review(Driver.Historian.Wonderware): AtTime fails over on connection-class errors

Re-review at 7286d320. -014 (Medium): ReadAtTimeAsync didn't classify StartQuery failures, so a connection-class failure left a dead connection, re-failed every timestamp, and returned Success=true with all-Bad (no failover); now resets+fails over via a shared classifier + tests. -015: refresh stale named-pipe comments to TCP (no wire change). -013 (silent cap truncation, ties OpcUaServer-002/Core.Abstractions-009) deferred cross-module. NOTE: the SDK-touching tests are net48 + native aahClientManaged and run only on Windows; macOS verifies build + the SDK-free subset only.
2026-06-19 11:47:11 -04:00
parent e07a4fbf52
commit b3907efa6e
8 changed files with 225 additions and 11 deletions
@@ -51,4 +51,54 @@ public sealed class HistorianDataSourceStartQueryClassificationTests
        HistorianDataSource.IsConnectionClassError(code).ShouldBeFalse(
            $"{code} is a query payload problem — must NOT tear down the SDK connection");
    }
+
+    // ── Driver.Historian.Wonderware-014: the at-time loop must classify a per-timestamp
+    //    StartQuery failure the same way the raw / aggregate / event paths do. The SDK
+    //    HistoryQuery type is sealed-by-non-virtual + has no interface, so the loop itself
+    //    can't be driven offline; the per-failure decision is therefore extracted into a
+    //    pure helper that the at-time loop calls and these tests pin directly. ──────────
+
+    /// <summary>
+    ///     A connection-class StartQuery error in the at-time loop must signal "reset the
+    ///     connection and abort the read" (true) — not silently record a Bad sample and keep
+    ///     hammering the dead connection for every remaining timestamp.
+    /// </summary>
+    /// <param name="code">The connection-class error code.</param>
+    [Theory]
+    [InlineData(HistorianAccessError.ErrorValue.FailedToConnect)]
+    [InlineData(HistorianAccessError.ErrorValue.NoReply)]
+    [InlineData(HistorianAccessError.ErrorValue.NotReady)]
+    public void AtTime_StartQuery_failure_with_connection_class_code_requests_connection_reset(
+        HistorianAccessError.ErrorValue code)
+    {
+        var error = new HistorianAccessError { ErrorCode = code };
+        HistorianDataSource.ShouldResetConnectionForStartQueryFailure(error).ShouldBeTrue(
+            $"{code} is a connection failure — the at-time loop must reset the connection, not record Bad");
+    }
+
+    /// <summary>
+    ///     A query-class StartQuery error (or a missing error) in the at-time loop must NOT
+    ///     reset the connection (false): a single bad/empty timestamp records a per-timestamp
+    ///     Bad sample and continues to the next without tearing down the shared connection.
+    /// </summary>
+    /// <param name="code">The query-class error code.</param>
+    [Theory]
+    [InlineData(HistorianAccessError.ErrorValue.InvalidArgument)]
+    [InlineData(HistorianAccessError.ErrorValue.NoData)]
+    [InlineData(HistorianAccessError.ErrorValue.NotApplicable)]
+    public void AtTime_StartQuery_failure_with_query_class_code_does_not_request_reset(
+        HistorianAccessError.ErrorValue code)
+    {
+        var error = new HistorianAccessError { ErrorCode = code };
+        HistorianDataSource.ShouldResetConnectionForStartQueryFailure(error).ShouldBeFalse(
+            $"{code} is a query/no-data problem — the at-time loop keeps the connection and records Bad");
+    }
+
+    /// <summary>A null error defaults to query-class (no reset) — the caller still records a Bad sample.</summary>
+    [Fact]
+    public void AtTime_StartQuery_failure_with_null_error_defaults_to_no_reset()
+    {
+        HistorianDataSource.ShouldResetConnectionForStartQueryFailure(null).ShouldBeFalse(
+            "a null error must not be promoted to a connection reset");
+    }
 }