[poissoncorp]

SubscriptionWorker now reports what it is doing through status and on_state_changed, and raises an event each time its connection to the server is established.

• Changed: new SubscriptionWorkerState and SubscriptionWorkerStatus (state, exception, since_utc, failing_since_utc), exported from ravendb. SubscriptionWorker gains status, add_/remove_on_state_changed and add_/remove_on_established_subscription_connection. The run loop sets Connecting, WaitingForDocuments, Processing, Retrying, Faulted and Stopped at the same points as the C# client.
• Not ported: Stopped on cancellation of the token passed to Run. Python run() takes no token, so stopping goes through close(), which reports Stopped.
• Tests: new test_ravenDB_27175.py with 5 server tests from RavenDB_27175 and 7 status transition tests that need no server. Skipped: the reflection checks for public API, the reconnection tests that need DropConnectionAsync, and the tests that use the Run cancellation token or SimulateUnexpectedException.
• Checks: black --check . passed; the 7 transition tests passed; the server tests run only in CI.

Decision needed

The worker waits for the next batch with read_from_server.result(time_to_wait_before_connection_retry), 5 seconds by default. On an idle subscription that wait can time out, and the worker treats the timeout as an error and reconnects. If so, an idle Python worker reports Retrying, Connecting and WaitingForDocuments on every cycle, the established connection event fires each time, and failing_since_utc is set briefly. The C# worker stays in WaitingForDocuments. This is read from the code and not confirmed against a server.

1. Merge as is and fix the idle read timeout in a separate change (recommended): the status reports what the worker does today, and changing reconnect behavior is a separate behavior change.
2. Fix the batch wait in this PR before merging.
3. Do not merge.

Timestamps are naive UTC datetimes, like the worker's existing _last_connection_failure. Timezone-aware UTC is the alternative.

---

Ticket: RDBC-1143
C# change: ravendb/ravendb#23271

SDK Sync Agent

[poissoncorp]

The failing check is not caused by this PR. No code change.

• Two tests fail, both in test_secured_subscription_basic: test_should_send_all_new_and_modified_docs and test_should_stream_all_documents_after_subscription_creation. Both time out waiting for the first batch (queue.Empty).
• The same two tests failed at the same time in five open PRs with unrelated changes (RDBC-1148 Merge release/quill into ravendb/v7.2 #313, RDBC-1141 RavenDB-25401 - Fix refusal on streaming #314, RDBC-1140 RavenDB-27181 : GPT-5.6 models unusable with AI agents - tools rejected on /v1/chat/completions #315, RDBC-1143 RavenDB-27175 Subscriptions - make it easier to inspect the current state #317, RDBC-1151 RavenDB-24824 Add per-call output schema override and string output for AI conversations #318). Each time they failed on only one or two Python versions and passed on the others with the same code.
• RDBC-1146 RavenDB-27173 Advanced.RegisterForConcurrencyCheck #312 passed on all Python versions earlier the same day.

A rerun of the failed job should pass. If it fails again, the secured test server setup in CI needs a look.

SDK Sync Agent

[poissoncorp]

The failing build job is not caused by this PR.

• Failed tests: test_should_send_all_new_and_modified_docs and test_should_stream_all_documents_after_subscription_creation in test_secured_subscription_basic, with queue.Empty. All other tests passed.
• Cause: the tests waited 4 seconds for a batch, less than the 5 second reconnect delay of the subscription worker.
• Fix: RDBC-1151 RavenDB-24824 Add per-call output schema override and string output for AI conversations #318 raises the wait to 15 seconds. This branch needs an update from v7.2 after RDBC-1151 RavenDB-24824 Add per-call output schema override and string output for AI conversations #318 is merged.

SDK Sync Agent

[poissoncorp]

The failing check is not caused by this PR. No code change.

• The failed job fails only on test_should_send_all_new_and_modified_docs and test_should_stream_all_documents_after_subscription_creation in test_secured_subscription_basic. Both time out waiting for the first batch (queue.Empty).
• The same two tests fail on one or two Python versions in RDBC-1148 Merge release/quill into ravendb/v7.2 #313, RDBC-1141 RavenDB-25401 - Fix refusal on streaming #314, RDBC-1140 RavenDB-27181 : GPT-5.6 models unusable with AI agents - tools rejected on /v1/chat/completions #315, RDBC-1156 RavenDB-22083 Report a proper IndexCompilationException for a C# index whose outer from clause does not enumerate the documents source #316, RDBC-1143 RavenDB-27175 Subscriptions - make it easier to inspect the current state #317 and RDBC-1151 RavenDB-24824 Add per-call output schema override and string output for AI conversations #318 and pass on the others with the same code.
• Cause and fix are in RDBC-1151 RavenDB-24824 Add per-call output schema override and string output for AI conversations #318: the tests wait 4 seconds for a batch, and the worker waits 5 seconds before it reconnects.
• This PR gets the fix when it is updated from v7.2 after RDBC-1151 RavenDB-24824 Add per-call output schema override and string output for AI conversations #318 is merged.

SDK Sync Agent

[poissoncorp]

Merged v7.2 into this branch (561525c) to pick up the test fix from #318. No change to the code of this PR.

• test_secured_subscription_basic: the tests now wait 15 seconds for a batch instead of 4, longer than the 5-second reconnect delay of the worker.
• This PR does not change the reconnect delay, so the fix holds with the worker changes here.
• black --check passes locally. The tests need a server and run only in CI.

SDK Sync Agent

[poissoncorp]

Merge as is and fix the idle read timeout in a separate change (recommended): the status reports what the worker does today, and changing reconnect behavior is a separate behavior change.