From f829df8087de9601825b01effe4f55ec96c34c22 Mon Sep 17 00:00:00 2001 From: Joseph Doherty Date: Mon, 9 Nov 2026 12:32:13 -0500 Subject: [PATCH] docs(worker): ReadBulk timeout comments describe one shared deadline (mxaccessgw#179) Claude-Session: https://claude.ai/code/session_01GT67AEHd2Acv1YPwGfXEoV --- .../MxAccess/MxAccessValueCacheTests.cs | 2 +- .../MxAccess/MxAccessCommandExecutor.cs | 2 +- .../MxAccess/MxAccessValueCache.cs | 8 ++++---- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/src/ZB.MOM.WW.MxGateway.Worker.Tests/MxAccess/MxAccessValueCacheTests.cs b/src/ZB.MOM.WW.MxGateway.Worker.Tests/MxAccess/MxAccessValueCacheTests.cs index 2b9151e2..2af99724 100644 --- a/src/ZB.MOM.WW.MxGateway.Worker.Tests/MxAccess/MxAccessValueCacheTests.cs +++ b/src/ZB.MOM.WW.MxGateway.Worker.Tests/MxAccess/MxAccessValueCacheTests.cs @@ -243,7 +243,7 @@ public sealed class MxAccessValueCacheTests /// fallback tick, so a timing-out wait still pumps repeatedly instead of /// once. This is what keeps the STA dispatching COM events in a process /// whose message queue never wakes the wait, and it is the safety net - /// behind the ReadBulk per-tag timeout. + /// behind the ReadBulk shared-deadline timeout. /// [Fact] public void TryWaitForUpdate_KeepsPumping_WhenPollIntervalExceedsTheFallbackTick() diff --git a/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessCommandExecutor.cs b/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessCommandExecutor.cs index a9cfd0ba..4d542ae0 100644 --- a/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessCommandExecutor.cs +++ b/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessCommandExecutor.cs @@ -12,7 +12,7 @@ namespace ZB.MOM.WW.MxGateway.Worker.MxAccess; /// public sealed class MxAccessCommandExecutor : IStaCommandExecutor { - /// Default per-tag timeout used when ReadBulkCommand.timeout_ms is zero. + /// Default timeout used when ReadBulkCommand.timeout_ms is zero. It is one deadline shared by the whole command, not a per-tag budget. internal static readonly TimeSpan DefaultReadBulkTimeout = TimeSpan.FromMilliseconds(1000); /// diff --git a/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessValueCache.cs b/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessValueCache.cs index 652b7ebc..fac45456 100644 --- a/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessValueCache.cs +++ b/src/ZB.MOM.WW.MxGateway.Worker/MxAccess/MxAccessValueCache.cs @@ -212,12 +212,12 @@ public sealed class MxAccessValueCache /// rather than /// Thread.Sleep: a sleeping STA pumps no Windows messages, so a /// ReadBulk waiting on the first OnDataChange would only dispatch it at - /// poll-tick granularity — and it holds the apartment for up to - /// timeout_ms per tag, stalling every other COM event with it. The + /// poll-tick granularity — and it holds the apartment for up to the + /// ReadBulk timeout_ms deadline, stalling every other COM event with it. The /// Win32 wait returns the instant a message needs pumping (or the value /// is recorded from another thread), so the apartment keeps dispatching - /// continuously for the whole wait. The per-tag timeout, the deadline - /// math, and both return shapes are unchanged. + /// continuously for the whole wait. The timeout (one deadline shared by the + /// whole ReadBulk command), the deadline math, and both return shapes are unchanged. /// /// MXAccess server handle. /// MXAccess item handle.