test: fix shared state, CI-vacuous tests and harness duplication (#2722)

Follow-up to #2721. It fixes how the remaining tests behave and changes no production code.

## CI
- Both `dotnet test` steps pass `--filter "Category!=Manual"`. Before this, Manual-tagged tests ran on CI anyway.
- The one surviving movement-throttle latency test is the only coverage of burst forgiveness over a full, wrapped history. It loses its Manual tag, so it runs on CI.

## Tests that passed on CI without testing anything
- **Corpse and throwing tests.** They relied on tiledata for the weapon `Layer`, so on CI the equip silently failed and they asserted against empty lists. They now set `Layer` explicitly and assert the equip took effect. `TestThrown` also pins its own strength and range values, so a shard retuning SoulGlaive no longer breaks them.
- **Healthbar poison rows.** Server.Tests registers no poisons, so every row tested `null`. They now use real poison levels.
- **Account packet tests.** The mock accounts reported `Length` 0, so the character-list tests only ever encoded empty slots. They now encode the real characters.
- **Pathing and AI tests that only mean something on real Trammel/Felucca terrain** are now guarded with `TileDataRequirement.SkipIfMissing()`: Britain desk and clutter scenes, house footprints (no MultiData on CI) and fingerprints. Tests that are valid on flat blank terrain stay unguarded. (A synthetic tiledata fixture to run these on CI is a planned follow-up.)
- **Assertions that couldn't fail now check their claim.** Affected tests include the `DefaultWalker` cache use, cache on/off parity requiring a path, the recorder capability flags, the idle-pet gate, the monthly DST recurrence, the AdvancedSearch `Record.Exception` wrappers, `CanFitItem` with no branch-guarded assert, the Firewall callback, and `Import_NoMap`. The Geometry Parse asserts also now use the same input as their TryParse.

## Shared state
Every test that changes process-wide state now restores it, even when an assert fails:
- **The clock.** A new `TimerWheelScope` saves and restores `Core._tickCount`, `Core._now` and the timer wheel. `SimulatedClockScope` covers decay tests.
- **StepCache settings and caches**, via `StepCacheStateScope`.
- **Network state.** The ManualAllowlist, LoginAllowlist, firewall and BanChannel reporters are restored. Global packet handlers are now registered per test and removed afterwards, so they no longer stay registered for later tests.
- **Data and config tables.** `ArchiveJournal` statics, `TileData.ItemTable`, `ObjectPropertyList.Enabled`, `Core.Expansion` and the localization table are restored. One `MobileStatusVersion` restore was resetting a different expansion than the one the test changed.
- **Entities.** Tests with fixed serials now delete what they create. Fixed mobile serials moved to `World.NewMobile`, because deleting a hard-coded serial could unregister a live entity.
- **Collection race.** `SpiralScanTests` no longer clears the spawn cache from a separate collection that could run in parallel with the cache-manager tests.
- **Shared statics.** Classes reading shared static allowlists joined the sequential collection.

## Duplication
New shared helpers replace copy-pasted harness code:
- **Server.Tests:** `PacketAssert.Sends`, used by about 130 packet tests; `TestClientHarness`, `PersistencePipelineScope`, `NetTestHelpers`, `EntityCleanup` and one `TestMulti`.
- **UOContent.Tests:** `TimerWheelScope`, which also holds one wraparound-safe `RunFor`/`RunUntil`; `AiTestStubs`, which merges six walker copies and five identical creature stubs; `MobileTestHarness`, `CommandTestHarness` and `TempDirectory`.

Dead oracle packet classes are removed.

## Cost
| Test | Before | After |
|---|---|---|
| `VirtualSerialTests` | 1.57s | 2ms (starts near the wrap via reflection instead of drawing ~17.9M serials) |
| Event scheduler recurrence loop | 473ms | 1ms |
| CrowdSec drain | 517ms | 16ms (event-driven instead of a fixed 500ms wait) |
| AuthId tests | 248ms | 99ms (SHA2 except the two GameLogin tests) |
| Container cap tests | 5,000 items | 3,449 / 3,277 (one past each protocol cap) |
| Bulk removal | 100k children | 20k children |
| OPL hash distribution | 60k builds | 10k builds |

## Shard-tuning decoupling
- AI assertions read `ReacquireDelay` and `FollowMoveSpeed` instead of hard-coding production tuning.
- The guard friendly-fire test uses an `AI_Mage` stub instead of WhiteWyrm.
- Argon2 assertions pin the `argon2id` variant and `v=19` but read the cost parameters from the configured hasher.

## Tests
| Run | Server.Tests | UOContent.Tests |
|---|---|---|
| With client files | 812 pass | 1,123 pass |
| Without client files (CI simulation) | 800 pass, 12 skip | 1,021 pass, 102 skip |

- The UOContent fixture reads `MODERNUO_TEST_DATA_DIR`; Server.Tests reads `MODERNUO_CLIENT_PATH`.
- The 15 new UOContent skips are the map-dependent guards above.
- Total run time with client files is about 5s, down from about 8s.
This commit is contained in:
Kamron Batman 2026-10-18 20:11:33 -07:00 • committed by GitHub
parent 79f736aa83
commit 00134cac33
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
135 changed files with 3164 additions and 4020 deletions

View file

@ -0,0 +1,152 @@
using System;
using System.Buffers;
using System.Diagnostics;
using System.Net.Sockets;
using System.Network;
using System.Runtime.CompilerServices;
using System.Threading;
using Server.Network;
using Xunit;
namespace Server.Tests.Network;
/// <summary>
/// Shared plumbing for tests that drive a real <see cref="NetState"/> over loopback.
/// </summary>
public static class NetTestHelpers
{
public const byte PingPacketId = 0x73;
/// <summary>The payload byte of the last ping delivered to the handler from <see cref="RegisterPing"/>.</summary>
public static byte LastPing { get; private set; }
/// <summary>
/// Restores the handler that was in a packet slot before a test replaced it. The handler table is
/// process-global, so a handler left behind would silently answer for that id in every later test.
/// </summary>
public readonly struct HandlerRegistration : IDisposable
{
private readonly int _packetId;
private readonly PacketHandler _previous;
internal HandlerRegistration(int packetId, PacketHandler previous)
{
_packetId = packetId;
_previous = previous;
}
public void Dispose() => IncomingPackets.Handlers[_packetId] = _previous;
}
public static unsafe HandlerRegistration RegisterHandler(
int packetId, int length, delegate*<NetState, SpanReader, void> onReceive
)
{
var previous = IncomingPackets.GetHandler(packetId);
IncomingPackets.Register(packetId, length, false, onReceive);
return new HandlerRegistration(packetId, previous);
}
/// <summary>Registers a two-byte ping whose payload byte is recorded in <see cref="LastPing"/>.</summary>
public static unsafe HandlerRegistration RegisterPing()
{
LastPing = 0;
return RegisterHandler(PingPacketId, 2, &RecordPing);
}
private static void RecordPing(NetState state, SpanReader reader) => LastPing = reader.ReadByte();
public static byte[] Pattern(int length, int seed)
{
var data = new byte[length];
new System.Random(seed).NextBytes(data);
return data;
}
/// <summary>
/// Pre-auth buffers are off only where the platform floor leaves no room below the base size
/// (the Windows Server 2012/2016 mapping path, or a 64 KiB-page kernel). Anywhere else a zero
/// initial size is a configuration regression and fails rather than skips.
/// </summary>
public static void RequirePreAuthBuffers()
{
var floor = IORingBuffer.MinimumSize;
Skip.If(
floor >= NetState.RecvBufferSize || floor >= NetState.SendBufferSize,
$"Pre-auth buffers are off on this platform: the {floor}-byte mapping floor leaves no room below the base buffers (recv {NetState.RecvBufferSize}, send {NetState.SendBufferSize})"
);
Assert.True(
NetState.InitialRecvBufferSize > 0 && NetState.InitialSendBufferSize > 0,
$"Pre-auth buffers are off although the {floor}-byte platform floor allows them (initial recv {NetState.InitialRecvBufferSize}, send {NetState.InitialSendBufferSize})"
);
}
/// <summary>Runs the network slice until <paramref name="done"/> holds, failing with the condition that timed out.</summary>
public static void SliceUntil(
Func<bool> done, int timeoutMs = 5000, [CallerArgumentExpression(nameof(done))] string condition = null
)
{
var deadline = Stopwatch.StartNew();
while (!done())
{
if (deadline.ElapsedMilliseconds >= timeoutMs)
{
Assert.Fail($"Timed out after {timeoutMs} ms waiting for {condition}");
}
NetState.Slice();
Thread.Sleep(5);
}
}
/// <summary>Reads exactly <paramref name="length"/> bytes from the peer, slicing the server while it waits.</summary>
public static byte[] ReadAll(Socket client, int length, int timeoutMs = 10000)
{
var received = new byte[length];
var total = 0;
var deadline = Stopwatch.StartNew();
while (total < length)
{
if (deadline.ElapsedMilliseconds >= timeoutMs)
{
Assert.Fail($"Timed out after {timeoutMs} ms with {total} of {length} bytes received by the peer");
}
NetState.Slice();
// Poll takes microseconds, not milliseconds
if (client.Poll(50_000, SelectMode.SelectRead))
{
var read = client.Receive(received, total, length - total, SocketFlags.None);
Assert.True(read != 0, $"Peer saw EOF after {total} of {length} bytes");
total += read;
}
}
return received;
}
/// <summary>Waits for the server's FIN to reach the peer, failing if any data arrives first.</summary>
public static void WaitForEof(Socket client, int timeoutMs = 10000)
{
var scratch = new byte[64];
var deadline = Stopwatch.StartNew();
while (true)
{
if (deadline.ElapsedMilliseconds >= timeoutMs)
{
Assert.Fail($"Timed out after {timeoutMs} ms waiting for the peer to see EOF");
}
NetState.Slice();
if (client.Poll(50_000, SelectMode.SelectRead))
{
var read = client.Receive(scratch);
Assert.True(read == 0, $"Peer received {read} unexpected bytes while waiting for EOF");
return;
}
}
}
}