Add task state store API to the Java SDK

[Andrushika]

Why

Airflow 3.3 added the task state store (AIP-103). A task can save a value under a key, and the value survives retries.

The supervisor already handles the four *TaskStateStore messages and the Java SDK schema.json already has them. But nothing on the Kotlin side sends them. So Java tasks cannot use the store and capabilities.yaml lists task-state-store as unsupported.

What

Add client.getTaskStateStore() with get, set, delete and clear. It is wired the same way as variables. get returns null when the key is not there. set takes an optional java.time.Duration retention, or TaskStateStore.NEVER_EXPIRE. Without a retention it uses [state_store] default_retention_days, read from AIRFLOW__STATE_STORE__DEFAULT_RETENTION_DAYS the same way as the Go SDK in #73420 (0 means never expire, a missing variable fails).

SetTaskStateStore is added to REQUIRED_NULLABLE_REQUESTS. Without it Jackson drops a null expires_at and the supervisor rejects the message. I checked this with the supervisor decoder.

One difference from Python is noted in java.rst: [workers] state_store_backend is not supported. The backend hooks run inside the Python task process today, so supporting it in the Java SDK needs a further design discussion.

related: #73420

---

Was generative AI tooling used to co-author this PR?

• Yes — Claude Code (Fable 5.1)

Generated-by: Claude Code (Fable 5.1) following the guidelines

[potiuk]

Hello @Andrushika - thank you for your contributions to Apache Airflow!

The Airflow community has introduced a limit of 5 open pull requests at a time for contributors without write access to the repository. You currently have 7 open pull requests, so - as a one-time step of introducing the limit - we closed the ones where maintainers have not engaged yet:

• Preserve AirflowTaskTimeout when timeout logging fails #69713 - Preserve AirflowTaskTimeout when timeout logging fails
• Fix mapped tasks does not validate task_id #70135 - Fix mapped tasks does not validate task_id
• Reject consecutive dots in Dag and task IDs in the Task SDK #70630 - Reject consecutive dots in Dag and task IDs in the Task SDK
• Enforce execution_timeout from the task supervisor #72493 - Enforce execution_timeout from the task supervisor
• Add breeze verify to list local checks for changed files #73241 - Add breeze verify to list local checks for changed files
• Add task state store API to the Java SDK #73464 - Add task state store API to the Java SDK

These pull requests stay open because maintainers are already engaged in them - they count towards your limit:

• Warn on suspicious Dag and task IDs at Java SDK build time #69937 - Warn on suspicious Dag and task IDs at Java SDK build time

This is not a judgement of you or of your changes. We never told contributors before that opening many pull requests at once was a problem, so there is nothing to feel bad about - and nothing is lost: your branches, commits and the review history stay where they are.

What we ask you to do is to make your first prioritization decision: choose which of the pull requests above matter most to you, and reopen them (up to 5 open at a time, including the ones still open) with the "Reopen pull request" button or gh pr reopen <PR_NUMBER> --repo apache/airflow. Reopen the ones you are ready to follow through - keep them rebased, respond to review comments and fix failing checks.

While your pull requests are waiting for review, the most valuable thing you can do is help in other ways - reviewing other contributors' pull requests, helping with issues, and taking part in the discussions on the devlist and Slack.

Why we introduced the limit, what it means for you and how to reopen or restore a pull request is explained in https://github.com/apache/airflow/blob/main/contributing-docs/32_open_pull_request_limit.rst.

---

Drafted-by: Claude Code (Opus 5); reviewed by @potiuk before posting

[FrankYang0529]

Thanks for the PR. Leave one comment.

[jason810496]

Nice, thanks! LGTM overall.

Let's also add a TODO before the serialization to respect the state_store/max_value_storage_bytes that will be passed from the coordinator side.

[Andrushika]

Also applied the two points from #73420 and #73618 that apply here:

• Removed the 30-day fallback. set without a retention now fails with a clear error when AIRFLOW__STATE_STORE__DEFAULT_RETENTION_DAYS is absent.
• Added a TODO before sending for the max_value_storage_bytes warning.

One note on ordering: the coordinator only passes that variable once #73420 is merged. If this PR goes in first, set without a retention will fail until then.

[jason810496]

Also applied the two points from #73420 and #73618 that apply here:

• Removed the 30-day fallback. set without a retention now fails with a clear error when AIRFLOW__STATE_STORE__DEFAULT_RETENTION_DAYS is absent.
• Added a TODO before sending for the max_value_storage_bytes warning.

One note on ordering: the coordinator only passes that variable once #73420 is merged. If this PR goes in first, set without a retention will fail until then.

Would it be better to add the explicit NEVER_EXPIRE sentinel instead of using 0 as the never expire?

[Andrushika]

Yes, we already have NEVER_EXPIRE for the Client API.

set without a retention now reads AIRFLOW__STATE_STORE__DEFAULT_RETENTION_DAYS, with the same rules as the Go SDK in #73420: absent means 30 days, 0 means never expire, and a malformed value fails. A zero or negative retention is rejected too

The 0 I described in the resolved comment is the config value (default_retention_days) passed from the coordinator, not a value users pass to set.

Sorry for the misleading wording :(

[jason810496]

Nice, thanks.