lmxopcua/tests/Drivers/ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient.Tests/OpcUaClientWriteHealthTests.cs

using System.Text.Json;
using Moq;
using Opc.Ua;
using Opc.Ua.Client;
using Shouldly;
using Xunit;
using ZB.MOM.WW.OtOpcUa.Core.Abstractions;

namespace ZB.MOM.WW.OtOpcUa.Driver.OpcUaClient.Tests;

/// <summary>
///     The health contract for the write path and the steady keep-alive: a write RESPONSE — whatever per-node
///     statuses it carries, a refusal included — is the server answering, so it restores Healthy and advances
///     <see cref="DriverHealth.LastSuccessfulRead"/>; a write that got NO response degrades health with the stamp
///     preserved; a caller-cancelled write is not a comms fault. A good keep-alive clears a Degraded left by a failed
///     read/write (it used to only move the stamp, leaving a subscription-only driver Degraded forever).
/// </summary>
[Trait("Category", "Unit")]
public sealed class OpcUaClientWriteHealthTests
{
    private const string SpeedRef = "Plant/dev/Speed";
    private static readonly DateTime Stale = new(2020, 1, 1, 0, 0, 0, DateTimeKind.Utc);

    private static readonly IReadOnlyList<RawTagEntry> Tags =
        [new(SpeedRef, JsonSerializer.Serialize(new { nodeId = "ns=0;s=Speed" }), WriteIdempotent: true)];

    private static (OpcUaClientDriver Driver, Mock<ISession> Session) NewConnectedDriver(DriverHealth health)
    {
        var drv = new OpcUaClientDriver(new OpcUaClientDriverOptions { RawTags = Tags }, "opcua-write-health");
        var session = new Mock<ISession>(MockBehavior.Loose);
        session.SetupGet(s => s.MessageContext).Returns(new ServiceMessageContext(SdkTelemetry.Context));
        session.SetupGet(s => s.NamespaceUris).Returns(new NamespaceTable());
        drv.SetSessionForTest(session.Object);
        drv.SetNamespaceMapForTest(NamespaceMap.FromTable(new NamespaceTable(), Tags));
        drv.SetHealthForTest(health);
        return (drv, session);
    }

    private static ZB.MOM.WW.OtOpcUa.Core.Abstractions.WriteRequest[] Write() => [new(SpeedRef, 7)];

    /// <summary>A refused write (BadNotWritable) is a reply: Healthy again, stamp advanced.</summary>
    [Fact]
    public async Task A_refused_write_is_a_reply_and_restores_healthy()
    {
        var (drv, session) = NewConnectedDriver(new DriverHealth(DriverState.Degraded, Stale, "read failed"));
        using var _ = drv;
        session.Setup(s => s.WriteAsync(It.IsAny<RequestHeader>(), It.IsAny<WriteValueCollection>(), It.IsAny<CancellationToken>()))
            .ReturnsAsync(new WriteResponse
            {
                ResponseHeader = new ResponseHeader(),
                Results = [StatusCodes.BadNotWritable],
                DiagnosticInfos = [],
            });

        var results = await drv.WriteAsync(Write(), TestContext.Current.CancellationToken);

        results.Single().StatusCode.ShouldBe(StatusCodes.BadNotWritable, "the upstream refusal still passes through verbatim");
        var h = drv.GetHealth();
        h.State.ShouldBe(DriverState.Healthy);
        h.LastSuccessfulRead.ShouldNotBeNull().ShouldBeGreaterThan(Stale);
        h.LastError.ShouldBeNull();
    }

    /// <summary>A write with no reply (transport failure) degrades health and preserves the stamp.</summary>
    [Fact]
    public async Task A_write_with_no_reply_degrades_and_preserves_the_stamp()
    {
        var (drv, session) = NewConnectedDriver(new DriverHealth(DriverState.Healthy, Stale, null));
        using var _ = drv;
        session.Setup(s => s.WriteAsync(It.IsAny<RequestHeader>(), It.IsAny<WriteValueCollection>(), It.IsAny<CancellationToken>()))
            .ThrowsAsync(new ServiceResultException(StatusCodes.BadConnectionClosed));

        var results = await drv.WriteAsync(Write(), TestContext.Current.CancellationToken);

        results.Single().StatusCode.ShouldBe(0x80050000u); // BadCommunicationError
        var h = drv.GetHealth();
        h.State.ShouldBe(DriverState.Degraded);
        h.LastSuccessfulRead.ShouldBe(Stale, "the comms-loss watchdog measures silence from this stamp");
    }

    /// <summary>
    ///     A write the CALLER cancelled says nothing about the upstream: health is left alone. (The positive control is
    ///     the test above — the same throw-path does write health when the failure is the transport's.)
    /// </summary>
    [Fact]
    public async Task A_caller_cancelled_write_leaves_health_alone()
    {
        var healthy = new DriverHealth(DriverState.Healthy, Stale, null);
        var (drv, session) = NewConnectedDriver(healthy);
        using var _ = drv;
        using var cts = new CancellationTokenSource();
        session.Setup(s => s.WriteAsync(It.IsAny<RequestHeader>(), It.IsAny<WriteValueCollection>(), It.IsAny<CancellationToken>()))
            .Returns(async (RequestHeader _, WriteValueCollection _, CancellationToken ct) =>
            {
                cts.Cancel();
                await Task.Delay(Timeout.Infinite, ct);
                return new WriteResponse();
            });

        var results = await drv.WriteAsync(Write(), cts.Token);

        results.Single().StatusCode.ShouldBe(0x800A0000u); // BadTimeout — outcome unknown
        drv.GetHealth().ShouldBe(healthy);
    }

    /// <summary>A good keep-alive is a reply: it clears a Degraded a failed read left behind.</summary>
    [Fact]
    public void A_good_keep_alive_clears_a_degraded_left_by_a_failed_read()
    {
        var (drv, _) = NewConnectedDriver(new DriverHealth(DriverState.Degraded, Stale, "read: BadTimeout"));
        using var __ = drv;

        drv.RaiseKeepAliveForTest(StatusCodes.Good);

        var h = drv.GetHealth();
        h.State.ShouldBe(DriverState.Healthy);
        h.LastSuccessfulRead.ShouldNotBeNull().ShouldBeGreaterThan(Stale);
        h.LastError.ShouldBeNull();
    }

    /// <summary>
    ///     Review m8: a write that TIMED OUT (an OperationCanceledException the caller did not cause) is a call with no
    ///     reply — Degraded, stamp preserved. The genuine positive control for the caller-cancelled test above (same
    ///     catch branch, the caller's token is the only difference).
    /// </summary>
    [Fact]
    public async Task A_write_that_timed_out_without_caller_cancellation_degrades()
    {
        var (drv, session) = NewConnectedDriver(new DriverHealth(DriverState.Healthy, Stale, null));
        using var _ = drv;
        session.Setup(s => s.WriteAsync(It.IsAny<RequestHeader>(), It.IsAny<WriteValueCollection>(), It.IsAny<CancellationToken>()))
            .ThrowsAsync(new OperationCanceledException("SDK request timeout"));

        var results = await drv.WriteAsync(Write(), TestContext.Current.CancellationToken);

        results.Single().StatusCode.ShouldBe(0x800A0000u); // BadTimeout
        drv.GetHealth().State.ShouldBe(DriverState.Degraded);
        drv.GetHealth().LastSuccessfulRead.ShouldBe(Stale);
    }

    /// <summary>
    ///     Review m7: a reply that was in flight across a bad keep-alive advances the stamp (it is evidence) but does not
    ///     declare the published outage over — health stays Degraded until the keep-alive recovers.
    /// </summary>
    [Fact]
    public async Task A_reply_during_a_published_outage_advances_the_stamp_but_keeps_degraded()
    {
        var (drv, session) = NewConnectedDriver(new DriverHealth(DriverState.Healthy, Stale, null));
        using var _ = drv;
        drv.RaiseKeepAliveForTest(StatusCodes.BadNoCommunication);
        drv.GetHealth().State.ShouldBe(DriverState.Degraded, "positive control: the outage is published");
        session.Setup(s => s.WriteAsync(It.IsAny<RequestHeader>(), It.IsAny<WriteValueCollection>(), It.IsAny<CancellationToken>()))
            .ReturnsAsync(new WriteResponse { ResponseHeader = new ResponseHeader(), Results = [StatusCodes.Good], DiagnosticInfos = [] });

        await drv.WriteAsync(Write(), TestContext.Current.CancellationToken);

        drv.GetHealth().State.ShouldBe(DriverState.Degraded);
        drv.GetHealth().LastSuccessfulRead.ShouldNotBeNull().ShouldBeGreaterThan(Stale);

        drv.RaiseKeepAliveForTest(StatusCodes.Good);
        drv.GetHealth().State.ShouldBe(DriverState.Healthy, "the keep-alive recovery is what ends the outage");
    }
}