fix(mqtt-contracts): strict retainSeed, real bounds instead of dead [Range]s, DateTime edges, closed topic formatting

Driver.Mqtt.Contracts-004/-005/-006/-007.

-004: retainSeed was the last behavioural key read leniently, so "retainSeed":"false"
— what a CSV import produces — silently became the default TRUE, the INVERSE of the
authored intent. Both mappers now read it through TryReadBoolStrict (the qos
absent/valid/present-but-invalid split) and DescribeRejection names it. The authoring
side needed no hand-mirrored rule: MqttTagConfigModel.Validate already consults this
same parse through TagConfigParseGate, so the save is blocked with the inspector's
wording — pinned by a new AdminUI parity theory pair.

-005: the seven [Range] attributes on MqttDriverOptions/MqttPlainOptions never ran —
the options are bound by JsonSerializer, and nothing in the four MQTT projects calls
Validator.TryValidateObject. Two of them CONTRADICTED the real authority (KeepAlive's
[Range(1,…)] forbade the zero Normalize admits as legal MQTT; MaxPayloadBytes' forbade
the non-positive value its own remarks document as falling back to 1 MiB), so wiring
them up would have broken deliberate behaviour. Deleted, with each property naming
where its bound actually lives. The one escape they papered over is closed: an
out-of-range defaultQos rode into the SUBSCRIBE verbatim, and MqttSubscriptionManager's
constructor now clamps it exactly as the AdminUI driver form already persists.

-006: the DateTime write arm cast a negative long to ulong, so 1969-12-31 published a
timestamp ~584 million years in the future; it is now refused as ValueNotRepresentable.
An Unspecified kind was handed to ToUniversalTime(), which reads it as LOCAL — the same
authored instant encoded differently on the two nodes of a redundant pair; it is now
read as UTC.

-007: Format/FormatState enforced only null/empty, and nothing at all on deviceId, so
the type was not round-trip-closed. They now apply the parser's IsValidSegment (plus
'/', which only the format side can see). MqttWriteEncoder's own guards stay — TryEncode
returns an error rather than throwing, and that contract does not change.
This commit is contained in:
Joseph Doherty
2026-08-10 08:18:30 -04:00
parent eddf5f57a5
commit 93f70504bd
11 changed files with 441 additions and 31 deletions
@@ -150,6 +150,78 @@ public sealed class SparkplugTopicTests
public void ToTopicString_RoundTrips(string topic)
=> SparkplugTopic.Parse(topic).ToTopicString().ShouldBe(topic);
/// <summary>
/// <see cref="SparkplugTopic.Format"/> must enforce the same per-segment rule
/// <see cref="SparkplugTopic.TryParse"/> does. It validated only null/empty on
/// <c>groupId</c>/<c>edgeNodeId</c> and NOTHING at all on <c>deviceId</c>, so the type was not
/// round-trip-closed: a <c>+</c>/<c>#</c> id produced a topic MQTT-3.3.2-2 forbids in a PUBLISH,
/// and a <c>/</c> in a <c>deviceId</c> silently added a segment. Every production caller is
/// already guarded, so the invariant belongs on the type that owns it rather than on the next
/// caller to arrive.
/// </summary>
/// <param name="groupId">The group id to format with.</param>
/// <param name="edgeNodeId">The edge-node id to format with.</param>
/// <param name="deviceId">The device id to format with, or <see langword="null"/>.</param>
[Theory]
[InlineData("Plant+1", "EdgeA", null)]
[InlineData("Plant#", "EdgeA", null)]
[InlineData("Plant1", "Edge+A", null)]
[InlineData("Plant1", "Edge#A", null)]
[InlineData("Plant1", "EdgeA", "Fill+er")]
[InlineData("Plant1", "EdgeA", "Fill#er")]
[InlineData("Ot/Sim", "EdgeA", null)] // two segments, not one — a DIFFERENT node than named
[InlineData("Plant1", "EdgeA", "Fill/er")]
[InlineData("Plant1", "EdgeA", "")] // would emit a trailing '/', which never re-parses
public void Format_SegmentThatIsNotALegalTopicSegment_Throws(string groupId, string edgeNodeId, string? deviceId)
=> Should.Throw<ArgumentException>(
() => SparkplugTopic.Format(groupId, SparkplugMessageType.DCMD, edgeNodeId, deviceId));
[Theory]
[InlineData("host+1")]
[InlineData("host#")]
public void FormatState_HostIdThatIsNotALegalTopicSegment_Throws(string hostId)
=> Should.Throw<ArgumentException>(() => SparkplugTopic.FormatState(hostId));
/// <summary>
/// Falsification control: the rule is the parser's <c>IsValidSegment</c>, not "reject anything
/// unusual". A guard that rejected every id would satisfy both theories above, and a Sparkplug id
/// is otherwise an opaque identifier — spaces, dots and dashes are all legitimate.
/// </summary>
/// <param name="id">A legal, if unusual, segment.</param>
[Theory]
[InlineData("Plant 1")]
[InlineData("plant.one")]
[InlineData("plant-one_2")]
[InlineData("usiné")]
public void Format_UnusualButLegalSegments_AreStillAccepted(string id)
=> SparkplugTopic.Parse(SparkplugTopic.Format(id, SparkplugMessageType.DCMD, id, id))
.ShouldBe(new SparkplugTopic(SparkplugMessageType.DCMD, id, id, id, null));
/// <summary>
/// The closure property the asymmetry broke: anything <see cref="SparkplugTopic.Format"/> emits
/// must parse back to the instance it was built from. It is asserted over the FORMATTER's own
/// output rather than over a hand-written topic list, so a future formatting change cannot leave
/// the corpus describing a shape the type no longer emits.
/// </summary>
[Fact]
public void Format_Output_AlwaysReParsesToTheSameInstance()
{
SparkplugMessageType[] deviceScoped =
[SparkplugMessageType.DBIRTH, SparkplugMessageType.DDATA, SparkplugMessageType.DDEATH, SparkplugMessageType.DCMD];
foreach (var type in Enum.GetValues<SparkplugMessageType>())
{
if (type == SparkplugMessageType.STATE) continue;
var deviceId = deviceScoped.Contains(type) ? "Filler 1" : null;
var topic = SparkplugTopic.Format("Plant 1", type, "Edge.A", deviceId);
SparkplugTopic.Parse(topic).ShouldBe(new SparkplugTopic(type, "Plant 1", "Edge.A", deviceId, null));
}
SparkplugTopic.Parse(SparkplugTopic.FormatState("otopcua host 1"))
.ShouldBe(new SparkplugTopic(SparkplugMessageType.STATE, null, null, null, "otopcua host 1"));
}
[Fact]
public void ToTopicString_LegacyStateTopic_NormalisesToV3Form()
// Parsed from the legacy no-prefix form, but formatting always emits the v3.0 shape.