (Jul-12) draft Arrow-writer base refactor by yihua; part of active Lance/Vortex io stack.
• #19255 refactor(core): consolidate per-format dispatch in file reader and writer factories (yihua, 2026-07-12, core): Recent (Jul-12) draft factory dispatch consolidation by yihua; part of active file-format stack.
• #19265 feat: agentic lakehouse v1 - hudi-lakehouse charts and hudi-agent-gateway (vinothchandar, 2026-07-12, agent-gateway): Recent (Jul-12) agentic lakehouse v1 feature by vinothchandar; 12 review rounds, author says will incorporate comments after wider discussion.
• #19270 docs(blog): explainer wave 2: streaming data lake, CDC, lakehouse comparisons (vinothchandar, 2026-07-13, asf-site, docs): Recent (Jul-13) draft blog explainer wave by vinothchandar; new 2026 content, part of stacked docs series.
• #19271 docs(blog): XTable interop post and Parquet/Hive migration guides (vinothchandar, 2026-07-13, asf-site, docs): Recent (Jul-13) draft XTable/migration blog guides by vinothchandar; new 2026 content, stacked docs series.
• #19272 docs(blog): Delta/Iceberg migration guides and CDC comparison (vinothchandar, 2026-07-13, asf-site, docs): Recent (Jul-13) draft Delta/Iceberg migration + CDC blog guides by vinothchandar; new 2026 content, stacked series.
• #19273 docs(blog): streaming, write-heavy, read-heavy and benchmark comparisons (vinothchandar, 2026-07-13, asf-site, docs): Recent (Jul-13) draft streaming/benchmark comparison blogs by vinothchandar; new 2026 content, stacked series.
• #19275 fix(ci): let validate-source's RAT check fail on unlicensed files outside Maven modules (wombatu-kun, 2026-07-13, ci): Recent (Jul-13) CI RAT-check fix by wombatu-kun; 4 review rounds, updated Jul-16, author requesting review.
• #19278 feat(ingest): ingest unstructured files into BLOB+VECTOR Hudi tables via Hudi Streamer (vinothchandar, 2026-07-13, utilities): Recent (Jul-13) draft unstructured-file ingest feature by vinothchandar; part of active RFC-100 unstructured-data work.
• #19280 fix(debezium): keep Postgres toasted columns on MOR read (v6/v8) (linliu-code, 2026-07-13, debezium): Recent (Jul-13) Debezium Postgres toasted-column MOR fix by linliu-code; 5 review rounds, mergeable.
• #19282 docs: add v10 into the tech-spec page (danny0405, 2026-07-14, asf-site, docs): Recent (Jul-14) tech-spec doc adding v10 table version by danny0405; current-content update, cc vinothchandar for review.
• #19284 refactor(common): utils consolidation sweep remnants (vinothchandar, 2026-07-14, common): Recent (Jul-14) draft common-utils consolidation refactor by vinothchandar; mergeable, recently opened.
• #19288 feat(trino): resolve merge-required columns from the table schema (voonhous, 2026-07-14, trino): Recent (Jul-14) draft Trino merge-column resolution feature by voonhous; active author work, mergeable.
• #19289 fix(hive-sync): sync column and partition column comments to HMS (rangareddy, 2026-07-14, hive-sync): Recent (Jul-14) hive-sync column-comment sync fix by rangareddy; 3 review rounds, updated Jul-16, active.
• #19291 docs(rfc): RFC-104 native vector search support in Apache Hudi (chrevanthreddy, 2026-07-14, docs): Recent (Jul-14) RFC-104 native vector search proposal by chrevanthreddy; active RFC under review.
• #19293 fix: fix the file size calibration for native log format (danny0405, 2026-07-15, core): Recent (Jul-15) native-log file-size calibration fix by danny0405; 17 review rounds, updated Jul-16, very active.
• #19295 test(trino): add MoR read tests for delete markers, custom payloads and commit-time ordering (voonhous, 2026-07-15, trino): Recent (Jul-15) draft Trino MOR read tests by voonhous; active author test work, stacked on trino infra.
• #19298 fix(trino): read uncompacted MDT HFILE log deltas and guard index pruning (voonhous, 2026-07-15, trino): 1-day-old draft in active stacked Trino chain (test(trino): add MoR read tests for delete markers, custom payloads and commit-time ordering #19295>feat(trino): resolve merge-required columns from the table schema #19288>feat(trino): Migrate the Trino-Hudi connector into the Hudi repo (RFC-105) #18837), closes [BUG] trino-hudi: queries fail on tables with uncompacted MDT deltas (HFILE log blocks unsupported, pruning path unguarded) #19279, self-authored, tests included
• #19299 docs: RFC-108 - Multi-dataset incremental reads in Hudi Streamer (ashokkumar-allu, 2026-07-15, docs): fresh RFC-108 doc (closes Multi-dataset incremental reads in Hudi Streamer #19281), status UNDER REVIEW, additive rfc/ files, in normal review flow
• #19300 feat(metadata): MDT posting-block schema foundation for vector index (chrevanthreddy, 2026-07-15, metadata-table): fresh additive RFC-104 vector-index MDT foundation (closes Add MDT row families and payload schema for vector indexes #19097), hudi-common only, unit-tested
• #19304 [HUDI-9666] Auto-deduce ComplexKeyGenerator single-field record-key encoding (table version 8 and below) (ad1happy2go, 2026-07-16, keygen): 1-day-old committer fix for HUDI-7001 keygen encoding mix, tv8-scoped, three new functional suites, e2e-verified
• #19305 [HUDI-9666] Auto-deduce ComplexKeyGenerator single-field record-key encoding on branch-0.x (ad1happy2go, 2026-07-16, branch-0.x, keygen): deliberate committer backport of [HUDI-9666] Auto-deduce ComplexKeyGenerator single-field record-key encoding (table version 8 and below) #19304 to live maintenance branch (last merged Apr 2026); bug is 0.14.x/0.15.0-specific, backport is the correct target