TODO(#16601) Make this method private once PerformUnlessClosing doesn't need it any more

returns the current approximate timestamp

whenever we get an update, we do set the approximate timestamp first to the sequencer time and use the synchronizer time tracker to advance the approximate time to the effective time after a timestamp greater than the effective time was received from the sequencer.

Wait for a condition to become true according to the current snapshot approximation

Finds the transaction with maximum effective time that has been sequenced before or at sequencedTime and yields the sequenced / effective time of that transaction. Potentially waits for sequencedTime to be observed.

Returns an optional future which will complete when the sequenced timestamp has been observed

If the timestamp is already observed, returns None.

Waits until a snapshot is available

The snapshot returned by this method should be used for validating transaction and reassignment requests (Phase 2 - 7). Use the request timestamp as parameter for this method. Do not use a response or result timestamp, because all validation steps must use the same topology snapshot.

Supervised version of awaitSnapshot

Returns an optional future which will complete when the effective timestamp has been observed

If the timestamp is already observed, returns None.

Blocks until all earlier tasks have completed and then prevents further tasks from being run.

Overloaded recent snapshot returning derived type

The order in which the subscriber should be executed among all the subscriptions. Lower values are executed first.

Possibly future dated head snapshot

As we future date topology transactions, the head snapshot is our latest knowledge of the topology state, but as it can be still future dated, we need to be careful when actually using it: the state might not yet be active, as the topology transactions are future dated. Therefore, do not act towards the sequencer using this snapshot, but use the currentSnapshotApproximation instead.

Check whether we're closing. Susceptible to race conditions; unless you're using this as a flag to the retry lib or you really know what you're doing, prefer performUnlessClosing and friends.

Set this to true to get detailed information about all futures that did not complete during shutdown.

current number of changes waiting to become effective

This must be called whenever a topology transaction is committed. It may be called at additional timestamps with transactions being empty. Calls must have strictly increasing sequencedTimestamp and effectiveTimestamp. The effectiveTimestamp must be the one computed by com.digitalasset.canton.topology.processing.TopologyTimestampPlusEpsilonTracker for sequencedTimestamp.

During crash recovery previous calls of this method may be replayed. Therefore, implementations must be idempotent.

Blocks until all earlier tasks have completed and then prevents further tasks from being run.

The synchronizer this client applies to

TODO(#16601) Make this method private once PerformUnlessClosing doesn't need it any more

Schedules the given task to be run upon closing.

Register a task to run when closing is initiated, or run it immediately if closing is already ongoing. Unlike runOnClose, this method does not guarantee that this task will have run by the time the LifeCycleManager's closeAsync method completes or AutoCloseable's close returns. This is because the task is run immediately if the component has already been closed.

Variant of runOnOrAfterClose that does not return a com.digitalasset.canton.lifecycle.LifeCycleRegistrationHandle.

Overloaded snapshot returning derived type

Returns whether a snapshot for the given timestamp is available.

Runs the computation f only if the component is not yet closing. If so, the component will delay releasing its resources until f has completed (as defined by the com.digitalasset.canton.util.Thereafter instance) or the synchronizeWithClosingPatience has elapsed.

DO NOT CALL this.close as part of f, because it will result in a deadlock. DO NOT PUT retries, especially indefinite ones, inside f.

Convenience method for synchronizing on Futures instead of FutureUnlessShutdowns. Equivalent to

synchronizeWithClosing(name)(FutureUnlessShutdown.outcomeF(f))

Runs the computation f only if the component is not yet closing. If so, the component will delay releasing its resources until f has finished or the synchronizeWithClosingPatience has elapsed.

DO NOT CALL this.close as part of f, because it will result in a deadlock. DO NOT PUT retries, especially indefinite ones, inside f.

Runs the computation f only if the component is not yet closing. If so, the component will delay releasing its resources until f has completed (as defined by the com.digitalasset.canton.util.Thereafter instance) or the synchronizeWithClosingPatience has elapsed.

DO NOT CALL this.close as part of f, because it will result in a deadlock. DO NOT PUT retries, especially indefinite ones, inside f.

Returns the topology information at a certain point in time

Fails with an exception if the state is not yet known.

The snapshot returned by this method should be used for validating transaction and reassignment requests (Phase 2 - 7). Use the request timestamp as parameter for this method. Do not use a response or result timestamp, because all validation steps must use the same topology snapshot.

Runs the computation fa unless isClosing returns true.

This method does not delay the closing while fa is running, unlike the methods in HasSynchronizeWithClosing. Accordingly, this method is useful for intermittent checks whether the result of the computation is still relevant.

Move the highest known (effective / approximate) timestamp ahead in the future.

May only be called if:

1. All committed topology transactions with effective time up to effectiveTimestamp have been persisted in the topology store.
2. All committed topology transactions with sequenced time up to sequencedTimestamp have been persisted in the topology store.
3. If this method is called with potentialTopologyChange == true, then for every subsequent committed topology transaction either updateHead(potentialTopologyChange == true, ...) or observed must be called again; such calls must occur with ascending effective timestamps.
4. sequencedTimestamp <= effectiveTimestamp
5. approximateTimestamp <= effectiveTimestamp
6. A sequenced event with timestamp at least approximateTimestamp has been received from the sequencer.