nsivabalan opened a new pull request, #19238:
URL: https://github.com/apache/hudi/pull/19238
### Change Logs
Adds `org.apache.hudi.tools.TimelineInspector`, a pure-Java CLI for
inspecting a Hudi table's active + archived timeline without needing Spark.
Lives in `hudi-hadoop-common` since it uses Hadoop `FileSystem` APIs for raw
archive log scanning and per-instant mtime lookups.
#### Modes
- `--show-instant` — dump one instant's content (all states + timeline
location)
- `--find-file-id` — find which instants touched a file id / partition /
file group
- `--commit-stats` — per-instant stats
(numWrites/numInserts/numUpdateWrites/numDeletes, bytes, partitions touched)
with optional filters
- `--parse-filename` — decode a base or log file name into {fileId,
commitTime, writeToken, ...}
- `--raw-archive` — dump raw archive log entries for a specific instant
- `--phase-timings` — per-instant phase durations
(requested→inflight→completed for data table + MDT sub-table) for latency
forensics
Modifiers: `--lifecycle`, `--output table|json`, `--state`, `--actions`,
`--start-instant`, `--end-instant`, `--limit`, `--sort asc|desc`, `--quiet`,
`--no-archived`.
#### Robustness
Handles archived-timeline storage quirks:
- Avro-container vs JSON-string sniff
- Union-tag fixup walker for malformed JSON
- `instantsRollback` checked alongside legacy `commitsRollback`
- JSON-tree aggregator fallback for `--commit-stats` when the POJO path
yields an empty shell
Handles V2 timeline layout:
- Reads instant files under `<base>/.hoodie/timeline/`
- Resolves completed-instant filenames with the `_<completionTime>` suffix
via the metaclient's `InstantFileNameGenerator`
- For `--phase-timings`, the MDT completed-file lookup globs the timeline
dir by `<requestedTime>` prefix + `.deltacommit` suffix (we don't know the
completion time a priori)
### Impact
- **New**: standalone CLI tool for timeline forensics. Useful for on-call
debugging (what did commit X do?) and post-incident analysis (which commits
touched file id Y?).
- **No breaking changes**: purely additive. No `@PublicAPIClass` /
`@PublicAPIMethod` touched. No storage format changes. No new `hoodie.*`
configs.
### Risk level
**low** — new tool with no changes to existing code paths.
### Documentation Update
- New file:
`hudi-hadoop-common/src/main/java/org/apache/hudi/tools/TimelineInspector.md` —
full runbook (classpath setup, per-mode usage + examples, troubleshooting).
- Class-level Javadoc on `TimelineInspector`.
### Contributor's checklist
- [x] Read through [contributor's
guide](https://hudi.apache.org/contribute/how-to-contribute)
- [x] Change Logs and Impact were stated clearly
- [x] Adequate tests were added if applicable
- [x] CI passed locally: `mvn -pl hudi-hadoop-common -o test
-Dtest=TestTimelineInspector` → **32/32 passing**
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
