getkyo/kyo

Problem Statement Currently kyo-sql-postgres does not support kerberos authentication for postgres Proposed Solution Add ability to configure and use kerberos authentication for postgres. This feature is important to us on the JVM platform. Alternative Solutions No response Current Workaround No response Additional Context currently using doobie with postgres jdbc driver. example configuration: //krb5.conf example [libdefaults] default_realm = YOURDOMAIN.COM dns_lookup_realm = false dns_lookup_kdc = false [realms] YOURDOMAIN.COM = { kdc = ://yourdomain.com admin_server = ://yourdomain.com } //etc //jaas.conf example pgjdbc { com.sun.security.auth.module.Krb5LoginModule required useKeyTab=true keyTab="/path/to/your.keytab" principal="[email protected]" storeKey=true useTicketCache=false; }; pass jvm flags -Djava.security.auth.login.config=/path/to/jaas.conf -Djava.security.krb5.conf=/path/to/krb5.conf specify jaasApplicat

Context Dict and OrderedDict are designed to avoid (K, V) tuple allocations: iteration, lookup, and transformation pass separate key and value arguments rather than tuples, to eliminate boxing in hot paths. Two APIs still break that premise by forcing tuple allocation: API Dict OrderedDict apply[K, V](entries: (K, V)*) kyo-data/shared/src/main/scala/kyo/Dict.scala:83 kyo-data/shared/src/main/scala/kyo/OrderedDict.scala:88 map[K2, V2](fn: (K, V) => (K2, V2)) kyo-data/shared/src/main/scala/kyo/Dict.scala:416 kyo-data/shared/src/main/scala/kyo/OrderedDict.scala:423 The vararg constructor allocates a tuple per entry at the call site; map allocates a tuple per element for the result of fn. Origin Raised by @fwbrasil while reviewing #1749 (OrderedDict). OrderedDict deliberately mirrors Dict's current design, so the tuple-allocating shape is pre-existing on Dict and inherited by OrderedDict. Split out as a separate follow-up per that review: #1749 (comment) (constructor) #1749 (c

Summary Add an opt-in, user-provided dead-letter sink to Actor.run so an actor's unprocessed backlog has a typed destination when the actor stops, instead of only being returned synchronously from close(). Follow-up carved out of the actor pub/sub work (#1689), where it was deliberately deferred to keep that PR focused. Motivation Actor.close() already returns the queued-but-unprocessed messages (Maybe[Seq[A]] < Sync), a synchronous single-shot dead-letter handoff to the closer. That covers graceful close but not: termination via unhandled E or panic (the backlog is dropped, not handed anywhere), a component that wants undelivered messages routed to a sink (a monitor, a retry queue, an audit log) rather than returned to whoever happened to call close. A configurable deadLetters sink generalizes the existing return-the-backlog behavior to any termination path and any consumer. Why this is thinner than Akka's dead letters Akka dead-letters broadly because tell is fire-and-forget with n

Problem Statement When curious about how to deal with a Seq[A < Sync], one is currently lost, because there's no foreach or similar for Sync. It turns out that method exists in Kyo, which a beginner won't even think to look into. Proposed Solution Export relevant methods like these to Sync and other relevant effects to simplify their discoverability. Alternative Solutions Mention the capabilities of the Kyo object specifically in the documentation. Current Workaround No response Additional Context No response