postgres_pgmq 0.1.0

postgres_pgmq: ^0.1.0 copied to clipboard

Breaking changes

• Renamed setVt/setVtBatch to setVisibilityTimeout/ setVisibilityTimeoutBatch.

• Renamed the vt named parameter to visibilityTimeout on all read methods and on watch.

• Renamed PgmqMessage.vt to PgmqMessage.visibleAt.

• Added Pgmq.queue and Pgmq.queueOf: queue-scoped PgmqQueue<T> handles that bind the queue name and, for typed handles, payload encoding and decoding once.

• Added Pgmq.watch and PgmqQueue.watch: a cancellable, pause-aware long-poll Stream<PgmqMessage<T>> for consuming messages.

• Added acquireQueueLock: transaction-scoped advisory queue lock (pgmq.acquire_queue_lock), matching the official Rust client and pgmq-go. Serializes queue-level DDL such as FIFO index creation.
• createPartitionedQueue is now idempotent by default (ifNotExists: true): it takes the queue lock and checks pgmq.meta inside runTx before creating, like the official Rust client. Pass ifNotExists: false for the previous direct call.
• Added queueMetadata / queueExists: single-queue metadata lookups from pgmq.meta.
• Added extensionVersion: the installed PGMQ extension version.
• Added listenNotifyInsert: broadcast Stream<String> of insert notifications for Connection sessions (Pool/TxSession throw).
• QueueMetrics.defaultPartitionLength parses the default_partition_length column added in PGMQ v1.13.0 (null on older servers).
• Targets PGMQ v1.13.0 (still compatible with v1.12.0).

• Initial release.
• Full PGMQ SQL API (v1.12.0) over any package:postgres Session (Connection, Pool, TxSession): extension setup, queue lifecycle (including unlogged and partitioned queues), send / send batch with headers and delays, read / long-poll / pop, grouped (FIFO) reads, delete / archive (single + batch), visibility timeouts, metrics, topic routing and bindings, insert notifications, FIFO indexes.
• Strongly typed PgmqMessage<T> with toJson / fromJson / mapPayload support; Map<String, dynamic> payloads work out of the box.
• Parameterized queries only — no SQL string interpolation of user input.
• Unit tests, Docker-based integration tests, and examples for Connection, Pool, transactions, FIFO and topics.