)]}'
{
  "log": [
    {
      "commit": "da023849414fbfdb5734be56a6f6160bd8f8089b",
      "tree": "612ffef21aa4122f2c082d042bd0b7a45e099427",
      "parents": [
        "2db21d8cfe76ed8d380cccd170eca083e3d21499"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Mon Nov 16 10:45:35 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 16 10:45:35 2026 +0100"
      },
      "message": "Allow server deadline enforcement to be turned off (#956)\n\n* Allow server deadline enforcement to be turned off\n\nMotivation:\nSince #872 generated handlers bound every call by the client\u0027s\ngrpc-timeout. Handlers before 2.0.0 always ran a call to completion, and a\nservice may rely on that, for example a streamed response that must run to\nits end. There was no way to keep the old behaviour.\n\nModification:\nAdd `pekko.grpc.server.enforce-deadline` (default on) and\n`GrpcServerSettings.enforceDeadline` / `withEnforceDeadline`. The internal\n`GrpcMarshalling.withServerDeadline` helpers now take the settings and\nreturn the response unchanged when enforcement is off. The generated Scala\nand Java handlers pass the settings they already resolve. Neither the\nsettings class nor the helper has been released yet, so the signature\nchange needs no MiMa filter.\n\nResult:\nDeadline enforcement can be turned off for all services in config, or for\none service through the settings passed to its generated handler.\n\nTests:\n- GrpcServerSettingsSpec, ServerDeadlineSpec and ServerTimeoutSpec cover\n  the default, the config key, the `with` method, and that unary and\n  streamed responses outlive the deadline when enforcement is off\n- sbt runtime/Test/compile plugin-tester-scala/Test/compile\n  plugin-tester-java/Test/compile: success\n- sbt runtime/mimaReportBinaryIssues: success\n- scalafmt --mode diff-ref: applied\n- Test runs left to CI\n\nReferences:\nRefs #929\n\n* Test the deadline opt-out through a generated Java handler\n\nMotivation:\nThe opt-out tests covered only the Scala helper and the generated Scala\nhandler, so the Java helper\u0027s early return and the settings passed by the\nJava handler template had no behavioural coverage.\n\nModification:\nAdd JavaServerTimeoutSpec to plugin-tester-java. It drives a generated\nJava handler with a service that replies after the client\u0027s grpc-timeout,\nchecking DEADLINE_EXCEEDED with enforcement on and an OK reply with\n`withEnforceDeadline(false)`.\n\nResult:\nJava and Scala DSL coverage of the deadline opt-out are in parity.\n\nTests:\n- sbt \"plugin-tester-java/testOnly example.myapp.helloworld.JavaServerTimeoutSpec\"\n\n* Hold the reply back in JavaServerTimeoutSpec instead of using a timer\n\nMotivation:\nThe opt-out case let the service reply on its own after 600 ms and then\nchecked the call was still pending after 400 ms. On a busy runner the\ntest thread can resume after the reply, so the check failed even though\nenforcement was correctly off.\n\nModification:\nThe service now returns a CompletableFuture the test completes itself.\nThe test checks the call is pending past its deadline, then releases the\nreply and expects OK. Both cases also check the service was entered.\n\nResult:\nThe test no longer depends on scheduling.\n\nTests:\n- sbt \"plugin-tester-java/testOnly example.myapp.helloworld.JavaServerTimeoutSpec\""
    },
    {
      "commit": "2db21d8cfe76ed8d380cccd170eca083e3d21499",
      "tree": "3a8ec1b71329f55edccf7f1e71da71f692993f53",
      "parents": [
        "614c050f6a035e7b6b1f251da9ef18dbec357b12"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Sun Nov 15 20:28:04 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Nov 15 20:28:04 2026 +0100"
      },
      "message": "Update maven-core, maven-plugin-api to 3.10.1 (#958)"
    },
    {
      "commit": "614c050f6a035e7b6b1f251da9ef18dbec357b12",
      "tree": "687bf792bd2fa20abd523765194c8d50ad1d2e02",
      "parents": [
        "f54736478163ef00b79952f55f5a9ade7c931664"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Sun Nov 15 20:27:55 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Nov 15 20:27:55 2026 +0100"
      },
      "message": "Update logback-classic to 1.6.7 (#957)"
    },
    {
      "commit": "f54736478163ef00b79952f55f5a9ade7c931664",
      "tree": "88ff83830843238a762c3aa04f9d62c80733b831",
      "parents": [
        "fbd9cd7f01686299a92bc493c1aa55c4c904a84c"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sat Nov 14 17:47:53 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Nov 14 17:47:53 2026 +0100"
      },
      "message": "docs: restore the dependencies table in the walkthroughs (#955)\n\nMotivation:\nThe server and client walkthroughs use @@dependencies to show\npekko-grpc-runtime\u0027s dependencies, but the table is empty. The\nsbt-paradox-dependencies plugin that renders the directive was dropped\nwhen the build moved from sbt-paradox-akka to pekko-sbt-paradox, which\ndoes not bring it in, and paradox ignores the unknown directive silently.\n\nModification:\nAdd sbt-paradox-dependencies 0.2.4 back to project/plugins.sbt. It\nenables itself wherever paradox is enabled.\n\nResult:\nBoth walkthroughs render the direct dependencies and dependency tree\ntables for pekko-grpc-runtime again.\n\nTests:\n- sbt docs/paradox: success; server/walkthrough.html and\n  client/walkthrough.html contain the tables\n\nReferences:\nFixes #697"
    },
    {
      "commit": "fbd9cd7f01686299a92bc493c1aa55c4c904a84c",
      "tree": "0d1657ca04e8ddf23d778592ae0452f449106790",
      "parents": [
        "9ecf21a790f64dd47c6a70c3b95b6162f2eaa1ce"
      ],
      "author": {
        "name": "He-Pin(kerr)",
        "email": "hepin1989@gmail.com",
        "time": "Sat Nov 14 18:25:40 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Nov 14 11:25:40 2026 +0100"
      },
      "message": "test: distinguish unresolved addresses from unreachable endpoints (#950)\n\n* test: distinguish unresolved addresses from unreachable endpoints\n\nMotivation:\nThe discovery fixture wrapped unresolved IP addresses in Some(null).\nCorrecting this exposed connection-failure tests that depended on that\ninvalid value and instead encountered DNS failures.\n\nModification:\nUse Option for nullable addresses with resolved/unresolved regression\ncoverage. Use resolved loopback endpoints at port 0 for connection-failure\ntests, assert DNS failure separately, and prevent retry tests from\nswallowing their own failures.\n\nResult:\nDiscovery fixtures follow the address contract without changing production\nresolver semantics. Existing endpoint-selection and status assertions are\npreserved; no tests are newly marked pending.\n\nTests:\nReproduced the helper regression and both Netty mixed-endpoint failures.\n2 resolver and 18 integration/fixture tests pass; 1 existing Pekko HTTP\ntest remains pending. Formatting, header checks and final independent\nadversarial review pass.\n\nReferences:\nThe one-line helper fix is adapted from akka/akka-grpc#1856 (3a18a259),\nnow under Apache 2.0 after its BSL Change Date. The regression tests and\nfollow-up test corrections were written for Pekko.\n\n* fix: skip unresolvable discovery targets instead of failing all\n\nMotivation:\nPekkoDiscoveryNameResolver resolved every target with InetAddress.getByName,\nso one unresolvable host aborted the whole list and the resolver reported\nUNKNOWN even when other targets were usable.\n\nModification:\nSkip targets whose host fails to resolve and rethrow the first\nUnknownHostException only when no target resolves.\n\nResult:\nA discovery result that mixes unresolvable and resolvable hosts now yields\nthe resolvable endpoints. An all-unresolvable result still reports UNKNOWN.\n\nTests:\n- Added PekkoDiscoveryNameResolverSpec cases for mixed and all-unresolvable\n  targets. Not run locally; left to CI.\n\nReferences:\nRefs #950 - review comment on unresolvable hosts in the resolver\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "9ecf21a790f64dd47c6a70c3b95b6162f2eaa1ce",
      "tree": "0c26f6c13602a73b88a93bfd638ed8e3b54733c3",
      "parents": [
        "2fd19503521f4b01d264e3304151242146c73f9f"
      ],
      "author": {
        "name": "He-Pin(kerr)",
        "email": "hepin1989@gmail.com",
        "time": "Sat Nov 14 18:09:32 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Nov 14 11:09:32 2026 +0100"
      },
      "message": "build: add Scalafix OrganizeImports support (#946)\n\nMotivation:\nProvide semantic import organization consistent with apache/pekko#3342.\n\nModification:\nAdd sbt-scalafix 0.14.10 and a sortImports command, preserve relative\nPekko grouping, scope unused-import diagnostics to SemanticDB sessions,\nremove the conflicting Scalafmt rewrite, and organize runtime imports.\nAdd contributor documentation and an isolated regression script.\n\nResult:\nRoot-aggregated projects can use OrganizeImports without changing the\nnormal compiler warning policy or generated sources.\n\nTests:\n- Regression passed on Scala 2.12.21 and 2.13.18.\n- Runtime OrganizeImports, compile and second-pass --check passed.\n- scalafmtAll, scalafmtSbtCheck, headerCheckAll and git diff --check passed.\n- Full test suites and all cross-Scala variants were not run.\n\nReferences:\nRefs apache/pekko#3342 and scalacenter/scalafix#2478."
    },
    {
      "commit": "2fd19503521f4b01d264e3304151242146c73f9f",
      "tree": "88e3a8712da91f7f5e8a4aa29c157743b6af5e04",
      "parents": [
        "1f37f524dd90f52f502e6a997e600d650c507b60"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Nov 13 20:00:13 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 20:00:13 2026 +0100"
      },
      "message": "Update sbt-pekko-build to 0.5.0 (#954)"
    },
    {
      "commit": "1f37f524dd90f52f502e6a997e600d650c507b60",
      "tree": "a879f428813b6cbcbc633e63a273c61d13e54da0",
      "parents": [
        "cd977575c00436342e3f417002cf9fb3d6a48ea5"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Fri Nov 13 13:44:30 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 13:44:30 2026 +0100"
      },
      "message": "build: update Pekko HTTP to 2.0.0 (#952)\n\nMotivation:\nPekko HTTP 2.0.0 is released.\n\nModification:\nUpdate the Pekko HTTP version from the 2.0.0 milestones to 2.0.0 in\nPekkoHttpDependency.scala and in the Gradle and Maven plugin-tester builds.\n\nResult:\nThe sbt build and the plugin testers use Pekko HTTP 2.0.0.\n\nTests:\n- Not run locally - CI builds and tests against 2.0.0\n\nReferences:\nNone - Pekko HTTP 2.0.0 release"
    },
    {
      "commit": "cd977575c00436342e3f417002cf9fb3d6a48ea5",
      "tree": "3a2e2305afcad9b2145710fc0ce8436b91553a70",
      "parents": [
        "3a6c05b2aa7027f55273f5032149a7dc23568d28"
      ],
      "author": {
        "name": "He-Pin(kerr)",
        "email": "hepin1989@gmail.com",
        "time": "Fri Nov 13 19:11:23 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 19:11:23 2026 +0800"
      },
      "message": "Align formatter configuration with Pekko core (#951)\n\nMotivation:\nPreserve Java import ordering across all modules and keep the Pekko core selector layout after the separate OrganizeImports migration.\n\nModification:\nMove javafmtSortImports to ThisBuild, set importSelectors to unfold, and update contributor documentation.\n\nResult:\nConsistent formatter configuration without source formatting churn or duplicating the import-organizer PR.\n\nTests:\nFormatter tasks and checks, runtime/compile, four-module setting assertions, temporary local formatter fixture, and git diff --check passed.\n\nReferences:\nRefs #946; apache/pekko@6049edf9"
    },
    {
      "commit": "3a6c05b2aa7027f55273f5032149a7dc23568d28",
      "tree": "e47b3ee674dbf463b0c2fd32ca12f021ddd85ad1",
      "parents": [
        "f87833f06c435f9bb483790ac60778b10173ee76"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Fri Nov 13 10:35:44 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 10:35:44 2026 +0100"
      },
      "message": "Improved and simplified unary server handlers (rebased #862) (#905)\n\n* Improve handler benchmarks.\n\nAdd coverage of all gRPC method styles with strict and streaming versions, for both Scala and Java frameworks.\n\n* Simplify GrpcMarshalling.handleUnary in Scala DSL.\n\nUsing FastFuture yields essentially the same speedups as manual Future introspection, with much simplified code.\n\n* Simplify GrpcMarshalling.handleUnaryResponse in Java DSL.\n\nCompleteableFuture already applies FastFuture like immediate dispatch on completed futures with equivalent performance.\n\n* Reduce duplication in Scala generated handlers.\n\nRework handler and partial to share a core implementation generated by a Twirl reusable block.\n\n* Remove special case unary strict handling in Java codegen.\n\nPerformance change is negligible over the standard unmarshal/compose path, and the implementation exposed internal details of GrpcProtocolReader that makes evolving internals harder.\n\n* Address the remaining review items from #862.\n\nFollows the replies on the review in\nhttps://github.com/apache/pekko-grpc/pull/862#issuecomment-5564587251\n\n- Build the exception handler only on the failure path. The new\n  GrpcExceptionHandler.recover returns a response that has already\n  succeeded as is, recovers an already failed one directly, and only\n  attaches a recover for responses that are still pending. The generated\n  Scala handlers use it for every RPC style, with the older .fast chains,\n  so GrpcMarshalling.handleUnary is removed.\n- Marshal the response on the handler\u0027s ExecutionContext rather than\n  ExecutionContext.parasitic, so a service implementation that completes\n  asynchronously does not have serialization, or error handling, run on\n  the thread that completed it.\n- Add OutboundDataFrame to the sealed frame hierarchy and narrow\n  encodeDataToResponse to it, so a TrailerFrame can no longer be passed\n  in and produce an empty response (native) or a duplicated trailer\n  (web). Data frame encoding is shared in AbstractGrpcProtocol.\n- Add Java DSL tests that a failed unary implementation is turned into a\n  gRPC error status, in the shape of the generated Java handlers.\n\n* Fix compilation after rebasing onto #877\n\nMotivation:\nAfter #877, javadsl GrpcMarshalling no longer uses InternalApi, and its\nnegotiated method is overloaded, so the explicit type argument in\nJavaHandlerBenchmark stops Scala inferring the lambda\u0027s parameter types.\n\nModification:\nDrop the unused import and the explicit type argument, as #877 did for the\nbenchmark this branch replaces.\n\nResult:\nruntime and benchmarks compile again.\n\n* Add unit tests for GrpcExceptionHandler.recover\n\nMotivation:\nrecover has three branches: already succeeded, already failed, and still\npending. Only the pending failure case was exercised, through an interop\nspec.\n\nModification:\nTest each branch in GrpcExceptionHandlerSpec: a successful response is\nreturned as the same instance without constructing the exception handler; a\nfailure is mapped by the handler or falls back on the default mapper, and the\nresponse fails if the handler throws; a pending response is recovered when it\nfails, and passed through without constructing the handler when it succeeds.\n\nResult:\nAll branches of recover are covered by unit tests.\n\n* Restore the InternalApi import lost when rebasing onto #872\n\n* Sort imports in files changed by this PR\n\nMotivation:\nSome imports touched by this branch were out of order: third-party and\npekko imports interleaved, and scala.concurrent.duration before\nscala.concurrent.\n\nModification:\nGroup the pekko imports together and sort them, ahead of io.grpc and\norg.openjdk, in GrpcExceptionHandler, the javadsl GrpcMarshallingSpec and\nthe handler benchmarks.\n\nResult:\nImport order matches the rest of the codebase. No behaviour change.\n\n* Share one actor system in the javadsl GrpcMarshallingSpec\n\nMotivation:\nEach test created and terminated its own ActorSystem, unlike the shared\nsystem shut down with TestKit that #945 settled on for the test suites.\n\nModification:\nCreate one system for the suite and shut it down in afterAll with\nTestKit.shutdownActorSystem. Also drop a stray blank line in\nDependencies.benchmarks.\n\nResult:\nThe spec follows the same actor system lifecycle as the other suites.\n\n* Leave fatal errors to propagate in GrpcExceptionHandler.recover\n\nMotivation:\nReview on #905: recover matched any Throwable, so a fatal error would be\nturned into a gRPC response instead of propagating. The interop test also\nran on ExecutionContext.global rather than the suite\u0027s actor system.\n\nModification:\nOnly recover NonFatal failures, in both the already failed and the pending\nbranch, and return a response that failed with a fatal error as is. Test the\nalready failed case with FastFuture.failed, since a Promise boxes fatal\nerrors in an ExecutionException. Use system.dispatcher in the interop\nGrpcMarshallingSpec.\n\nResult:\nFatal errors are not hidden behind a gRPC status, and the interop test runs\non its own actor system.\n\n* Fix GrpcExceptionHandlerSpec compilation\n\nMotivation:\nFastFuture.failed is a function value, not a method, so it takes no type\nargument, and runtime tests failed to compile.\n\nModification:\nGive the result a type ascription instead.\n\nResult:\nruntime Test compiles, and GrpcExceptionHandlerSpec passes.\n\n---------\n\nCo-authored-by: Tim Whittington \u003ctim.whittington@indexity.io\u003e"
    },
    {
      "commit": "f87833f06c435f9bb483790ac60778b10173ee76",
      "tree": "019ecd226e21ca4510a91c6077397926968d1548",
      "parents": [
        "bf5da005cd6b3c785cf4065fcace043b9917b773"
      ],
      "author": {
        "name": "He-Pin(kerr)",
        "email": "hepin1989@gmail.com",
        "time": "Fri Nov 13 12:06:00 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 12:06:00 2026 +0800"
      },
      "message": "docs: align build-tool guides with the Java 17 baseline (#947)\n\nMotivation:\nThe build targets Java 17, but the build-tool guides still describe JDK 8\nsupport and adding an ALPN agent.\n\nModification:\nDocument Java 17 as the minimum and built-in ALPN support. Correct the\nserver walkthrough link description.\n\nResult:\nThe guides match the supported runtime instead of recommending an\nunsupported JVM.\n\nTests:\nNot run - docs only. git diff --check and independent static review passed.\n\nReferences:\nAdapted from akka/akka-grpc#1837 (78210d82), whose Apache 2.0 Change Date\nhas passed."
    },
    {
      "commit": "bf5da005cd6b3c785cf4065fcace043b9917b773",
      "tree": "20019b2f3012a41f3ae6f4215adb85f7a208607a",
      "parents": [
        "6de5b775ca0fbd461c5f9500e5d1b8f76cadf857"
      ],
      "author": {
        "name": "He-Pin(kerr)",
        "email": "hepin1989@gmail.com",
        "time": "Fri Nov 13 04:05:21 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 04:05:21 2026 +0800"
      },
      "message": "fix(maven): recompute proto paths after module reconfiguration (#949)\n\nMotivation:\nA reused Maven mojo keeps the first module\u0027s normalized proto paths in a\nlazy val even when its project and protoPaths configuration changes.\n\nModification:\nCompute normalizedProtoPaths on each access, as upstream does. Add a\nregression test for changing the module and then the proto path list.\n\nResult:\nProto discovery and protoc include paths use the current configuration.\n\nTests:\nThe regression failed on the original lazy val, retaining first-module\npaths. All 6 Maven plugin tests pass after the fix. Scala formatting,\n+headerCheckAll, git diff --check and independent static review passed.\n\nReferences:\nAdapted from akka/akka-grpc#1822 (ddc302e4), now under Apache 2.0 after its\nBSL Change Date."
    },
    {
      "commit": "6de5b775ca0fbd461c5f9500e5d1b8f76cadf857",
      "tree": "58d8b656d52667e274d3bdb3662617172cef1f21",
      "parents": [
        "40c2287c87bff24a657202fc32f044c57c896bdf"
      ],
      "author": {
        "name": "He-Pin(kerr)",
        "email": "hepin1989@gmail.com",
        "time": "Fri Nov 13 03:14:52 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 13 03:14:52 2026 +0800"
      },
      "message": "build: link POM metadata to the Apache Pekko release notes (#948)\n\nMotivation:\nArtifact consumers should be able to find Pekko gRPC\u0027s maintained release\nnotes from the POM, using the project\u0027s ASF-hosted documentation.\n\nModification:\nSet releaseNotesURL to the canonical Apache Pekko gRPC release-notes\nindex for non-SNAPSHOT versions. Leave the property absent for development\nsnapshots. Add an explicit checkReleaseNotesPom verification task.\n\nResult:\nRelease POMs link to official project documentation, not GitHub tag pages.\nPublishing behavior is unchanged.\n\nTests:\nGenerated POM checks pass for 2.0.0 and 2.0.1-SNAPSHOT. Canonical notes URL\nverified. scalafmtSbt and git diff --check pass. The embedded benchmark-java\nbuild was verified when the setting and verification task were introduced.\n\nReferences:\nAdapted from akka/akka-grpc#1812 (6cd7c842), now under Apache 2.0 after its\nBSL Change Date, with the destination adjusted to Pekko\u0027s ASF documentation."
    },
    {
      "commit": "40c2287c87bff24a657202fc32f044c57c896bdf",
      "tree": "c0ba1c4b4f392e90d62890723d3ec1b07d2650a8",
      "parents": [
        "10a3a4d32a9905ea9ffd9a3e46d3f26722f75f78"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Nov 12 14:50:34 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 12 14:50:34 2026 +0100"
      },
      "message": "Add a config to stop grpc-web CORS allowing credentials from any origin (#871)\n\n* Add a config to stop grpc-web CORS allowing credentials from any origin\n\nMotivation:\n`WebHandler.defaultCorsSettings` sets `withAllowCredentials(true)` and leaves\n`allowed-origins` at the pekko-http-cors default of `\"*\"`. A POST from an\narbitrary origin is answered with:\n\n    Access-Control-Allow-Origin: https://evil.example\n    Access-Control-Allow-Credentials: true\n\nso a browser sends cookies and HTTP authentication on a cross-origin call and\nlets the calling page read the response. For a service authenticated that way,\nany page a user visits can act as them. Every gRPC-Web deployment is served by\npekko-http, so there is no backend to switch to that avoids this.\n\nModification:\nAdded `pekko.grpc.server.grpc-web.allow-credentials-from-any-origin`, default\n`true`, so existing deployments are unchanged. While it is `true` and the\ncombination is actually in effect, each handler logs a warning naming the two\nways out. Set it to `false` and the credentials permission is dropped: the\nhandler answers `Access-Control-Allow-Origin: *` with no credentials header, and\nbrowsers withhold cookies cross-origin.\n\nThe check is on `HttpOriginMatcher.*` by identity, which is what\n`allowed-origins \u003d \"*\"` parses to. Settings that list origins are a deliberate\nchoice, so they are neither warned about nor modified - credentials keep working\nonce the caller has said who may use them.\n\nBoth entry points are covered. The javadsl handler builds its own CORS route, so\nit would otherwise have bypassed this; `scaladsl.CorsSettings` extends the\njavadsl one, so it reaches the same policy through a match.\n\nResult:\nThe insecure combination is reported wherever it is live, and can be turned off\nwith a config setting rather than a code change. The default is unchanged\nbehaviour.\n\nTests:\n- `WebHandlerCorsSpec` asserts the response headers in both modes, for actual\n  and preflight requests, plus that restricted-origin settings are left alone\n- Confirmed the guard bites: not applying the policy fails 2 tests\n- sbt \"runtime/testOnly ...WebHandlerCorsSpec\" - 8 passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt docs/paradox - passed\n- sbt scalafmtAll scalafmtSbt - applied\n- sbt \"runtime/test\" - not run locally, left to CI\n\nCo-Authored-By: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e\n\n* scalafmt\n\n* Make grpc-web CORS secure by default\n\nMotivation:\nThe default grpc-web CORS settings allow any origin with credentials, which\nlets any web page make credentialed cross-origin calls and read the responses.\n2.0.0 is the release to stop shipping that as the default.\n\nModification:\nDefault pekko.grpc.server.grpc-web.allow-credentials-from-any-origin to false.\nSetting it to true keeps the 1.x behaviour, with the existing warning. Update\nreference.conf, scaladoc and the grpc-web docs, and test the default config.\n\nResult:\nOut of the box, credentials are dropped while the origins are unrestricted.\nRestricted origins keep credentials, and users can opt back in to 1.x.\n\n* Shut down the extra actor systems in WebHandlerCorsSpec\n\n---------\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "10a3a4d32a9905ea9ffd9a3e46d3f26722f75f78",
      "tree": "920cf48254415ec3939d13643643f45e612d955f",
      "parents": [
        "981728a1f1d64c2b43702f0f02c8161db4f57870"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Nov 12 14:50:10 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 12 14:50:10 2026 +0100"
      },
      "message": "Shut down the actor systems created by tests (#945)\n\nMotivation:\nSeveral specs create an ActorSystem and never shut it down, so its\nthreads and any bound ports linger until the forked test JVM exits.\n\nModification:\nAdd an afterAll that shuts the system down to the TestKit specs that\nlacked one, shut the system down as well as the server binding in\nHostnameVerificationSpec and PowerApiSpec, move the interop\nGrpcMarshallingSpec system to the suite so it can be shut down, close\nthe client, binding and system in the scala3 scripted GreeterServiceSpecs,\nand wait for the termination that a few specs fired and forgot.\n\nResult:\nEvery test that creates an actor system shuts it down when it finishes.\n\nTests:\n- scalafmt run on the changed files; test suites left to CI\n\nReferences:\nRefs #871 - review comment about shutting down extra actor systems"
    },
    {
      "commit": "981728a1f1d64c2b43702f0f02c8161db4f57870",
      "tree": "d5c83a346b9842e9c80005cb22258193808eee18",
      "parents": [
        "6c6464d8fbd60b0be96ec20f866733fdcb608ae1"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Nov 12 13:05:20 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 12 13:05:20 2026 +0100"
      },
      "message": "feat(runtime): Support setting a deadline per call (#922)\n\n* feat(runtime): Support setting a deadline per call\n\nMotivation:\nA deadline could only be set for all calls of a client, through\nGrpcClientSettings.\n\nModification:\nPort akka/akka-grpc#1838: add setDeadline to the scaladsl and javadsl\nSingleResponseRequestBuilder and StreamResponseRequestBuilder.\nDifferences from upstream:\n- The builder replaces the deadline of its client settings rather than\n  its CallOptions, so the deadline is counted from when the call is\n  invoked, not from when it is set.\n- The request builders now apply the deadline for every call (including\n  unary invoke), and the Netty channel no longer re-applies the client\n  deadline, so a per-call deadline overrides the client deadline on both\n  backends.\n- Document deadlines in the client details page.\n\nResult:\nUsers can set a deadline for a single call with\nclient.sayHello().setDeadline(...).invoke(request).\n\nTests:\n- New PerCallDeadlineSpec (netty and pekko-http) and\n  PerCallDeadlineTest (javadsl). Without the change they do not compile.\n  With upstream\u0027s approach, \"override the deadline of the client\n  settings\" and \"start when the call is invoked\" fail; both pass here\n- sbt runtime/test: 189 passed\n- sbt interop-tests/test: 264 passed\n- sbt runtime/mimaReportBinaryIssues: passed\n- sbt headerCheckAll, interop-tests/Test/javafmt (JDK 17), docs/paradox\n\nReferences:\nRefs akka/akka-grpc#1838\n\n* Handle Java deadlines too large for a FiniteDuration\n\nMotivation:\nConverting a Java deadline with Duration.fromNanos(d.toNanos) throws an\nArithmeticException for durations beyond about 292 years, such as\nDuration.ofSeconds(Long.MaxValue), which callers use to mean \"no deadline\".\nGrpcClientSettings.withDeadline(java.time.Duration) had the same problem.\n\nModification:\nConvert with scala.jdk.DurationConverters in one internal helper used by both\nthe per-call setDeadline and GrpcClientSettings.withDeadline. A duration too\nlong for a FiniteDuration means no deadline, one too negative means already\nexpired, and null means no deadline in both places.\n\nResult:\nLarge Java deadlines no longer throw."
    },
    {
      "commit": "6c6464d8fbd60b0be96ec20f866733fdcb608ae1",
      "tree": "4f55f55fc23ef0256ba951a7f339bc3327c0640e",
      "parents": [
        "9997a2915fb4e1c337b0bdce3ed9ef10d00640f4"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Nov 12 12:53:11 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 12 12:53:11 2026 +0100"
      },
      "message": "feat(runtime): Support client-side request compression (#927)\n\n* feat(runtime): Support client-side request compression\n\nMotivation:\nPekko gRPC clients always sent uncompressed requests, with no setting\nto enable compression on either backend.\n\nModification:\n- Add a compression client setting (GrpcClientSettings.withCompression\n  and pekko.grpc.client.\"*\".compression): gzip, deflate, or identity\n  (the default) for no compression. Unsupported names are rejected.\n- pekko-http backend: use the matching codec for the request writer.\n- netty backend: set the compression in the CallOptions, and register\n  a deflate compressor with the channel, since grpc-java only ships\n  gzip.\n- Document request compression in the client configuration page.\n\nResult:\nClients can compress requests with gzip or deflate on both backends.\n\nTests:\n- New ClientCompressionSpec (netty and pekko-http) checks the\n  grpc-encoding header and compressed flag received by the server;\n  without the backend wiring the 8 compression tests fail\n- New GrpcClientSettingsSpec tests for the setting\n- sbt runtime/test: 205 passed\n- sbt interop-tests/test: 266 passed\n- sbt runtime/mimaReportBinaryIssues, headerCheckAll, docs/paradox: passed\n\nReferences:\nFixes #923. Depends on #926\n\n* Share the deflate codec between backends and check every compressed frame\n\nMotivation:\nAfter #926, the netty client had two deflate implementations: the\nDeflateCodec added here for request compression, and the DeflateDecompressor\nfrom #926 for responses. Both duplicated the streams in Deflate. Looking up a\ncodec by name was also repeated, as Codecs.supportedCodecs.find, in\nPekkoHttpClientUtils and GrpcClientSettings. ClientCompressionSpec only checked\nthe compressed flag of the first request frame.\n\nModification:\nUse one grpc-java DeflateCodec built on Deflate\u0027s own streams for both the\ncompressor and decompressor registries, and drop DeflateDecompressor. Add\nCodecs.byName and use it in both places. ClientCompressionSpec now parses\nevery request frame and checks each one\u0027s compressed flag, with a\nthree-message stream.\n\nResult:\nOne deflate implementation serves both backends and both directions, and\nevery frame of a compressed request is checked."
    },
    {
      "commit": "9997a2915fb4e1c337b0bdce3ed9ef10d00640f4",
      "tree": "4f6ef947535d5366914c57a7d2982ddec5637d74",
      "parents": [
        "0b94ea1ff83e161c73caab1c8d5c9743c93849cf"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Nov 12 12:06:24 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 12 12:06:24 2026 +0100"
      },
      "message": "Honour grpc-timeout on the server, and send it from the pekko-http client (#872)\n\n* Honour grpc-timeout on the server, and send it from the pekko-http client\n\nMotivation:\n`grpc-timeout` appeared in this repository only in the gRPC-Web CORS allow-list.\nNothing parsed it, so a server kept working on a reply the client had already\ngiven up on: the client stops waiting when its own deadline expires, and the\nserver was never told there was one.\n\nThe pekko-http client made that worse by never sending the header. `applyDeadline`\nbounded the wait on the client side only, so a server talking to it had no way to\nlearn of a deadline at all. The netty client sends it via grpc-java.\n\nModification:\n- Added the `grpc-timeout` header model: a positive integer of at most 8 digits\n  and a unit character, per the HTTP/2 protocol document. Rendering picks the\n  finest unit whose value still fits in 8 digits.\n- `GrpcMarshalling.withServerDeadline` bounds a response by the requested\n  deadline, completing with `DEADLINE_EXCEEDED` on expiry. Generated Scala and\n  Java handlers apply it inside the existing `negotiatedWithMaxSize` block, so\n  the protocol writer is already in scope and the request is not parsed twice.\n- The pekko-http client sends `grpc-timeout` derived from `CallOptions`. An\n  already-expired deadline sends nothing: such a call is failed outright by\n  `applyDeadline`, and a non-positive timeout is not representable on the wire.\n\nA malformed value is ignored rather than rejected. The timeout is a hint from the\npeer, and refusing to serve the request would be a worse outcome than serving it\nwithout a deadline.\n\nResult:\nA server stops producing a response once the client\u0027s deadline has passed, and\nboth backends now tell the server what that deadline is.\n\nTests:\n- `TimeoutSpec` covers every unit, the 8-digit cap, malformed values, and that\n  encoding round-trips without exceeding the original duration\n- `ServerDeadlineSpec` covers the enforcement helper, including that it returns\n  well before a slow handler would have\n- `ClientTimeoutHeaderSpec` covers what the client puts on the wire\n- `ServerTimeoutSpec` drives a generated handler end to end, so the codegen\n  wiring is covered and not just the helper it calls; it asserts the service was\n  entered, so a pass cannot come from a routing miss\n- Confirmed the guard bites: with the deadline never applied, 2 tests fail\n- sbt \"runtime/testOnly ...ServerDeadlineSpec ...ClientTimeoutHeaderSpec ...TimeoutSpec\" - 20 passed\n- sbt \"plugin-tester-scala/testOnly ...ServerTimeoutSpec\" - 2 passed\n- sbt \"plugin-tester-scala/Test/compile\" \"plugin-tester-java/Test/compile\" \"interop-tests/Test/compile\" - passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt scalafmtAll scalafmtSbt - applied\n- sbt \"runtime/test\" - not run locally, left to CI\n\nNote: the deadline bounds the response, not the work behind it. A service that\nhas started an expensive operation keeps running it; only the reply is abandoned.\nDocumented in server/details.md.\n\nCo-Authored-By: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e\n\n* Bound streamed bodies by grpc-timeout and cancel the deadline timer\n\nMotivation:\nwithServerDeadline raced the response Future against pattern.after. The timer\nwas never cancelled, so each request kept a scheduler task until its\ndeadline even after replying. Only the response headers were bounded: a\nstreamed body could outlive the deadline indefinitely. The comments claimed\nthe server stops working at the deadline, which it did not. Timeout.parse also\naccepted a signed amount such as \"+5S\".\n\nModification:\nUse scheduleOnce and cancel the timer when the response completes. Bound a\nchunked body by the remaining deadline: on expiry it ends with\nDEADLINE_EXCEEDED trailers and cancels the source producing it. Describe what\nthe deadline does and does not stop. Reject a grpc-timeout amount that is not\nall digits. Fix the rebase onto #877 (negotiated with GrpcServerSettings).\n\nResult:\nServer streams stop at the client\u0027s deadline, no timers outlive the calls\nthey bound, and signed timeouts are rejected.\n\n* Describe streamed-response deadlines in the server docs\n\n---------\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "0b94ea1ff83e161c73caab1c8d5c9743c93849cf",
      "tree": "bdf7b85432beaa904d09826610f05c29bab6bbcf",
      "parents": [
        "8a1511f282efba4bfa686d17aab9efd3ae4b1ebf"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Nov 11 21:41:37 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 11 21:41:37 2026 +0100"
      },
      "message": "Minimum TLS version setting (#823)\n\n* Minimum TLS version setting\n\nRebased onto main, which now carries #818 and #869. The conflicts were purely\nadditive: `maxInboundMessageSize` and `minimumTlsVersion` both extend the\n`GrpcClientSettings` constructor and `copy`, and `PercentEncoding` and\n`SSLContextUtils` are both imported by PekkoHttpClientUtils.\n\nThe floor is applied inside `PekkoHttpClientUtils.sslContextFor`, so both the\nverifying and the `verify-hostname \u003d false` connection contexts inherit it, and\nin `NettyClientUtils.nettySslContext`, which also keeps #818\u0027s\n`maxInboundMessageSize` wiring on the same builder.\n\nCo-Authored-By: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e\n\n* Keep the TLS helpers internal, validate the version early, and document it\n\nMotivation:\nThe new SSLContextUtils methods, including the SSLContextSpi wrapper, were\npublic API on a public object. An unknown minimum-tls-version was only\nrejected when a channel was created. The setting was not documented, nor was\nthe JVM-wide alternative.\n\nModification:\nMake enabledProtocols and withMinimumTlsVersion private[grpc] and\n@InternalApi. Validate the version in GrpcClientSettings.withMinimumTlsVersion\nand fromConfig, sharing the check with the protocol filtering. Add a \"Minimum\nTLS version\" section to the client configuration docs, covering both backends,\na supplied SSLContext, and jdk.tls.client.protocols.\n\nResult:\nNo new public TLS helper API, typos fail when the settings are built, and the\nsetting is documented.\n\n* Respect JDK-enabled TLS protocols for the minimum TLS version\n\nMotivation:\nenabledProtocols read getSupportedSSLParameters, so a TLSv1.2 floor on a JVM\nstarted with -Djdk.tls.client.protocols\u003dTLSv1.3 re-enabled TLSv1.2, widening\nwhat the JDK allows instead of only narrowing it.\n\nModification:\nRead getDefaultSSLParameters (the enabled protocols) instead. Replace the test\nthat asserted the supported-set behaviour with one using an SSLContext that\nsupports TLSv1.2 but enables only TLSv1.3, and document that the setting never\nre-enables a JDK-disabled version.\n\nResult:\nminimum-tls-version can only narrow the JDK\u0027s enabled TLS versions.\n\nTests:\n- scalafmt run on the changed Scala files; test run left to CI\n\nReferences:\nhttps://github.com/apache/pekko-grpc/pull/823#discussion_r4540943113\n\n* Shut down the actor system after MinimumTlsVersionSpec\n\nMotivation:\nafterAll unbound the server but never terminated the TestKit actor system,\nleaking it for the rest of the test run.\n\nModification:\nShut the actor system down in afterAll, after unbinding, even if the unbind\nfails.\n\nResult:\nThe spec releases its actor system when it finishes.\n\nTests:\n- scalafmt run on the changed file; test run left to CI\n\nReferences:\nhttps://github.com/apache/pekko-grpc/pull/823#discussion_r4540971008\n\n---------\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "8a1511f282efba4bfa686d17aab9efd3ae4b1ebf",
      "tree": "cd48f19cbafb45fa3d8e88e1ef63ce73be3c74e9",
      "parents": [
        "e5fd9e55de5dd1178acad29eee5a3c1d4bb40542"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Nov 11 18:19:17 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 11 18:19:17 2026 +0100"
      },
      "message": "feat(runtime): Support deflate message compression (#926)\n\n* feat(runtime): Support deflate message compression\n\nMotivation:\nOnly gzip and identity were supported. A client compressing requests\nwith deflate (e.g. a gRPC C-core based client) got UNIMPLEMENTED.\nCodecs.negotiate only read the first grpc-accept-encoding header, so\nencodings in later headers were ignored.\n\nModification:\n- Add a Deflate codec using the zlib format (RFC 1950), and add it to\n  Codecs.supportedCodecs.\n- Move Gzip\u0027s size-bounded decompression into StreamingZipCodec, which\n  Gzip and Deflate extend, so deflate also enforces\n  max-inbound-message-size while decompressing.\n- Message-Accept-Encoding.findIn returns the encodings of all\n  grpc-accept-encoding headers, in order, and negotiate picks the\n  first supported one.\n- Document message compression in the server details page.\n\nResult:\nThe server accepts deflate requests and sends deflate responses when\ndeflate is the first supported encoding the client accepts.\n\nTests:\n- New tests in CodecsSpec, MaxInboundMessageSizeSpec and a new\n  DeflateCompressionSpec; without Deflate in supportedCodecs the\n  negotiation and end-to-end tests fail, and without the findIn change\n  the multiple-headers test fails\n- sbt runtime/test: 200 passed\n- sbt interop-tests/test: 256 passed\n- sbt runtime/mimaReportBinaryIssues, headerCheckAll, docs/paradox: passed\n\nReferences:\nFixes #924\n\n* Accept deflate responses on the netty client and advertise encodings on UNIMPLEMENTED\n\nMotivation:\nWith deflate in Codecs.supportedCodecs, the pekko-http client advertises and\ndecodes deflate responses, but the netty client uses grpc-java\u0027s default\nDecompressorRegistry, which only knows gzip. The two backends accepted\ndifferent response encodings.\n\nA request with an unsupported grpc-encoding was rejected with UNIMPLEMENTED\nbut without grpc-accept-encoding. The gRPC compression spec asks for it, so a\ncompressing client can fall back to an encoding the server supports.\n\nModification:\nGive the netty channel a DecompressorRegistry with gzip, deflate (a zlib\nInflaterInputStream decompressor) and identity, advertised in the same order\nas the pekko-http client. Attach grpc-accept-encoding, listing the supported\nencodings, to the UNIMPLEMENTED error from Codecs.detect. Its description now\nnames the encoding rather than printing an Option.\n\nResult:\nBoth client backends accept the same response encodings, and a client using\nan unsupported encoding is told which ones the server supports.\n\n* Test both client backends receiving deflate responses end to end\n\nMotivation:\nDeflateCompressionSpec drives the generated handler with hand-built HTTP\nrequests and decodes the frames itself, so nothing showed that the real\nnetty and pekko-http clients decode deflate responses.\n\nModification:\nAdd DeflateResponseClientSpec, run with each backend against a generated\nhandler. Both clients list gzip before deflate, so the server side checks that\nthe client advertised deflate, then keeps only deflate from the\ngrpc-accept-encoding it advertised. The test checks that the responses really\nwere deflate-encoded, for a unary call and a server stream.\n\nResult:\nBoth client backends are tested decoding deflate responses."
    },
    {
      "commit": "e5fd9e55de5dd1178acad29eee5a3c1d4bb40542",
      "tree": "f87533a3ebca2aa74e0cf6e4954bed7ba7232c9c",
      "parents": [
        "0e7a2d80568359c795a4255b11259e44a392baf0"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Nov 11 11:04:30 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 11 11:04:30 2026 +0100"
      },
      "message": "Add a javadsl telemetry test (#878)\n\nMotivation:\n`TelemetrySpi.onRequest` is called from both the Scala and the Java generated\nserver handlers, but only the Scala path is covered by a test, so a regression\nin the Java handler template would go unnoticed.\n\nModification:\nAdd `TelemetryTest`, the javadsl counterpart of the existing scaladsl\n`TelemetrySpec`: it configures `pekko.grpc.telemetry-class`, calls the handler\nbuilt by the generated `GreeterServiceHandlerFactory` with a unary request, and\nasserts the collected prefix, method and content type.\n\nResult:\nThe telemetry hook is covered on both generated handler paths.\n\nTests:\n- sbt \"interop-tests/testOnly org.apache.pekko.grpc.javadsl.TelemetryTest\n  org.apache.pekko.grpc.scaladsl.TelemetrySpec\" - passed\n- Removing `spi.onRequest` from the Java handler template fails the new test\n  (expected: \u003c1\u003e but was: \u003c0\u003e); the template is unchanged on this branch\n\nReferences:\nNone - test coverage for the javadsl server handler telemetry hook"
    },
    {
      "commit": "0e7a2d80568359c795a4255b11259e44a392baf0",
      "tree": "8b1cccb186a71d6b9efcaefe81787433b59a73b5",
      "parents": [
        "a64c1912895c5cc303c000a3b8055970cc8ebd13"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Nov 11 07:05:38 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 11 07:05:38 2026 +0100"
      },
      "message": "fix(runtime): Avoid NPE for a StatusRuntimeException without trailers (#920)\n\n* fix(runtime): Avoid NPE for a StatusRuntimeException without trailers\n\nMotivation:\nA StatusRuntimeException created without trailers has null trailers.\nGrpcServiceException(ex) and the javadsl GrpcExceptionHandler wrapped\nthem in a GrpcMetadataImpl, which failed with a NullPointerException\nwhen the trailers were written to the response.\nGrpcServiceException(code, message, details) also used a plain\nGrpcMetadataImpl, so its metadata was not a MetadataStatus.\n\nModification:\nPort akka/akka-grpc#1855:\n- GrpcServiceException(ex) uses empty metadata when there are no\n  trailers.\n- GrpcServiceException(code, message, details) uses\n  RichGrpcMetadataImpl.\n- GrpcMetadataImpl requires a non-null delegate.\nPekko change on top of upstream: the javadsl GrpcExceptionHandler\nfalls back to empty metadata, as the scaladsl handler already does.\n\nResult:\nA StatusRuntimeException without trailers maps to a gRPC status\nresponse instead of an HTTP 500.\n\nTests:\n- New GrpcServiceExceptionSpec: all 3 tests fail without the fix\n  (2 with the NPE, 1 not a MetadataStatus), pass with it\n- sbt runtime/test: 192 passed\n- sbt runtime/mimaReportBinaryIssues: passed\n\nReferences:\nRefs akka/akka-grpc#1855, akka/akka-grpc#1854\n\n* refactor(runtime): Let GrpcMetadataImpl handle missing trailers\n\nMotivation:\nReview feedback: requiring a non-null delegate pushed null handling\nonto every caller.\n\nModification:\nGrpcMetadataImpl treats a null delegate as empty metadata, and\nRichGrpcMetadataImpl.raw uses Option. The null checks in\nGrpcServiceException.apply and the javadsl GrpcExceptionHandler are\nreverted.\n\nResult:\nSimpler fix with the same behaviour.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n\n---------\n\nCo-authored-by: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "a64c1912895c5cc303c000a3b8055970cc8ebd13",
      "tree": "9660b1152ae7402b403fd0e1fc419e712ecd1eff",
      "parents": [
        "3c59910623e2a12e61b342c9b88e94869b82edcf"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Nov 10 18:45:10 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Nov 10 18:45:10 2026 +0100"
      },
      "message": "Pass GrpcServerSettings to the negotiate entry points (#877)\n\nMotivation:\nmaximum inbound message size, and #747 proposes another `negotiate` variant for\ncompression. One entry point per setting does not scale, and each new one is a\nchange to the generated handlers as well.\n\nModification:\n`GrpcProtocol.negotiate` and `GrpcMarshalling.negotiated` (both DSLs) now take a\n`GrpcServerSettings` instead of an `Int`, as an overload of `negotiated` rather\nthan a separate name. `negotiatedWithMaxSize` is removed; the settings-less\nforms delegate with the new package-private `GrpcServerSettings.defaults`. The\ngenerated handlers resolve their settings once and pass them straight through,\nso their own signatures are unchanged.\n\nThe javadsl overload does not resolve when a Scala caller applies an explicit\ntype argument, so `JavaUnaryHandlerBenchmark` now lets `T` be inferred.\n\nResult:\nFurther server handler configuration is added to `GrpcServerSettings` rather\nthan as another negotiate entry point and another generated-handler parameter.\n\nTests:\n- sbt \"runtime/testOnly org.apache.pekko.grpc.*\" - 174 passed, including the new\n  GrpcMarshallingNegotiationSpec\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt \"runtime/Test/compile\" \"benchmarks/Test/compile\" \"plugin-tester-scala/Test/compile\"\n  \"plugin-tester-java/Test/compile\" \"interop-tests/Test/compile\" - passed\n- sbt \"++3.3.8 runtime/Test/compile\" \"++3.3.8 plugin-tester-scala/Test/compile\"\n  \"++3.3.8 benchmarks/Test/compile\" - passed\n\nReferences:\nRefs #875"
    },
    {
      "commit": "3c59910623e2a12e61b342c9b88e94869b82edcf",
      "tree": "53afe0be1cb32ea2bd8636c283d48c234a771ea2",
      "parents": [
        "acd0d41c56c8467d240802de214e2aabead68364"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Nov 10 18:44:18 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Nov 10 18:44:18 2026 +0100"
      },
      "message": "fix(runtime): Cancel server streams in the pekko-http client backend (#921)\n\nMotivation:\nWith the pekko-http client backend, cancelling a server streaming call\n(e.g. taking only the head of the response) never reached the server.\nThe CancellationBarrierGraphStage kept reading the response so the\ntrailers could be read, so the server stream was never stopped.\n\nModification:\nPort akka/akka-grpc#1832: responseToSource takes a streamingResponse\nflag, and applies the cancellation barrier only to single responses.\nThe pekko-http channel passes the flag for each call.\n\nResult:\nCancelling a server streaming call with the pekko-http backend\ncompletes the stream on the server.\n\nTests:\n- New PekkoHttpClientCancelSpec (copied from akka/akka-grpc#1832):\n  fails without the fix (server stream not completed within 5s),\n  passes with it\n- sbt runtime/test: 189 passed\n- sbt interop-tests/test: 255 passed\n- sbt runtime/mimaReportBinaryIssues: passed\n\nReferences:\nRefs akka/akka-grpc#1832"
    },
    {
      "commit": "acd0d41c56c8467d240802de214e2aabead68364",
      "tree": "46dbab48dd5aa7272509bd44f05f4b364aed8a25",
      "parents": [
        "43f340807b0179ae6f7f170d356bab13abd27533"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Nov 10 09:57:36 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Nov 10 09:57:36 2026 +0100"
      },
      "message": "feat(runtime): Widen Java API collection parameters (#935)\n\nMotivation:\nSeveral Java API methods take java.util.List or invariant generic\ncollections, so Java callers have to copy or up-cast their collections.\nFor example, GrpcServiceException.create rejects a List\u003cLocalizedMessage\u003e,\nand our own docs snippet works around it with ArrayList\u003cGeneratedMessage\u003e.\n\nModification:\n- GrpcServiceException.create takes Iterable\u003c? extends GeneratedMessage\u003e.\n- WebHandler.grpcWebHandler takes an Iterable of handlers.\n- MetadataBuilder.fromHeaders takes Iterable\u003c? extends HttpHeader\u003e.\n- ServerReflection.create takes Iterable\u003c? extends ServiceDescription\u003e.\n- ServiceHandler.handler and concatOrNotFound gain Iterable overloads\n  next to the varargs ones.\n- GrpcClientSettings gains withResolveTimeout(java.time.Duration).\n- Simplify the rich error model Java docs snippet.\n- Add MiMa excludes for the changed signatures.\n\nResult:\nJava callers can pass any Iterable, and collections of subtypes, without\ncopying or casting.\n\nTests:\n- sbt \"runtime/testOnly org.apache.pekko.grpc.javadsl.*\": passed\n  (new JavaApiParameterTypesTest fails to compile against the old API)\n- sbt \"+runtime/Test/compile\": passed (2.13, 3.3)\n- sbt \"interop-tests/testOnly *RichErrorModel*\": passed\n- sbt \"runtime/mimaReportBinaryIssues\": passed\n- sbt runtime/scalafmtCheckAll, runtime/javafmtAll, interop-tests/javafmtAll: run\n\nReferences:\nNone - Java API cleanup for 2.0.0"
    },
    {
      "commit": "43f340807b0179ae6f7f170d356bab13abd27533",
      "tree": "08c18aab58696c6e570e42717cf8175f8894a764",
      "parents": [
        "e740d063339fe62936a40675c13fcbca974492cf"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Nov 09 06:54:53 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 09 06:54:53 2026 +0100"
      },
      "message": "build(deps): bump actions/setup-java from 6.1.0 to 6.2.0 (#944)\n\nBumps [actions/setup-java](https://github.com/actions/setup-java) from 6.1.0 to 6.2.0.\n- [Release notes](https://github.com/actions/setup-java/releases)\n- [Commits](https://github.com/actions/setup-java/compare/b843ebf9b913146b38da2eaa7aeddf76074ae1ad...77dd5b2ce225ea11469928ac317850245a74ae78)\n\n---\nupdated-dependencies:\n- dependency-name: actions/setup-java\n  dependency-version: 6.2.0\n  dependency-type: direct:production\n  update-type: version-update:semver-minor\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "e740d063339fe62936a40675c13fcbca974492cf",
      "tree": "25640cdf5d7692e004b9f04c8ad999802e3e9e73",
      "parents": [
        "1ef9c9873f786b6160cacf8140c10892cb501dea"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Nov 09 06:54:46 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 09 06:54:46 2026 +0100"
      },
      "message": "build(deps): bump coursier/setup-action from 3.0.4 to 3.1.0 (#943)\n\nBumps [coursier/setup-action](https://github.com/coursier/setup-action) from 3.0.4 to 3.1.0.\n- [Release notes](https://github.com/coursier/setup-action/releases)\n- [Commits](https://github.com/coursier/setup-action/compare/648df969f41ef15fda2baba8b37f9fa3d16390a3...d5a17f9aaeb260f9ba336a5b582fb49f79c64ac4)\n\n---\nupdated-dependencies:\n- dependency-name: coursier/setup-action\n  dependency-version: 3.1.0\n  dependency-type: direct:production\n  update-type: version-update:semver-minor\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "1ef9c9873f786b6160cacf8140c10892cb501dea",
      "tree": "c5c436633ef0e06422fabffe6af4fe9f01ea5958",
      "parents": [
        "b029070808e2b6001e793b176938b05357d0660e"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Nov 09 06:54:39 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 09 06:54:39 2026 +0100"
      },
      "message": "build(deps): bump sbt/setup-sbt from 1.5.13 to 1.5.14 (#942)\n\nBumps [sbt/setup-sbt](https://github.com/sbt/setup-sbt) from 1.5.13 to 1.5.14.\n- [Release notes](https://github.com/sbt/setup-sbt/releases)\n- [Commits](https://github.com/sbt/setup-sbt/compare/b113b960ba361e87e08ce02a390d51871b7ee459...67d461eba2c4753196c230e02da850757ba4775e)\n\n---\nupdated-dependencies:\n- dependency-name: sbt/setup-sbt\n  dependency-version: 1.5.14\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "b029070808e2b6001e793b176938b05357d0660e",
      "tree": "e757aa7d169443c9115c3be73a6210d508b2a55b",
      "parents": [
        "846d3766e1556c3f5eeb8d1333cb3e327732b670"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Nov 05 23:57:22 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 05 23:57:22 2026 +0100"
      },
      "message": "Update Pekko core to 2.0.0 (#940)\n\n* Test against Pekko 2.0.0-RC1 from Apache staging\n\nMotivation:\nPekko 2.0.0-RC1 is staged in the Apache staging repository and needs to be\ntested against this module before the release vote.\n\nModification:\nAdd Resolver.ApacheMavenStagingRepo as a resolver and set the Pekko core\ndependency to 2.0.0.\n\nResult:\nCI builds and tests this module against the staged Pekko 2.0.0 artifacts.\n\nTests:\n- Not run locally - CI runs the build against the staged artifacts\n\nReferences:\nNone - Pekko 2.0.0-RC1 testing, not to be merged\n\n* Remove the Apache staging resolver now that Pekko 2.0.0 is released\n\n* Update Pekko to 2.0.0 in the plugin testers\n\nMotivation:\nThe Gradle tests fail on this PR because plugin-tester-scala still pins\nPekko 2.0.0-M4 while pekko-grpc-runtime now brings in 2.0.0. The mixed\nversions trip Pekko\u0027s version check when the ActorSystem starts.\n\nModification:\nSet the Pekko version to 2.0.0 in the build.gradle and pom.xml of\nplugin-tester-java and plugin-tester-scala.\n\nResult:\nThe plugin testers use a single Pekko version, matching the sbt build.\n\nTests:\n- Not run locally - relies on the CI Gradle and Maven jobs\n\nReferences:\nNone - follow-up to the Pekko 2.0.0 upgrade in this PR"
    },
    {
      "commit": "846d3766e1556c3f5eeb8d1333cb3e327732b670",
      "tree": "6b31ecb8a44339bd5102e5fa8bf062beaad18963",
      "parents": [
        "29b02a05b12eab231673465880914f5fe9e91521"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Thu Nov 05 19:16:31 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 05 19:16:31 2026 +0100"
      },
      "message": "Update org.eclipse.jgit to 7.8.1.202611041320-r (#941)"
    },
    {
      "commit": "29b02a05b12eab231673465880914f5fe9e91521",
      "tree": "6dd102f4745d72ab5550291bc4db6eaf42d7b4c2",
      "parents": [
        "447498b75c566124325cd504c11fa66e32efc3d9"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Nov 02 14:14:44 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 02 14:14:44 2026 +0800"
      },
      "message": "build(deps): bump sbt/setup-sbt from 1.5.12 to 1.5.13 (#938)\n\nBumps [sbt/setup-sbt](https://github.com/sbt/setup-sbt) from 1.5.12 to 1.5.13.\n- [Release notes](https://github.com/sbt/setup-sbt/releases)\n- [Commits](https://github.com/sbt/setup-sbt/compare/ecad189dd245444b9ae0a0cd5922fd938b88906f...b113b960ba361e87e08ce02a390d51871b7ee459)\n\n---\nupdated-dependencies:\n- dependency-name: sbt/setup-sbt\n  dependency-version: 1.5.13\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "447498b75c566124325cd504c11fa66e32efc3d9",
      "tree": "33fc6f1cdbae436a368d6c43553cd2fc24f693bf",
      "parents": [
        "f19e62605f389b854303556831e7542c54f0d9a1"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Nov 02 14:14:15 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 02 14:14:15 2026 +0800"
      },
      "message": "build(deps): bump coursier/cache-action from 8.2.0 to 9.0.1 (#939)\n\nBumps [coursier/cache-action](https://github.com/coursier/cache-action) from 8.2.0 to 9.0.1.\n- [Release notes](https://github.com/coursier/cache-action/releases)\n- [Commits](https://github.com/coursier/cache-action/compare/057d08c2f9a2c4e7a4d33df3aae657b7db88b8cc...ee0650f98295f27a505e4bfd284c9bfaa913a465)\n\n---\nupdated-dependencies:\n- dependency-name: coursier/cache-action\n  dependency-version: 9.0.1\n  dependency-type: direct:production\n  update-type: version-update:semver-major\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "f19e62605f389b854303556831e7542c54f0d9a1",
      "tree": "818c554c45f8bf67ece9fafc38a08972c5b3b56f",
      "parents": [
        "6e66cab99c298b3a8c6d716ca5a3aeb8288bad9e"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Wed Oct 28 19:36:00 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Oct 28 18:36:00 2026 +0000"
      },
      "message": "Update scala3-library to 3.3.9 (#936)"
    },
    {
      "commit": "6e66cab99c298b3a8c6d716ca5a3aeb8288bad9e",
      "tree": "743a604d5ac09444be69c7be6fd585093f66f3e3",
      "parents": [
        "6dccb21e7d4f297b4ee3afdfc7a1c7b79ac33af0"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Wed Oct 28 19:35:53 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Oct 28 18:35:53 2026 +0000"
      },
      "message": "Update sbt, scripted-plugin to 1.13.1 (#937)"
    },
    {
      "commit": "6dccb21e7d4f297b4ee3afdfc7a1c7b79ac33af0",
      "tree": "d10b15647768c75e6392b72248ef9bf99fb81705",
      "parents": [
        "c61dcd25cd262e327a5bc34f96e9341ed2e0913a"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Mon Oct 26 18:18:27 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 26 17:18:27 2026 +0000"
      },
      "message": "Update logback-classic to 1.6.6 (#934)"
    },
    {
      "commit": "c61dcd25cd262e327a5bc34f96e9341ed2e0913a",
      "tree": "379e987d8b90b39fde146734b56979fac64fb55c",
      "parents": [
        "6c8dc2baf9d9c689c92e8d907e8af9489134e856"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Oct 20 21:55:57 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Oct 20 20:55:57 2026 +0100"
      },
      "message": "Update grpc-core, grpc-interop-testing, ... to 1.85.1 (#919)\n\n* Update grpc-core, grpc-interop-testing, ... to 1.85.1\n\n* Update grpc-java version to 1.85.1 in remaining version-synced files\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e\nCo-authored-by: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "6c8dc2baf9d9c689c92e8d907e8af9489134e856",
      "tree": "61a07dc0380292020e7bc3c46ef5a82306bfacdd",
      "parents": [
        "e9a1769e5a21885ba6c9558ffea271963237f803"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Oct 20 17:19:26 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Oct 20 17:19:26 2026 +0100"
      },
      "message": "Encode response data frames through GrpcProtocolWriter with deferred writes (#909)\n\n* Generalise unary response optimisation.\n\nEncapsulate optimisation of unary responses into a GrpcProtocol/GrpcProtocolWriter concern, allowing it to be applied to other variants.\nThis removes the need for the frame encoder, as this is just a map with encodeFrame.\n\n* Keep the frame serializer types internal.\n\nMotivation:\nThe deferred data frame change made ProtobufFrameSerializer a public\n@ApiMayChange trait, which would expose the internal DeferredDataWriter\nto anyone extending the public serializers.\n\nModification:\n- Keep ProtobufFrameSerializer private[grpc], as it is on main.\n- Make DeferredDataWriter\u0027s methods, and the ScalaPB and Google\n  serializers\u0027 implementations of them, private[grpc].\n- Add the MiMa filters for the Scala 3 positional accessors of\n  GrpcProtocolReader and GrpcProtocolWriter, and rename the filter file\n  after the change.\n- Remove debug println output from the serializer specs.\n\nResult:\nNo new public API. Binary compatibility passes on Scala 2.13 and 3.\n\nTests:\n- sbt runtime/test: 180 passed\n- sbt +runtime/mimaReportBinaryIssues: passed\n\nReferences:\nRefs #905\n\n* Add a benchmark for encoding response data frames.\n\nMotivation:\nThe deferred data frame change affects how response messages are\nwritten into gRPC frames, but no benchmark covered gRPC-Web responses.\n\nModification:\nAdd FrameWriteBenchmark, which marshals unary and streamed responses\nfor the native and gRPC-Web protocols without compression, with 16 B,\n1 KiB and 64 KiB bytes payloads.\n\nResult:\nFrame encoding can be compared with main.\n\nTests:\n- sbt benchmarks/Jmh/compile: passed\n\nReferences:\nRefs #905\n\n* Fix FrameWriteBenchmark formatting and Scala 3.9 warnings.\n\nReplace the deprecated \u003d _ initializers, which fail the next build under\n-Werror, and apply scalafmt.\n\n* Use the pekko import prefix for DeferredDataFrame imports.\n\n* Fix the frane parameter name typo in GrpcProtocolWeb.\n\n* Use OutboundFrame in GrpcProtocolWeb encodeFrame and encodeFrameToBytes, matching GrpcProtocolNative.\n\n---------\n\nCo-authored-by: Tim Whittington \u003ctim.whittington@indexity.io\u003e\nCo-authored-by: PJ Fanning \u003cfanningpj@gmail.com\u003e"
    },
    {
      "commit": "e9a1769e5a21885ba6c9558ffea271963237f803",
      "tree": "ea328ce9d198616ac9af793e7550438574b415d0",
      "parents": [
        "1bdf0e5b32174412da4beb8587c8f4e9b60a9de8"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Oct 19 16:25:23 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 19 16:25:23 2026 +0100"
      },
      "message": "build(deps): bump coursier/cache-action from 8.1.1 to 8.2.0 (#918)\n\nBumps [coursier/cache-action](https://github.com/coursier/cache-action) from 8.1.1 to 8.2.0.\n- [Release notes](https://github.com/coursier/cache-action/releases)\n- [Commits](https://github.com/coursier/cache-action/compare/95e5b1029b6b86e7bac033ee44a0697d8a527d2d...057d08c2f9a2c4e7a4d33df3aae657b7db88b8cc)\n\n---\nupdated-dependencies:\n- dependency-name: coursier/cache-action\n  dependency-version: 8.2.0\n  dependency-type: direct:production\n  update-type: version-update:semver-minor\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "1bdf0e5b32174412da4beb8587c8f4e9b60a9de8",
      "tree": "f394559334ca1514103e3b6fe36874c61bf6caf3",
      "parents": [
        "4eac6f8543a875334591e4b1475099c62ee4e580"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Oct 19 16:25:15 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 19 16:25:15 2026 +0100"
      },
      "message": "build(deps): bump actions/setup-java from 6.0.1 to 6.1.0 (#917)\n\nBumps [actions/setup-java](https://github.com/actions/setup-java) from 6.0.1 to 6.1.0.\n- [Release notes](https://github.com/actions/setup-java/releases)\n- [Commits](https://github.com/actions/setup-java/compare/de7274f081f381c8f8158605e0321c36c376e2e6...b843ebf9b913146b38da2eaa7aeddf76074ae1ad)\n\n---\nupdated-dependencies:\n- dependency-name: actions/setup-java\n  dependency-version: 6.1.0\n  dependency-type: direct:production\n  update-type: version-update:semver-minor\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "4eac6f8543a875334591e4b1475099c62ee4e580",
      "tree": "5d537c9f3a158ba1cad026eb6c22f752cf6f2f5a",
      "parents": [
        "2b2e2d80ac4f6e29d66298464ca4365084210d65"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Mon Oct 19 09:50:45 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 19 08:50:45 2026 +0100"
      },
      "message": "Update grpc-core, grpc-interop-testing, ... to 1.84.3 (#914)\n\n* Update grpc-core, grpc-interop-testing, ... to 1.84.3\n\n* Update remaining gRPC 1.84.2 references to 1.84.3.\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "2b2e2d80ac4f6e29d66298464ca4365084210d65",
      "tree": "6f6747aba4685c609926ec673c93414a2da63fdd",
      "parents": [
        "a670bbcec5b3f00d6ec01be1d34edeae0bba8dbd"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Oct 19 08:47:34 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 19 08:47:34 2026 +0100"
      },
      "message": "build(deps): bump sbt/setup-sbt from 1.5.8 to 1.5.12 (#916)\n\nBumps [sbt/setup-sbt](https://github.com/sbt/setup-sbt) from 1.5.8 to 1.5.12.\n- [Release notes](https://github.com/sbt/setup-sbt/releases)\n- [Commits](https://github.com/sbt/setup-sbt/compare/c7d2d6258b4bd0d3ec5129e6b3453199d3c79729...ecad189dd245444b9ae0a0cd5922fd938b88906f)\n\n---\nupdated-dependencies:\n- dependency-name: sbt/setup-sbt\n  dependency-version: 1.5.12\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "a670bbcec5b3f00d6ec01be1d34edeae0bba8dbd",
      "tree": "0f0e184a879d03f3e2c88dd9f5472e58d993f22f",
      "parents": [
        "233b16b85fc8a0dff3da07004f2f79d36903e7c2"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Oct 13 10:34:26 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Oct 13 10:34:26 2026 +0100"
      },
      "message": "Avoid buffer copies with a non-synchronized ByteStringOutputStream (#907)\n\n* Avoid buffer copy on Gzip compress/uncompress.\n\n* Use the non-synchronized ByteStringOutputStream from Pekko HTTP.\n\nCopied from apache/pekko-http http-core ByteStringOutputStream, which hands the\nbuffered data over as a ByteString without copying and avoids the synchronized\nwrites of ByteArrayOutputStream.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n\n* Avoid private[this], which Scala 3.9 warns about.\n\nCo-Authored-By: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\n\n* Add a benchmark for gzip and InputStream buffering.\n\nMotivation:\nThe buffers used by Gzip.compress, Gzip.uncompress and\nByteStringUtils.fromInputStream had no benchmark, so changes to them\ncould not be measured.\n\nModification:\nAdd GzipBenchmark, covering compress, uncompress and fromInputStream\nfor 1 KiB, 64 KiB and 1 MiB payloads.\n\nResult:\nThe ByteStringOutputStream change can be compared with main.\n\nTests:\n- sbt benchmarks/Jmh/compile: passed\n- sbt \"benchmarks/Jmh/run -prof gc org.apache.pekko.grpc.GzipBenchmark\": run on main and this branch\n\nReferences:\nRefs #905\n\n* Avoid deprecated \u003d _ initializers in GzipBenchmark.\n\nScala 3.9 deprecates them, which fails the next build under -Werror.\n\n---------\n\nCo-authored-by: Tim Whittington \u003ctim.whittington@indexity.io\u003e\nCo-authored-by: Claude Opus 5.5 \u003cnoreply@anthropic.com\u003e\nCo-authored-by: PJ Fanning \u003cfanningpj@gmail.com\u003e"
    },
    {
      "commit": "233b16b85fc8a0dff3da07004f2f79d36903e7c2",
      "tree": "a0c7960a62a88a2f3461177d12d825b9a9ffbee6",
      "parents": [
        "891cd1b4a5c09b4f3ab99f70ee04c44a0992ce78"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Sat Oct 10 17:13:09 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Oct 10 16:13:09 2026 +0100"
      },
      "message": "Update grpc-core, grpc-interop-testing, ... to 1.84.2 (#913)\n\n* Update grpc-core, grpc-interop-testing, ... to 1.84.2\n\n* Update remaining grpc-java references to 1.84.2\n\nMotivation:\nThe Scala Steward update left 1.84.1 in the Gradle plugin, its unit test\nand the Maven plugin-tester poms.\n\nModification:\nReplace 1.84.1 with 1.84.2 in the remaining files tracked by\nVersionSyncCheckPlugin.\n\nResult:\nAll gRPC version references agree on 1.84.2.\n\nTests:\n- sbt grpcVersionSyncCheck: success\n\nReferences:\nRefs #913\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "891cd1b4a5c09b4f3ab99f70ee04c44a0992ce78",
      "tree": "d1140b109f9472995289165246e4bcfbf4049f1b",
      "parents": [
        "ae69587a68ca3cf0b8b1adca5fcc40a1a2771fbe"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Fri Oct 09 11:23:54 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Oct 09 11:23:54 2026 +0100"
      },
      "message": "Update PekkoHttpDependency version to 2.0.0-M2 (#906)"
    },
    {
      "commit": "ae69587a68ca3cf0b8b1adca5fcc40a1a2771fbe",
      "tree": "acaf0416bf2d394cf910483af612567c2f88f120",
      "parents": [
        "5d1e773a481a0bedb30a11359180d6b6064518b9"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Thu Oct 08 20:57:52 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Oct 08 19:57:52 2026 +0100"
      },
      "message": "Update grpc-core, grpc-interop-testing, ... to 1.84.1 (#911)\n\n* Update grpc-core, grpc-interop-testing, ... to 1.84.1\n\n* Update remaining grpc-java references to 1.84.1\n\nMotivation:\nThe Scala Steward update left 1.84.0 in the Gradle plugin, its unit test\nand the Maven plugin-tester poms.\n\nModification:\nReplace 1.84.0 with 1.84.1 in the remaining files tracked by\nVersionSyncCheckPlugin.\n\nResult:\nAll gRPC version references agree on 1.84.1.\n\nTests:\n- sbt grpcVersionSyncCheck: success\n\nReferences:\nRefs #911\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "5d1e773a481a0bedb30a11359180d6b6064518b9",
      "tree": "faa4a958ed6e5854a0a4759f65959defdbdb94b7",
      "parents": [
        "5ca512a3f890ae3c8eb4c4a1aba806b2f2318936"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Oct 07 12:12:04 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Oct 07 13:12:04 2026 +0200"
      },
      "message": "Run the MiMa check for every library Scala version. (#908)\n\nMotivation:\nThe CI binary-compatibility step runs mimaReportBinaryIssues, which\nonly checks the default Scala version (2.13). Scala 3 case classes have\npositional accessors (_1, _2, ...) that Scala 2.13 ones don\u0027t, so\nchanging a case class can break Scala 3 binary compatibility and still\npass CI. The documented `sbt +mimaReportBinaryIssues` command doesn\u0027t\nwork either: the Scala 2.12 plugin projects don\u0027t define the task.\n\nModification:\nRun `+runtime/mimaReportBinaryIssues` in CI, and give that command in\nAGENTS.md, CLAUDE.md and CONTRIBUTING.md. runtime is the only project\nwith MiMa enabled.\n\nResult:\nCI checks binary compatibility on Scala 2.13 and Scala 3.\n\nTests:\n- sbt grpcVersionSyncCheck googleProtobufVersionSyncCheck +runtime/mimaReportBinaryIssues on main: passed (2.13.18 and 3.3.8)\n- sbt +runtime/mimaReportBinaryIssues with commit ee37b1fb from #905 applied to main: fails with 4 Scala 3 problems that the current CI check misses\n- sbt +mimaReportBinaryIssues on main: fails with \"No such setting/task\" for the Scala 2.12 projects\n\nReferences:\nRefs #905\n\nCo-authored-by: PJ Fanning \u003cfanningpj@gmail.com\u003e"
    },
    {
      "commit": "5ca512a3f890ae3c8eb4c4a1aba806b2f2318936",
      "tree": "cd6ba4f23ffda49687301fa2f8e468771739647d",
      "parents": [
        "8159c66bf04bf54677b239f35d6dc13d11479435"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Oct 07 10:19:07 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Oct 07 11:19:07 2026 +0200"
      },
      "message": "Use the pekko-http Trailer header instead of a private copy. (#910)\n\nMotivation:\npekko-grpc kept its own private `Trailer` custom header to pre-announce\nthe `grpc-status` trailer. That header was moved upstream into\npekko-http (apache/pekko-http#684, since 1.3.0), and pekko-grpc main\nalready requires pekko-http 2.0.0-M1, so the private copy is redundant.\npekko-http also parses incoming `trailer` headers into its modeled\ntype, so two classes existed for the same header name.\n\nModification:\nRemove the private `Trailer` header from `scaladsl.headers` and use\n`pekko.http.scaladsl.model.headers.Trailer` in `GrpcResponseHelpers`.\nDrop the now-unused `scala.collection.compat.immutable.ArraySeq`\nimport, which was never needed for the 2.13/3-only runtime module.\n\nResult:\nServer streaming responses pre-announce trailers with pekko-http\u0027s\nmodeled and validated `Trailer` header.\n\nTests:\n- sbt \"runtime/testOnly org.apache.pekko.grpc.internal.GrpcResponseHelpersSpec\" - passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt \"runtime/scalafmtAll\" \"runtime/Test/scalafmtAll\" - no changes\n\nReferences:\nFixes #470"
    },
    {
      "commit": "8159c66bf04bf54677b239f35d6dc13d11479435",
      "tree": "1cfdda8622bcc1f40d97159b52bd7bf9c78c131f",
      "parents": [
        "bb29e6ba8bbff457a40f0d5a875bc56c11843ef5"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Oct 05 10:15:44 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 05 10:15:44 2026 +0100"
      },
      "message": "build(deps): bump coursier/setup-action from 3.0.3 to 3.0.4 (#904)\n\nBumps [coursier/setup-action](https://github.com/coursier/setup-action) from 3.0.3 to 3.0.4.\n- [Release notes](https://github.com/coursier/setup-action/releases)\n- [Commits](https://github.com/coursier/setup-action/compare/3174df5c784a0c8724f06b7924b849551c40c39e...648df969f41ef15fda2baba8b37f9fa3d16390a3)\n\n---\nupdated-dependencies:\n- dependency-name: coursier/setup-action\n  dependency-version: 3.0.4\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "bb29e6ba8bbff457a40f0d5a875bc56c11843ef5",
      "tree": "4dfcfc352a77fa6dd137b0796430aec47c681bf7",
      "parents": [
        "ff247d3f3e846f67054990b6444d06c2d9ae3ccc"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Oct 02 18:13:29 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Oct 02 17:13:29 2026 +0100"
      },
      "message": "Update maven-core, maven-plugin-api to 3.10.0 (#903)"
    },
    {
      "commit": "ff247d3f3e846f67054990b6444d06c2d9ae3ccc",
      "tree": "ae0980177851d70a13be3d23644e87d5e2970112",
      "parents": [
        "62226b380a2d42d8d325aad7e808c5b64b57651c"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Oct 02 18:13:22 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Oct 02 17:13:22 2026 +0100"
      },
      "message": "Update sbt-paradox, sbt-paradox-theme to 0.11.1 (#902)"
    },
    {
      "commit": "62226b380a2d42d8d325aad7e808c5b64b57651c",
      "tree": "e719538ee88e54e9ff44414bf5256c5060411d64",
      "parents": [
        "76adc6282be9eef8d2f69b5d8cf6e1695b3a28fd"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Oct 02 18:13:16 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Oct 02 17:13:16 2026 +0100"
      },
      "message": "Update sbt-java-formatter to 0.14.0 (#901)"
    },
    {
      "commit": "76adc6282be9eef8d2f69b5d8cf6e1695b3a28fd",
      "tree": "70a47a8ea6e3972a74a6c011e3f0383c5f2273f8",
      "parents": [
        "9aeccbac4b91afe46382f5e10f542c43aedb37b1"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Oct 02 18:13:08 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Oct 02 17:13:08 2026 +0100"
      },
      "message": "Update logback-classic to 1.6.5 (#900)"
    },
    {
      "commit": "9aeccbac4b91afe46382f5e10f542c43aedb37b1",
      "tree": "cc1286361cfe841bc8e5bdbecf9983c2d9c7cd11",
      "parents": [
        "53169ba42825011bb34a36a019e419899289aa9a"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Oct 01 15:05:46 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Oct 01 15:05:46 2026 +0100"
      },
      "message": "Remove leftover Scala 2.12 collection idioms (#886)\n\nMotivation:\nThe 2.13/3-only modules still carried Scala 2.12-era idioms:\n`immutable.Seq` qualifiers (`scala.Seq` has been `immutable.Seq`\nsince 2.13) and `.toIterator` in the scripted interop test.\nEquivalent of apache/pekko#3539.\n\nModification:\n- Replace `immutable.Seq` with `Seq` in runtime, interop-tests,\n  benchmarks and the `gen-scala-server/00-interop` scripted test\n  (all built for 2.13/3 only) and drop the now-unused\n  `scala.collection.immutable` imports.\n- `requests.toIterator` -\u003e `requests.iterator` in\n  `PekkoGrpcClientTester` (scripted test).\n- `codegen`, `maven-plugin`, `sbt-plugin` main sources and\n  `scalapb-protoc-plugin` still cross-build for 2.12 where\n  `scala.Seq` is `collection.Seq`, so they are left unchanged.\n\nResult:\nNo remaining Scala 2.12 collection idioms in the 2.13/3-only modules.\nSource and binary compatible: `scala.Seq` is a type alias for\n`scala.collection.immutable.Seq` on 2.13 and 3, so the erasure and\ngeneric signatures are unchanged.\n\nTests:\n- native scalafmt on all changed files\n- sbt runtime/mimaReportBinaryIssues on 2.13.18 and 3.3.8: passes\n- sbt runtime/test on 2.13.18: 178 passed\n- sbt runtime/Test/compile interop-tests/Test/compile benchmarks/compile\n  on 2.13.18 and 3.3.8\n- sbt \"sbt-plugin/scripted gen-scala-server/00-interop\": passes\n\nReferences:\nNone - equivalent of apache/pekko#3539"
    },
    {
      "commit": "53169ba42825011bb34a36a019e419899289aa9a",
      "tree": "ed9064b229797520e4ca0b5b1325d82eacf46464",
      "parents": [
        "aad2f3f4c8a5ae891684a0162d4f1a5efc68bbe1"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Sep 29 11:18:04 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 29 10:18:04 2026 +0100"
      },
      "message": "Update logback-classic to 1.6.4 (#898)"
    },
    {
      "commit": "aad2f3f4c8a5ae891684a0162d4f1a5efc68bbe1",
      "tree": "d32ef60864d32958dfe45a7968bcc015decb5fcc",
      "parents": [
        "33e309367c7608ea5e0d67c757e431bbefec83ee"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Thu Sep 24 17:29:14 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Sep 24 16:29:14 2026 +0100"
      },
      "message": "Update slf4j-api to 2.0.20 (#897)"
    },
    {
      "commit": "33e309367c7608ea5e0d67c757e431bbefec83ee",
      "tree": "31cbc4a4cbd3653952b92ed6d433cfaf214c76a2",
      "parents": [
        "21ec5c8c5244015e1d1f5a0cb7953e116d0f0b8f"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Thu Sep 24 11:21:12 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Sep 24 10:21:12 2026 +0100"
      },
      "message": "Update sbt-mima-plugin to 1.2.1 (#894)"
    },
    {
      "commit": "21ec5c8c5244015e1d1f5a0cb7953e116d0f0b8f",
      "tree": "251d52b0b08f1a497a83067a636f8e110885bffa",
      "parents": [
        "32b428627b8884b1c7c6036591a1ae972619cf07"
      ],
      "author": {
        "name": "Matthias Kurz",
        "email": "m.kurz@irregular.at",
        "time": "Wed Sep 23 16:50:04 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Sep 23 15:50:04 2026 +0100"
      },
      "message": "fix: Avoid duplicate gRPC resource directories in sbt builds (#895)\n\n* test: Detect duplicate gRPC resource directories in sbt 2\n\nMotivation:\nThe gRPC plugin duplicates resource directories, which can cause downstream packaging to fail.\n\nModification:\nExtend the existing sbt 2 scripted fixture to assert unique Compile and Test resource directories and package its JAR.\n\nResult:\nThe assertion fails against the unfixed plugin with each resource directory listed twice.\n\nTests:\n- sbt --batch \u0027++3.8.4\u0027 \u0027sbt-plugin / scripted scala3/01-basic-client-server\u0027 (expected failure: duplicate directories)\n- scalafmt --test sbt-plugin/src/sbt-test/scala3/01-basic-client-server/build.sbt (passed)\n\nReferences:\nNone - reproduces the Play gRPC resource-packaging regression.\n\n* fix: Avoid duplicate gRPC resource directories in sbt builds\n\nMotivation:\nThe plugin appends sbt-protoc resource paths even when sbt already lists them. Downstream packaging can then encounter duplicate resources.\n\nModification:\nDeduplicate the combined resource directories in the existing Compile and Test plugin settings while retaining custom sbt-protoc paths.\n\nResult:\nBoth configurations expose each resource directory once, without changing the set of directories.\n\nTests:\n- sbt scripted scala3/01-basic-client-server on sbt 2.0.0: passed.\n- sbt scripted gen-scala-server/01-gen-basic-server on sbt 1.12.15: passed.\n- sbt 2.1.0-M2 checkResourceDirectories and Compile/packageBin: passed.\n- sbt checkCodeStyle: passed.\n- sbt validatePullRequest: unavailable; this build does not define the command.\n- A combined sbt 2.1.0-M2 test/check/package run passed test and check, then exited with a Pekko classloader LinkageError; check/package passed separately.\n\nReferences:\nNone - fixes the Play gRPC duplicate-resource packaging regression."
    },
    {
      "commit": "32b428627b8884b1c7c6036591a1ae972619cf07",
      "tree": "8f05cb0489967e2ee249ef477b7c69dd1dddcadb",
      "parents": [
        "fd44957f13006e772204d842da7a7f4c35a98bea"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Sep 23 11:20:19 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Sep 23 11:20:19 2026 +0100"
      },
      "message": "fix: publish apiURL as scaladoc root (#891)\n\nMotivation:\nThe published apiURL pointed at the org/apache/pekko/grpc/index.html\npage rather than the scaladoc root. Downstream builds that use\nautoAPIMappings (e.g. pekko-projection) pass this to scaladoc as the\nexternal doc base, and scaladoc appends the fully-qualified class path,\nproducing 404 links such as\nhttps://pekko.apache.org/api/pekko-grpc/2.0.0-M2/org/apache/pekko/grpc/org/apache/pekko/grpc/GrpcChannel.html\n\nModification:\nSet apiURL to https://pekko.apache.org/api/pekko-grpc/\u003cversion\u003e/ so\nthat appending org/apache/pekko/grpc/\u003cClass\u003e.html resolves correctly.\n\nResult:\nDownstream scaladoc links to pekko-grpc classes resolve.\n\nTests:\n- sbt runtime/makePom -\u003e info.apiURL is https://pekko.apache.org/api/pekko-grpc/snapshot/\n- sbt scalafmtSbtCheck passes\n\nReferences:\nNone - found via pekko-projection link-validator failure\nhttps://github.com/apache/pekko-projection/actions/runs/35568694883/job/106235549655"
    },
    {
      "commit": "fd44957f13006e772204d842da7a7f4c35a98bea",
      "tree": "86d73145d608f61c17bcd329e286d0b3681316ed",
      "parents": [
        "ce74e511a1b6c54c0252b89113d9a887754e9c1d"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Sep 22 17:24:03 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 22 16:24:03 2026 +0100"
      },
      "message": "Update protobuf-java to 4.36.2 (#893)\n\n* Update protobuf-java to 4.36.2\n\n* more changes\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "ce74e511a1b6c54c0252b89113d9a887754e9c1d",
      "tree": "366e18e71c50024917890dcf86ca0a81926f43a6",
      "parents": [
        "c2145425ed1f9c39f963bc375c34242460afdcd6"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Sep 22 16:22:33 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 22 15:22:33 2026 +0100"
      },
      "message": "Update sbt-buildinfo to 0.13.2 (#892)"
    },
    {
      "commit": "c2145425ed1f9c39f963bc375c34242460afdcd6",
      "tree": "643d4aadf21ce6272ddd6493dce4c530d135bb22",
      "parents": [
        "91e4e7724c396e8e6c621bb5099107b5ca735ee7"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Mon Sep 21 12:23:18 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Sep 21 12:23:18 2026 +0100"
      },
      "message": "Remove note about server reflection (#867)"
    },
    {
      "commit": "91e4e7724c396e8e6c621bb5099107b5ca735ee7",
      "tree": "d5418a4cf388a877784bf5e3dcc3da381ef84207",
      "parents": [
        "41fe7727fa65dc4202800ea94ec6c12e31094f39"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Sep 21 08:26:03 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Sep 21 08:26:03 2026 +0100"
      },
      "message": "build(deps): bump coursier/setup-action from 3.0.2 to 3.0.3 (#889)\n\nBumps [coursier/setup-action](https://github.com/coursier/setup-action) from 3.0.2 to 3.0.3.\n- [Release notes](https://github.com/coursier/setup-action/releases)\n- [Commits](https://github.com/coursier/setup-action/compare/9b7939bf01fd1185ce2babe16135168361bf2c62...3174df5c784a0c8724f06b7924b849551c40c39e)\n\n---\nupdated-dependencies:\n- dependency-name: coursier/setup-action\n  dependency-version: 3.0.3\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "41fe7727fa65dc4202800ea94ec6c12e31094f39",
      "tree": "265d6aab7fefd6f734d509624cef5879c0906247",
      "parents": [
        "7f8b0d846c1b8b103f15320009ea36e9ec6cf659"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Sep 14 08:13:29 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Sep 14 08:13:29 2026 +0100"
      },
      "message": "build(deps): bump actions/setup-java from 6.0.0 to 6.0.1 (#887)\n\nBumps [actions/setup-java](https://github.com/actions/setup-java) from 6.0.0 to 6.0.1.\n- [Release notes](https://github.com/actions/setup-java/releases)\n- [Commits](https://github.com/actions/setup-java/compare/dd06d9cba3e5552c54d9f8ea23572deb30010f7c...de7274f081f381c8f8158605e0321c36c376e2e6)\n\n---\nupdated-dependencies:\n- dependency-name: actions/setup-java\n  dependency-version: 6.0.1\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "7f8b0d846c1b8b103f15320009ea36e9ec6cf659",
      "tree": "09fbd7736e1f1d89505e8041f1a3893a1e7c6109",
      "parents": [
        "707e5b6a4c466e763cf89338a213b7307fae5534"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Sep 08 21:14:27 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 08 20:14:27 2026 +0100"
      },
      "message": "Update sbt-mima-plugin to 1.2.0 (#884)"
    },
    {
      "commit": "707e5b6a4c466e763cf89338a213b7307fae5534",
      "tree": "150268e5147b816b9de5094886b3991976c43bc4",
      "parents": [
        "922fbb536b8cb3a8dcbbb1af5f1cad6786b595d2"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Sep 08 21:08:16 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 08 20:08:16 2026 +0100"
      },
      "message": "Update org.eclipse.jgit to 7.8.0.202609011348-r (#885)"
    },
    {
      "commit": "922fbb536b8cb3a8dcbbb1af5f1cad6786b595d2",
      "tree": "001d6f7a78b18d037d12c6398ac75aa9c6da8d7d",
      "parents": [
        "c9a15638a685bfdf431b0e3de65cb93b27b108fa"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Sep 04 21:34:50 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Sep 04 20:34:50 2026 +0100"
      },
      "message": "Update scala-library to 3.9.0 (#881)\n\n* Update scala-library to 3.9.0\n\n* fix\n\n* Drop stray $ from PekkoHttpServerProviderJava test object name\n\nMotivation:\nScala 3.9.0 added syntax warning E230, which flags identifiers containing\n`$` as reserved for internal compiler use. The interop-tests test sources\nare compiled with `-Werror`, so `object PekkoHttpServerProviderJava$` turns\nthat warning into a build failure.\n\nThe trailing `$` is not intentional. It is a leftover from the Akka to\nPekko rename in 7ad0b5d6; the equivalent object in the sbt scripted test\ncopy of this spec is named `PekkoHttpServerProviderJava` with no `$`.\n\nModification:\nRename the object to `PekkoHttpServerProviderJava` and update its single\nreference in `Servers.Pekko.Java`.\n\nResult:\ninterop-tests test sources compile under Scala 3.9.0 with `-Werror`.\n\nTests:\n- sbt -batch \"++3.9.0!\" Test/compile - success\n\nReferences:\nRefs #883\n\n* Keep the sbt 2.x plugin on the Scala version sbt itself is built with\n\nMotivation:\nBumping scala3Next to 3.9.0 broke the `sbt scripted tests (scala3, next)`\njob. scala3Next was doing two unrelated jobs at once: it is the newest\nScala 3 release the library is compiled and tested against, and it is also\nthe Scala version the sbt 2.x plugin is cross-built with.\n\nThe second is not ours to choose. sbt uses its own Scala compiler to\ncompile build.sbt, and every sbt 2.x release to date, up to and including\n2.0.8, depends on scala3-library_3 3.8.4. A plugin built with 3.9.0 emits\nTASTy 28.9, which that compiler cannot read, so loading any build that\nuses the plugin fails:\n\n  value autoImport is not a member of object\n  org.apache.pekko.grpc.sbt.PekkoGrpcPlugin\n    error while loading IArray$package,\n    TASTy file scala/IArray$package.tasty could not be read, failing with:\n      Forward incompatible TASTy file has version 28.9, produced by\n      Scala 3.9.0-bin-nonbootstrapped, expected stable TASTy from\n      28.0 to 28.8.\n\nModification:\nSplit the two roles apart. Add Versions.scala3ForSbtPlugin, pinned to the\nScala version sbt 2.x is built with, and use it in CrossScalaForSbtPlugin.\nscala3Next stays free to track the newest Scala 3 release.\n\nAdd a `sbt-plugin-scala3` case to resolve-scala-version.sh and point the\n`scala3` scripted matrix entry at it, so the scripted job builds the plugin\nwith a version sbt can load.\n\nThe scripted test projects still receive scala3Next through\n-Dpekko.grpc.scala3.next.version, so their generated sources are still\ncompiled with the newest Scala 3 - which is the part of that job worth\ntesting. Publishing is unaffected: `+publish` picks up the cross version\nfrom CrossScalaForSbtPlugin.\n\nResult:\nThe sbt plugin is loadable by sbt 2.x again, and scala3Next can be bumped\nindependently of sbt\u0027s release cadence.\n\nTests:\n- sbt -batch \"++3.9.0!\" Test/compile - success\n- bash .github/scripts/resolve-scala-version.sh sbt-plugin-scala3 - 3.8.4\n- bash .github/scripts/resolve-scala-version.sh next - 3.9.0\n- sbt \"sbt-plugin/scripted scala3/*\" - not run locally, relying on CI\n\nReferences:\nRefs #881\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "c9a15638a685bfdf431b0e3de65cb93b27b108fa",
      "tree": "dfd6572bac2352bf73424db95fe954865690d64f",
      "parents": [
        "022414594595ee3e71476bd0b522ea2bf108bb47"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Sep 04 19:25:46 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Sep 04 18:25:46 2026 +0100"
      },
      "message": "Update slf4j-api to 2.0.19 (#882)"
    },
    {
      "commit": "022414594595ee3e71476bd0b522ea2bf108bb47",
      "tree": "ba1ea2035f79f522738b4ea8d50a790aacaffb8e",
      "parents": [
        "9167279035f401016d2f3dbb61f234772c58addf"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Wed Sep 02 18:29:57 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Sep 02 17:29:57 2026 +0100"
      },
      "message": "Update protobuf-java to 4.36.1 (#879)\n\n* Update protobuf-java to 4.36.1\n\n* more refs\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "9167279035f401016d2f3dbb61f234772c58addf",
      "tree": "937c84e9693a3bdc94166ac924f7bb63385e3f8c",
      "parents": [
        "02a70d2f064e808c2bea4916da72d6f585a661b9"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Wed Sep 02 18:29:20 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Sep 02 17:29:20 2026 +0100"
      },
      "message": "Update grpc-core, grpc-interop-testing, ... to 1.84.0 (#880)\n\n* Update grpc-core, grpc-interop-testing, ... to 1.84.0\n\n* more refs\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "02a70d2f064e808c2bea4916da72d6f585a661b9",
      "tree": "3d547ca2ca3a1cc5b55046e9fc6c0123fbe4d556",
      "parents": [
        "2629e0af13344bd82bee60114d4c0b0de8d0b5ce"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Sep 01 11:21:24 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 01 11:21:24 2026 +0100"
      },
      "message": "Report malformed response headers as a gRPC status (#870)\n\nMotivation:\nTwo spots in the pekko-http client turned a malformed response from the peer\ninto a raw JVM exception rather than a `StatusRuntimeException`, so a caller\nmatching on gRPC status saw something it could not handle.\n\n`Status.fromCodeValue(statusCode.toInt)` threw `NumberFormatException` for a\nserver sending `grpc-status: abc`. Building the response metadata\nbase64-decodes every `-bin` header, so a malformed one threw\n`IllegalArgumentException` - and `mapToStatusException` builds that metadata for\nevery error response, so an invalid binary header replaced the error actually\nbeing reported with an exception about itself.\n\nBoth are reachable from any peer. The netty backend rejects these at the\ntransport layer, so this was pekko-http only.\n\nModification:\n- A non-numeric `grpc-status` is reported as `INTERNAL`, quoting the offending\n  value. `parseResponseStatus` already matches on `Some(\"0\")`, so such a\n  response was on the failure path regardless; it now fails with a status.\n- Metadata construction falls back to empty metadata if a header cannot be\n  represented, so the status and message still reach the caller.\n\nResult:\nA broken or hostile peer produces a `StatusRuntimeException` like any other\nfailure, instead of an exception the caller is not expecting.\n\nTests:\n- 2 cases in `PekkoHttpClientUtilsSpec`: a non-numeric `grpc-status`, and a\n  malformed `-bin` header alongside a real error that must still be reported\n- Confirmed both guards bite: restoring `toInt` fails the first, removing the\n  metadata fallback fails the second\n- sbt \"runtime/testOnly ...PekkoHttpClientUtilsSpec\" - 8 passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt scalafmtAll scalafmtSbt - applied\n- sbt \"runtime/test\" - not run locally, left to CI\n\nNote: `HeaderMetadataImpl.getBinary` and `asList` still throw on a malformed\nbinary header when a user calls them directly on a successful response. That is\nleft as is - an explicit accessor call is a different contract from the error\npath, where the exception displaces an unrelated failure.\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "2629e0af13344bd82bee60114d4c0b0de8d0b5ce",
      "tree": "dba89defab2bbcb6bcbdefff57d1749751df5ad1",
      "parents": [
        "5d0a3446ad77639ea050a4828b1ae4abfad4a880"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Sep 01 11:20:59 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 01 11:20:59 2026 +0100"
      },
      "message": "Use FastFuture for already completed marshalling futures (#863)\n\n* Use FastFuture for completed futures.\n\nUse FastFuture.apply and FastFuture.successful/failed for trivially completed futures to optimise subsequent transforms.\nThis speeds up a lot of service invocations in existing generated code that depend on these, especially GrpcMarshalling.unmarshal on strict request entities.\n\n* Benchmark and test FastFuture completion of marshalling futures\n\nMotivation:\nThe FastFuture change is only observable as a threading/scheduling\nproperty: the returned futures are already completed, so transforms\nchained onto them run directly instead of being scheduled on an\nExecutionContext. Nothing asserted or measured that.\n\nModification:\nAdd scaladsl GrpcMarshallingSpec, which asserts that unmarshal of a\nstrict entity (both success and failure), unmarshalStream,\nGrpcExceptionHandler.from and the ServiceHandler notFound /\nunsupportedMediaType responses are already completed, and that a\ntransform chained onto each still completes when given an\nExecutionContext that throws if anything is scheduled on it.\n\nAdd GrpcMarshallingBenchmark.unmarshallStrictChained, covering the\nunmarshal-then-transform chain that generated unary handlers use, and\nScalaUnaryHandlerBenchmark.oldStyleUnaryStrictRequestProcessingFastImplementation,\na variant whose service implementation returns an already completed\nFastFuture so the whole handler chain can run directly.\n\nResult:\nThe intent of the change is pinned by tests, and the benchmarks show\nwhere the saved dispatches actually come from.\n\nTests:\n- sbt runtime/test - 139 succeeded, 0 failed\n- sbt checkCodeStyle - success\n- sbt \"benchmarks/Jmh/run -f 2 -wi 5 -i 5 ...\" - see PR body\n\nReferences:\nRefs #862\n\n---------\n\nCo-authored-by: Tim Whittington \u003ctim.whittington@indexity.io\u003e"
    },
    {
      "commit": "5d0a3446ad77639ea050a4828b1ae4abfad4a880",
      "tree": "f2e4a3d3353bb4cdc14cd3d872a3ab44cbb63bbd",
      "parents": [
        "3dd4c68219394fe2f64da86702e309063ea092e1"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Sep 01 10:10:08 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 01 10:10:08 2026 +0100"
      },
      "message": "Replace the resolver\u0027s assert with an IllegalArgumentException (#873)\n\nMotivation:\nThe pekko-http transport resolver checked the host it was asked to resolve\nagainst the configured `override-authority` with `assert`. Two problems:\n\n`assert` throws `AssertionError`, which is an `Error` and not an `Exception`, so\nit passes straight through a handler that catches `Exception` and is not what a\ncaller inspecting a failed connection would expect to match on.\n\n`assert` is also elided entirely under `-Xdisable-assertions`. That is the\ncalling application\u0027s compiler flag, not this project\u0027s, so the check silently\ndisappears in a build this project does not control.\n\nThe message was also just \"assertion failed\", naming neither the host nor the\nauthority it was compared against.\n\nModification:\nExtracted `checkAuthority`, which throws an `IllegalArgumentException` naming\nboth hosts and the client it belongs to. Extracting it also gives the check a\nseam a test can reach; it previously lived inside a closure passed to\n`ClientTransport.withCustomResolver`.\n\nResult:\nThe check runs regardless of the caller\u0027s compiler flags, fails with an\nexception rather than an error, and says what mismatched.\n\nTests:\n- 3 cases in `PekkoHttpClientUtilsSpec`: a matching authority, no configured\n  authority, and a mismatch asserting the failure is an `Exception` and names\n  both hosts\n- Confirmed the guard bites: restoring `assert` fails the mismatch case, since\n  `AssertionError` is not an `Exception`\n- sbt \"runtime/testOnly ...PekkoHttpClientUtilsSpec\" - 12 passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt scalafmtAll scalafmtSbt - applied\n- sbt \"runtime/test\" - not run locally, left to CI\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "3dd4c68219394fe2f64da86702e309063ea092e1",
      "tree": "bc28e81e26174f4005118c6da0d06da5293b3c5e",
      "parents": [
        "1d2a2f6f00d39de1a1f71976d2d8f66628d81d74"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Tue Sep 01 10:07:48 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Sep 01 10:07:48 2026 +0100"
      },
      "message": "Do not return NotImplementedError messages to the caller (#874)\n\nMotivation:\n`defaultMapper` logged and scrubbed the message of every failure it did not\nrecognise, but two cases were different:\n\n    case e: NotImplementedError           \u003d\u003e Trailers(Status.UNIMPLEMENTED.withDescription(e.getMessage))\n    case e: UnsupportedOperationException \u003d\u003e Trailers(Status.UNIMPLEMENTED.withDescription(e.getMessage))\n\nThose messages come from whatever the service implementation threw, so an\nexception raised deep inside a handler put its text on the wire. Neither case was\nlogged either, so the detail went to the caller and nowhere else - the opposite\nof every other branch.\n\n`???` in Scala throws `NotImplementedError`, and `UnsupportedOperationException`\ncomes out of many collection and stream operations, so these are easy to hit\nwithout meaning to report anything.\n\nModification:\nBoth cases now log at warning and answer with a bare `UNIMPLEMENTED`, matching\nhow the mapper treats every other unrecognised failure, in both the scaladsl and\njavadsl handlers.\n\n`GrpcServiceException` remains the way to send a description deliberately, so the\ngenerated handlers now raise that for an unknown method rather than relying on\nthe message of a `NotImplementedError` reaching the caller. The behaviour a\nclient sees for an unknown method is unchanged.\n\nResult:\nAn accidental `???` or unsupported collection operation no longer describes\nitself to the caller, and the detail is logged where an operator can see it.\nDeliberate descriptions still work.\n\nTests:\n- 3 new cases in `GrpcExceptionHandlerSpec`: the message of each exception type\n  is withheld, and a deliberate `GrpcServiceException` description still arrives\n- The existing table-driven expectation is updated, which is the behaviour change\n- The existing end-to-end `ErrorReportingSpec` in both plugin-testers already\n  asserts a client calling an unknown method receives\n  `grpc-message: Not implemented: UnknownMethod`; both still pass, which is what\n  covers the codegen half of this\n- Confirmed the guard bites: restoring the leak fails 3 tests\n- sbt \"runtime/testOnly ...GrpcExceptionHandlerSpec\" - 29 passed\n- sbt \"plugin-tester-scala/testOnly ...ErrorReportingSpec\" - 2 passed\n- sbt \"plugin-tester-java/testOnly ...ErrorReportingSpec\" - 2 passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt scalafmtAll scalafmtSbt - applied\n- sbt \"runtime/test\" - not run locally, left to CI\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "1d2a2f6f00d39de1a1f71976d2d8f66628d81d74",
      "tree": "66067167e1607ec1dcc0c37b4592bb25f3e87a4d",
      "parents": [
        "66b15d92fec9b2b934acc45286cc3cdfac54384c"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Mon Aug 31 12:52:13 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 31 12:52:13 2026 +0100"
      },
      "message": "Ship a complete LICENSE and NOTICE in the assembly jars (#865)\n\n* Ship a complete LICENSE and NOTICE in the assembly jars\n\nMotivation:\npekko-grpc-codegen and pekko-grpc-scalapb-protoc-plugin publish assembly jars\nand bat archives that bundle classes from 3rd party projects, so their\nMETA-INF/LICENSE and META-INF/NOTICE have to account for those projects.\nlegal/AssemblyLicense.txt and legal/AssemblyNotice.txt exist for that, but:\n\n- The codegen assembly jar and bat shipped no META-INF/LICENSE at all. The\n  merge strategy discarded META-INF/LICENSE to stop guava/failureaccess\n  colliding, and that also discarded the copy the ApacheSonatypePlugin\n  resource generator puts in the jar.\n- legal/AssemblyLicense.txt had drifted from what is actually bundled. It\n  listed org.checkerframework:checker-qual, which guava dropped in favour of\n  org.jspecify:jspecify, and jspecify was not listed anywhere.\n- The scalapb-protoc-plugin assembly bundled protobuf-java 4.35.0, pulled in\n  transitively by compilerplugin, rather than the pinned 4.36.0 that codegen\n  and runtime bundle.\n\nNothing failed the build when the legal files fell behind a dependency change.\n\nModification:\nMerge META-INF/LICENSE and META-INF/NOTICE with MergeStrategy.preferProject in\nboth assemblies, so our own files win over the ones the dependencies bring in.\n\nSwap checker-qual for jspecify in legal/AssemblyLicense.txt and drop the MIT\ntext that was only there for checker-qual.\n\nPin protobuf-java in scalapb-protoc-plugin via a new Dependencies group, the\nsame way codegen and runtime already do.\n\nAdd project/AssemblyLicenseCheck.scala with two tasks, wired into both\nassembly projects and run in CI as the checkAssemblyLicenses alias:\n\n- assemblyLicenseCheck fails when a module on the assembly classpath is not\n  named in the assembly LICENSE file, and warns about listed modules that are\n  not bundled for the Scala version being built.\n- assemblyMetaInfCheck opens the published assembly jar and bat archive and\n  fails unless each carries META-INF/LICENSE and META-INF/NOTICE matching the\n  files in legal/.\n\nResult:\nBoth assembly jars and both bat archives ship a LICENSE and NOTICE covering\nthe 3rd party classes they bundle, both bundle protobuf-java 4.36.0, and CI\nfails if a dependency change leaves the legal files behind.\n\nTests:\n- sbt checkAssemblyLicenses - passed. assemblyMetaInfCheck fails on the\n  codegen assembly before the merge strategy fix (META-INF/LICENSE is\n  missing), and assemblyLicenseCheck fails before the AssemblyLicense.txt fix\n  (org.jspecify:jspecify is not accounted for).\n- sbt scalafmtSbtCheck - passed.\n\nReferences:\nNone - assembly jar license completeness\n\n* Update AssemblyLicenseCheck.scala\n\n* Name the archive when it cannot be read as a zip\n\nMotivation:\n`assemblyMetaInfCheck` opened each archive with `new ZipFile`, which throws\n`ZipException: error in opening zip file` without naming the file, so a corrupt\nor truncated archive in CI gave no clue which one it was.\n\nModification:\nOpen the archive and its entries through `sbt.io.Using`, and report `IOException`\nwith the archive name.\n\nResult:\nA failure to read an archive names it, like the license problems reported below it.\n\nTests:\n- sbt checkAssemblyLicenses - passed for both assembly projects (jar and bat)\n\nReferences:\nRefs #865\n\n* Assert the shipped LICENSE and NOTICE still say something\n\nMotivation:\n`assemblyMetaInfCheck` only compared the archived files with the curated ones in\n`legal/`, which says the merge strategy picked our copy but not that the copy is\nstill a real LICENSE or NOTICE. An emptied or truncated curated file would ship\nand pass.\n\nModification:\nCheck the shipped bytes for the markers each file has to carry: the Apache\nLicense 2.0 heading and terms plus the 3rd party section in LICENSE, and the\nproduct name, the ASF attribution and the 3rd party section in NOTICE. Missing\nmarkers are reported alongside the existing problems for that archive.\n\nResult:\nAn assembly that ships a gutted LICENSE or NOTICE fails the check.\n\nTests:\n- sbt checkAssemblyLicenses - passed for both assembly projects (jar and bat)\n- With legal/AssemblyLicense.txt cut down to its first two lines,\n  codegen/assemblyMetaInfCheck fails: \"META-INF/LICENSE does not mention\n  \u0027TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION\u0027,\n  \u0027contain classes from 3rd party projects\u0027, \u0027Apache License Version 2.0:\u0027\"\n\nReferences:\nRefs #865"
    },
    {
      "commit": "66b15d92fec9b2b934acc45286cc3cdfac54384c",
      "tree": "40a779a948f3e4be6045337d5682924936032e8c",
      "parents": [
        "6477f539792c1a96fe799fc9b3e7abd9990d3a55"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Aug 31 08:20:01 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 31 08:20:01 2026 +0200"
      },
      "message": "build(deps): bump actions/setup-java from 5.7.0 to 6.0.0 (#876)\n\nBumps [actions/setup-java](https://github.com/actions/setup-java) from 5.7.0 to 6.0.0.\n- [Release notes](https://github.com/actions/setup-java/releases)\n- [Commits](https://github.com/actions/setup-java/compare/b6effb05e454b25005698d916606bdc6ffcbf961...dd06d9cba3e5552c54d9f8ea23572deb30010f7c)\n\n---\nupdated-dependencies:\n- dependency-name: actions/setup-java\n  dependency-version: 6.0.0\n  dependency-type: direct:production\n  update-type: version-update:semver-major\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "6477f539792c1a96fe799fc9b3e7abd9990d3a55",
      "tree": "928aeb84b89c08797fbf5adc310a305982ae7853",
      "parents": [
        "8c11091f8a97ae1e4a26302d1dd0598f33c39bfa"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sun Aug 30 08:38:53 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Aug 30 09:38:53 2026 +0200"
      },
      "message": "apply a max message len (#818)\n\n* apply a max message len\n\nMotivation:\ngRPC message size limits protect both clients and servers from decompression\nbombs and memory exhaustion. grpc-java defaults to 4 MiB but Pekko gRPC had no\nlimit, leaving it vulnerable to oversized frames and gzip decompression attacks.\n\nModification:\nProtocol layer - `GrpcFramingDecoderStage` and the strict decoder validate the\ndeclared frame length against `maxInboundMessageSize` before the payload is\nbuffered. `Codec.uncompress` gained overloads that bound the decompressed size\nand throw `RESOURCE_EXHAUSTED` when it is exceeded; the compression-bit overload\ndelegates to the size-bounded one so `Gzip` can fail fast while decompressing\nrather than after. `Gzip` checks cumulative output during streaming\ndecompression using a `Long` accumulator and a clamped initial buffer. Negative\nframe lengths are reported as `INTERNAL` on both the streaming and the strict\npath.\n\nClient side - added `max-inbound-message-size` to `pekko.grpc.client.\"*\"`\n(default 4 MiB) and `GrpcClientSettings.withMaxInboundMessageSize`. Both\nbackends honour it.\n\nServer side - added `pekko.grpc.server.max-inbound-message-size` and a new\n`GrpcServerSettings` class. Generated Scala and Java handlers resolve the\nsetting once per `partial`/`handler` call rather than per request.\n\nResult:\nBoth clients and servers reject inbound messages exceeding the configured limit\nwith `RESOURCE_EXHAUSTED`. Oversized frames are rejected from the declared\nlength before the payload is buffered, and gzip frames that inflate past the\nlimit are rejected while decompressing rather than after.\n\nCo-Authored-By: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e\n\n* Deprecate the unbounded Codec.uncompress overloads\n\nMotivation:\n`uncompress(bytes)` and `uncompress(compressedBitSet, bytes)` place no bound on\nthe decompressed output, so a caller that reaches for them is exposed to the\ndecompression bomb this branch otherwise closes. No production call site uses\nthem any more - the framing decoder went through the size-bounded overloads -\nbut they remain public and are the shape through which a future caller would\nreintroduce the problem.\n\nModification:\nMarked both overloads `@deprecated` in `Codec`, `Gzip` and `Identity`, pointing\nat the size-bounded replacements. The default `uncompress(bytes, max)` in\n`Codec` and `Gzip`\u0027s compression-bit overload still delegate to them, so both\ncarry `@nowarn(\"cat\u003ddeprecation\")`; the build runs with `-Xfatal-warnings`.\nDocumented on the default `uncompress(bytes, max)` that it checks after\ndecompressing and that codecs able to fail fast should override it, as `Gzip`\ndoes.\n\nResult:\nThe unbounded entry points still work for source compatibility but now warn,\nand the bounded overloads are the discoverable ones.\n\nCo-Authored-By: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e\n\n---------\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "8c11091f8a97ae1e4a26302d1dd0598f33c39bfa",
      "tree": "1f3d90a0cbc0f59292b6119d4c1a9519a0baf14b",
      "parents": [
        "f978f7e6f0c9f455c663c5f270757faa83fb0993"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sun Aug 30 08:37:39 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Aug 30 09:37:39 2026 +0200"
      },
      "message": "Percent-decode grpc-message on the pekko-http client (#869)\n\nMotivation:\n`grpc-message` travels UTF-8 percent-encoded on the wire. The server side\nencodes it, and `Status-Message.parse` exists to decode it, but the pekko-http\nclient\u0027s `mapToStatusException` read the header value raw. A non-ASCII error\ndescription reached the caller as `%D0%BF...` rather than text. The netty\nbackend decodes it in grpc-java, so this was pekko-http only.\n\nFixing that exposed a second bug underneath. `PercentEncoding.Decoder.decode`\nconverted its input with `getBytes(US_ASCII)`, which replaces every non-ASCII\ncharacter with \u0027?\u0027. A conforming grpc-message is pure ASCII so this never bit on\nvalid input, but the protocol requires that an invalid value is neither errored\non nor thrown away, and this silently destroyed it. Only values containing a\npercent reach the slow path, which is why the existing round-trip tests missed\nit: they always encode first, so the non-ASCII is already escaped by then.\n\ngrpc-java does not have this bug. Its `parseAsciiString` takes the wire bytes\ndirectly and uses US-ASCII only for the two hex digits after a percent, which\nare ASCII by definition. The pekko port takes a String, because\n`ModeledCustomHeaderCompanion.parse` hands it one, so it needs a String-to-bytes\nstep that grpc-java does not have - and that step picked US-ASCII, following the\nname `parseAsciiString` rather than its data flow.\n\nModification:\n- `mapToStatusException` decodes the header through `PercentEncoding.Decoder`.\n- `Decoder.TransferEncoding` is UTF-8. It is byte-identical to US-ASCII for\n  conforming input, and preserves the characters of a non-conforming one.\n  `decodeSlow` already copies payload bytes through one at a time and decodes\n  once as UTF-8 at the end, so multi-byte sequences reassemble correctly.\n\nResult:\nA percent-encoded `grpc-message` reaches the caller as text on both backends,\nand a non-conforming one is no longer mangled.\n\nTests:\n- 3 cases in `PekkoHttpClientUtilsSpec`: an encoded message, a plain one that\n  must pass through untouched, and a broken `%ZZ` escape that must not be\n  thrown away\n- 1 case in `HeadersSpec` for the decoder, covering non-ASCII alongside a\n  percent - the shape the existing round-trip tests could not reach\n- Confirmed both guards bite: reverting the client-side decode fails the\n  encoded-message case, and restoring US-ASCII fails the decoder case\n- sbt \"runtime/testOnly ...PekkoHttpClientUtilsSpec ...HeadersSpec\" - 14 passed\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt scalafmtAll scalafmtSbt - applied\n- sbt \"runtime/test\" - not run locally, left to CI\n\nCo-authored-by: Claude Opus 5 (1M context) \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "f978f7e6f0c9f455c663c5f270757faa83fb0993",
      "tree": "7997d56dc42139af4b5e40fc5a6d9002b0f92640",
      "parents": [
        "150f0179b015be4dd4aabeca78ea9168b4dc2f10"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Sat Aug 29 17:33:47 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Aug 29 16:33:47 2026 +0100"
      },
      "message": "Update sbt-assembly to 2.5.0 (#864)"
    },
    {
      "commit": "150f0179b015be4dd4aabeca78ea9168b4dc2f10",
      "tree": "98b62c5c5d897add84ccd2c6a25a15bec0644641",
      "parents": [
        "954a105a68c48df0891de86b78c7e4052dea5ed6"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Aug 25 18:19:37 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 25 17:19:37 2026 +0100"
      },
      "message": "Update protobuf-java to 4.36.0 (#854)\n\n* Update protobuf-java to 4.36.0\n\n* more refs\n\n* Stop duplicate com.google.protobuf gencode shadowing protobuf-java\n\nMotivation:\nThe gen-scala-server/00-interop scripted test fails with protobuf-java 4.36.0:\n\n  IllegalArgumentException: Edition EDITION_PROTO2 is greater than the\n  maximum supported edition EDITION_UNKNOWN!\n\nThat test depends on grpc-interop-testing in the \"protobuf-src\" configuration\nand generates Java as well as Scala, so protoc also compiles the\ngoogle/protobuf/*.proto extracted from the transitive protobuf-java into\ncom.google.protobuf gencode. Those classes shadow the real protobuf-java on\nthe classpath, and the \"protobuf-src\" configuration resolves protobuf-java\nindependently of the pinned version, so the gencode came from 4.34.0 while the\nruntime was 4.36.0. 4.34.0\u0027s Edition enum has no EDITION_2026, which 4.36.0\ndeclares as its maximum edition, so the shadowing FeatureSetDefaults reported\nEDITION_UNKNOWN. With 4.35.1 the maximum was EDITION_2024, which that gencode\nknows, so the duplicate classes stayed benign.\n\nModification:\nPass the pinned protobuf-java version to scripted tests and use it for the\n\"protobuf-src\" dependency in 00-interop, so the extracted well-known-type\nprotos match the runtime. Also keep the duplicate com.google.protobuf gencode\nout of that test\u0027s compilation, as the interop-tests project already does.\nExtract the interop-tests well-known-type protos from the same pinned\nprotobuf-java rather than the older one grpc brings in.\n\nResult:\nThe gencode no longer shadows protobuf-java, and the interop scripted test\npasses on protobuf-java 4.36.0.\n\nTests:\n- sbt \"sbt-plugin / scripted gen-scala-server/00-interop\" - passed (failed\n  before this change with the error above)\n- sbt \"interop-tests / Test / compile\" - passed\n- sbt scalafmtSbtCheck - passed\n\nReferences:\nRefs #854\n\n---------\n\nCo-authored-by: PJ Fanning \u003cpjfanning@users.noreply.github.com\u003e"
    },
    {
      "commit": "954a105a68c48df0891de86b78c7e4052dea5ed6",
      "tree": "0180f4bb205e1f911a0c784997ff6b908ead3418",
      "parents": [
        "4084ae5ee9bb9ad8904d94d43e2379e70583f14f"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Tue Aug 25 16:18:40 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 25 16:18:40 2026 +0100"
      },
      "message": "build(deps): bump sbt/setup-sbt from 1.5.7 to 1.5.8 (#861)\n\nBumps [sbt/setup-sbt](https://github.com/sbt/setup-sbt) from 1.5.7 to 1.5.8.\n- [Release notes](https://github.com/sbt/setup-sbt/releases)\n- [Commits](https://github.com/sbt/setup-sbt/compare/8feba82adc7f01ddcf8165b86f778bdb5b82cebc...c7d2d6258b4bd0d3ec5129e6b3453199d3c79729)\n\n---\nupdated-dependencies:\n- dependency-name: sbt/setup-sbt\n  dependency-version: 1.5.8\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "4084ae5ee9bb9ad8904d94d43e2379e70583f14f",
      "tree": "4f8d68ccc53a6f7b3bd955be9ce11601fb1267d5",
      "parents": [
        "fbb7f41eb20b77d7984a355a0704fd62bbef67d7"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Mon Aug 24 11:45:35 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 24 11:45:35 2026 +0100"
      },
      "message": "Harden and test TLS hostname verification for the pekko-http client (#857)\n\nMotivation:\nFollow-up to #820 and #830. `ConnectionContext.httpsClient(SSLContext)`,\nwhich the client used before #820, already sets both client mode and the\n`https` endpoint identification algorithm, so hostname verification was\nnever absent. What #820 changed was to move to the\n`httpsClient((host, port) \u003d\u003e ...)` overload, which pekko-http marks\n`@ApiMayChange` and documents as leaving SNI and hostname verification to\nthe caller. The secure default now depends on pekko-grpc reproducing that\nsetup correctly, and nothing tests or documents it.\n\nModification:\n- The verifying path delegates to `ConnectionContext.httpsClient(SSLContext)`\n  again, so the default rides on pekko-http\u0027s own maintained implementation.\n  The hand-built engine is confined to the `verify-hostname \u003d false` opt-out,\n  where it now also sets client mode: `Http.sslTlsStage` leaves the engine\n  entirely to the creator, so nothing sets it on the HTTP/1.1 path. Only\n  `Http2.outgoingConnection` happens to set it today, which is an internal\n  detail rather than a contract.\n- Extracted `sslContextFor`, `insecureSslEngineCreator` and `connectionContext`\n  as internal helpers so the verification contract is reachable from a test.\n- The netty backend logs a warning when `verify-hostname \u003d false` is set.\n  grpc-java\u0027s netty transport always verifies and exposes no switch, so the\n  setting was silently ignored on what is the default backend.\n- Reworded the pekko-http warning: the only readers of it are users who\n  explicitly disabled verification, so pointing at the default is clearer\n  than telling them to set what they just overrode.\n- New `HostnameVerificationSpec` performs real TLS handshakes against a local\n  server whose certificate carries a single `DNS:localhost` SAN, asserting\n  that reaching it as `127.0.0.1` fails when verification is on and succeeds\n  when it is off. The pair is a differential guard: with endpoint\n  identification unset the rejecting case passes, since the certificate is\n  otherwise valid and trusted.\n- Documented the setting, which hostname is actually checked (the authority,\n  not the discovered address), the production warning, and that it does not\n  apply to the netty backend.\n\nThe certificates under runtime/src/test/resources/certs are copied from\nplugin-tester-scala/src/main/resources/certs in this repository.\n\nResult:\nThe secure default is provided by pekko-http rather than reimplemented,\nthe opt-out no longer depends on an unrelated part of pekko-http to set\nclient mode, a silently ignored setting on the default backend is now\nreported, and the verification contract is covered by tests and docs.\n\nTests:\n- sbt \"runtime/test\" - 134 tests passed, 5 of them new\n- sbt \"runtime/mimaReportBinaryIssues\" - passed\n- sbt scalafmtCheckAll scalafmtSbtCheck - passed\n- sbt docs/paradox - passed\n- sbt headerCreateAll - applied\n- sbt \"sbt-plugin / scripted\" - Not run, left to CI\n- sbt +mimaReportBinaryIssues - Not run, single-version run passed, cross\n  version left to CI\n\nReferences:\nRefs #820, Refs #830"
    },
    {
      "commit": "fbb7f41eb20b77d7984a355a0704fd62bbef67d7",
      "tree": "46ff1edb6e2572de85cd7c3f60a87155a7ee3a65",
      "parents": [
        "85ec44b046f79b46c57a25296c48b33f8375df51"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Sun Aug 23 17:38:07 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Aug 23 16:38:07 2026 +0100"
      },
      "message": "Update scripted-plugin to 1.12.15 (#860)"
    },
    {
      "commit": "85ec44b046f79b46c57a25296c48b33f8375df51",
      "tree": "25219c4685a424e73c3f05dd0b03260a91c5312a",
      "parents": [
        "e457fd078355b17ec7a2d9af2436620c251de3d7"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sat Aug 22 12:01:39 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Aug 22 12:01:39 2026 +0100"
      },
      "message": "Update pekkoVersion to 2.0.0-M4 (#856)\n\n* Update pekkoVersion to 2.0.0-M4\n\n* avoid deprecated method\n\n* more versions to update"
    },
    {
      "commit": "e457fd078355b17ec7a2d9af2436620c251de3d7",
      "tree": "9e657ad43a3f6faca7d285236a9ab10cc23b3ae1",
      "parents": [
        "1b3e197cd1633dbe499237ed9d95797579eb738a"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Aug 21 16:24:43 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 21 15:24:43 2026 +0100"
      },
      "message": "Update sbt, scripted-plugin to 1.13.0 (#855)"
    },
    {
      "commit": "1b3e197cd1633dbe499237ed9d95797579eb738a",
      "tree": "b52a730d0d62c66207309c9e3ecd4c08b56a8ac5",
      "parents": [
        "10ff8fcf4ad54791df5d97650a3d80a41bdacaa4"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Fri Aug 21 10:11:50 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 21 10:11:50 2026 +0100"
      },
      "message": "CI jobs for 1.3.x (#853)"
    },
    {
      "commit": "10ff8fcf4ad54791df5d97650a3d80a41bdacaa4",
      "tree": "bc90fbf5f0a98124b44593802f56427daf3c6f74",
      "parents": [
        "24c42730804b190b0c2f78aad0ac448abc0024e3"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Thu Aug 20 14:27:59 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Aug 20 14:27:59 2026 +0100"
      },
      "message": "csvparser util for headers (#852)\n\n* csvparser util for headers\n\n* Update headers.scala"
    },
    {
      "commit": "24c42730804b190b0c2f78aad0ac448abc0024e3",
      "tree": "204aa25c7caf5644bb1c245440659292d2244cfd",
      "parents": [
        "72cbb022de2e42bb5cc672ec80d892fb2f8bb7b6"
      ],
      "author": {
        "name": "haroldb1",
        "email": "harold.bergson@pirum.com",
        "time": "Thu Aug 20 10:16:22 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Aug 20 10:16:22 2026 +0100"
      },
      "message": "Fix Codecs.negotiate() dropping supported encodings with whitespace (#850)\n\nMotivation:\ngrpc-accept-encoding headers formatted with a space after the comma\n(e.g. \"deflate, gzip\", as sent by grpc-go/grpc-python/grpcurl) were\nnever matched against supported codec names, silently downgrading\nnegotiation to Identity even when the client offered gzip.\n\nModification:\nTrim each split token in `Message-Accept-Encoding`.findIn before\ncodec-name lookup.\n\nResult:\nCodecs.negotiate now correctly picks a supported codec regardless of\nwhitespace around commas in grpc-accept-encoding.\n\nTests:\n- sbt \"runtime / testOnly org.apache.pekko.grpc.CodecsSpec\" - pass\n\nReferences:\nFixes #847"
    },
    {
      "commit": "72cbb022de2e42bb5cc672ec80d892fb2f8bb7b6",
      "tree": "0af80986aa4befbc12641265364d5aca65b0bdf5",
      "parents": [
        "23d6b2c38c9b669dadff39255cecec49b7b2bd77"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Wed Aug 19 22:59:39 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Aug 19 22:59:39 2026 +0100"
      },
      "message": "Protect 1.3.x branch (#844)"
    },
    {
      "commit": "23d6b2c38c9b669dadff39255cecec49b7b2bd77",
      "tree": "78e0d31e0d9bc78e7189b0bffdb941fbcb218e92",
      "parents": [
        "f33c583116e70dfa4c81bd30b9a0028ef381b345"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Wed Aug 19 20:13:34 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Aug 19 19:13:34 2026 +0100"
      },
      "message": "Update sbt-assembly to 2.4.2 (#843)"
    },
    {
      "commit": "f33c583116e70dfa4c81bd30b9a0028ef381b345",
      "tree": "f6fb35fa15e5b230290658623777b03bea5efb87",
      "parents": [
        "f5f08e0d50adacb5e5701f7879c6f2c0c78c1b7e"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Aug 17 10:29:23 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 17 10:29:23 2026 +0100"
      },
      "message": "build(deps): bump sbt/setup-sbt from 1.5.6 to 1.5.7 (#841)\n\nBumps [sbt/setup-sbt](https://github.com/sbt/setup-sbt) from 1.5.6 to 1.5.7.\n- [Release notes](https://github.com/sbt/setup-sbt/releases)\n- [Commits](https://github.com/sbt/setup-sbt/compare/bfea3c5f48abd221b04a6df4798aa5eb8b6a2baf...8feba82adc7f01ddcf8165b86f778bdb5b82cebc)\n\n---\nupdated-dependencies:\n- dependency-name: sbt/setup-sbt\n  dependency-version: 1.5.7\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "f5f08e0d50adacb5e5701f7879c6f2c0c78c1b7e",
      "tree": "d892d1b2ba16980a867b1ff0dcf13021b2026491",
      "parents": [
        "cbb8dad99d40c17fd380c047064a798edfbb3b4a"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Mon Aug 17 10:28:48 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 17 10:28:48 2026 +0100"
      },
      "message": "add gzip header regression tests (#839)"
    },
    {
      "commit": "cbb8dad99d40c17fd380c047064a798edfbb3b4a",
      "tree": "5b3a3945f0d2f6fd93f41355d6b0522f49e1d4d5",
      "parents": [
        "2a35abf81b1225e474bf757aa88071e959cd00e2"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Sat Aug 15 21:55:47 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Aug 15 20:55:47 2026 +0100"
      },
      "message": "Update logback-classic to 1.6.3 (#840)"
    },
    {
      "commit": "2a35abf81b1225e474bf757aa88071e959cd00e2",
      "tree": "da44dcfbc1cf3869873889a7a43c3159bf4e6d96",
      "parents": [
        "77978a7be1276de439edf481cf6f70b457e2327a"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Aug 11 17:35:46 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 11 17:35:46 2026 +0200"
      },
      "message": "Update logback-classic to 1.6.2 (#836)"
    },
    {
      "commit": "77978a7be1276de439edf481cf6f70b457e2327a",
      "tree": "2150393bcfebffb136b30c30d3ffd64de0258df4",
      "parents": [
        "cbf4097173f1955ea1cb8f7b15e5955b75522b46"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Tue Aug 11 16:35:06 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 11 16:35:06 2026 +0200"
      },
      "message": "Update sbt-site-paradox to 1.8.0 (#837)"
    },
    {
      "commit": "cbf4097173f1955ea1cb8f7b15e5955b75522b46",
      "tree": "eb77ec3ae0bb7b32a4960b417a98a88d1e1aa536",
      "parents": [
        "faabc60a155d660a4ac748f2a1bf5c0ea9e3a8f9"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Mon Aug 10 12:54:00 2026 +0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Aug 10 12:54:00 2026 +0800"
      },
      "message": "build(deps): bump coursier/setup-action from 3.0.0 to 3.0.2 (#835)\n\nBumps [coursier/setup-action](https://github.com/coursier/setup-action) from 3.0.0 to 3.0.2.\n- [Release notes](https://github.com/coursier/setup-action/releases)\n- [Commits](https://github.com/coursier/setup-action/compare/fd1707a76b027efdfb66ca79318b4d29b72e5a02...9b7939bf01fd1185ce2babe16135168361bf2c62)\n\n---\nupdated-dependencies:\n- dependency-name: coursier/setup-action\n  dependency-version: 3.0.2\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    },
    {
      "commit": "faabc60a155d660a4ac748f2a1bf5c0ea9e3a8f9",
      "tree": "371a99414e059b42c0777a4924b522c56cd449c8",
      "parents": [
        "5d4932115c7e5f861c3ef9e02c447ad00ef9c9b3"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sat Aug 08 22:04:24 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Aug 08 22:04:24 2026 +0100"
      },
      "message": "trailer frame support in GrpcProtocol appears broken (#822)\n\n* trailer frame support in GrpcProtocol appears broken\n\n* Update GrpcProtocolWebSpec.scala\n\n* support LF only splits in trailer\n\n* Update GrpcProtocolWeb.scala\n\n* Update GrpcProtocolWebSpec.scala\n\n* test issue\n\n* remove return from decodeTrailer"
    },
    {
      "commit": "5d4932115c7e5f861c3ef9e02c447ad00ef9c9b3",
      "tree": "ac74d6cb85bc1a512b3dc0242f34457109c36146",
      "parents": [
        "bcbacbb035e82f556793387d5acbda9fcbe78856"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sat Aug 08 08:37:45 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Aug 08 08:37:45 2026 +0100"
      },
      "message": "apply deadline on pekko-http backend (#824)\n\n* apply deadline on pekko-http backend\n\n* Refactor applyDeadline call for clarity\n\n* Update PekkoHttpClientUtils.scala"
    },
    {
      "commit": "bcbacbb035e82f556793387d5acbda9fcbe78856",
      "tree": "c61ac6f28033565a92032a4a93213f5e6b7bb199",
      "parents": [
        "5014c835b3cb628c79aea1684f13255aa0616ab3"
      ],
      "author": {
        "name": "PJ Fanning",
        "email": "pjfanning@users.noreply.github.com",
        "time": "Sat Aug 08 07:12:41 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Aug 08 07:12:41 2026 +0100"
      },
      "message": "tidy up close logic (#825)"
    },
    {
      "commit": "5014c835b3cb628c79aea1684f13255aa0616ab3",
      "tree": "70260e860f464ec28a03b6973e4b3441ed7dc832",
      "parents": [
        "fbd673fb6b5ad2c46e5ea44f891b96fd71360309"
      ],
      "author": {
        "name": "Scala Steward",
        "email": "43047562+scala-steward@users.noreply.github.com",
        "time": "Fri Aug 07 20:22:08 2026 +0200"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 07 19:22:08 2026 +0100"
      },
      "message": "Update sbt-paradox, sbt-paradox-theme to 0.11.0 (#831)"
    },
    {
      "commit": "fbd673fb6b5ad2c46e5ea44f891b96fd71360309",
      "tree": "17104139fdf6d18514869524d33c783b5594043e",
      "parents": [
        "90a561d00f9f3ee9474c9dcbe31b0571390c52d3"
      ],
      "author": {
        "name": "dependabot[bot]",
        "email": "49699333+dependabot[bot]@users.noreply.github.com",
        "time": "Fri Aug 07 17:09:54 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 07 17:09:54 2026 +0100"
      },
      "message": "build(deps): bump sbt/setup-sbt from 1.5.4 to 1.5.6 (#828)\n\nBumps [sbt/setup-sbt](https://github.com/sbt/setup-sbt) from 1.5.4 to 1.5.6.\n- [Release notes](https://github.com/sbt/setup-sbt/releases)\n- [Commits](https://github.com/sbt/setup-sbt/compare/f6db8ab474efba716fc7f04441ab33714e4825af...bfea3c5f48abd221b04a6df4798aa5eb8b6a2baf)\n\n---\nupdated-dependencies:\n- dependency-name: sbt/setup-sbt\n  dependency-version: 1.5.6\n  dependency-type: direct:production\n  update-type: version-update:semver-patch\n...\n\nSigned-off-by: dependabot[bot] \u003csupport@github.com\u003e\nCo-authored-by: dependabot[bot] \u003c49699333+dependabot[bot]@users.noreply.github.com\u003e"
    }
  ],
  "next": "90a561d00f9f3ee9474c9dcbe31b0571390c52d3"
}
