Files
ClaudeDo/src/ClaudeDo.Worker/State/TaskStateService.cs
Mika Kuns 8823265e5a refactor(worker/state): introduce TaskStateService and route mutations through it
Slice 2 of the worker state consolidation refactor (spec sections 2 and 8).

Adds Worker/State/ITaskStateService + TaskStateService as the single component
that mutates Status, PlanningPhase, and BlockedByTaskId. Each transition is one
atomic ExecuteUpdate with a WHERE filter on the expected source status, so
parallel claims are TOCTOU-free. Side effects (queue wake on -> Queued, hub
TaskUpdated broadcast, chain advance + parent completion on terminal child)
are owned by the service so callers no longer need to remember them.

Migrated callers (mechanical, behavior preserved):
- TaskRunner: HandleSuccess/HandleFailure/MarkFailed/RunAsync/ContinueAsync
- StaleTaskRecovery: bulk recover stale Running tasks
- TaskResetService: status flip (worktree cleanup stays in service)
- PlanningSessionManager.StartAsync: status flip via state, token write via repo
- PlanningChainCoordinator.OnChildFinishedAsync: routes the next-sibling write
  through state.UnblockAsync (Slice 4 finishes the rewrite)
- ExternalMcpService.UpdateTaskStatus: Queued case via state.EnqueueAsync

Repo Mark*Async helpers (MarkRunning/MarkDone/MarkFailed/FlipAllRunningToFailed)
are now internal; ClaudeDo.Data grants InternalsVisibleTo to ClaudeDo.Worker
and ClaudeDo.Worker.Tests for the existing repo-level tests.

DI: TaskStateService is registered as Singleton in both the main app and the
external-MCP app; the queue-wake delegate captures sp -> QueueService.WakeQueue
to break the TaskStateService -> QueueService -> TaskRunner -> TaskStateService
construction cycle. PlanningChainCoordinator takes Func<ITaskStateService> for
the same reason; Slice 3 will replace both with IQueueWaker.

Tests: TaskStateServiceTests covers happy + reject for every transition, the
parallel StartRunningAsync claim race, child-terminal chain advancement, and
stale recovery. Existing service/repo tests are updated to construct the new
state-service via a TaskStateServiceBuilder helper. Pre-existing constructor
drift in QueueService/ExternalMcp/PlanningHub tests is patched to keep the
test project building (the surrounding test logic is otherwise untouched).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 11:31:57 +02:00

259 lines
10 KiB
C#

using ClaudeDo.Data;
using ClaudeDo.Data.Models;
using ClaudeDo.Data.Repositories;
using ClaudeDo.Worker.Hub;
using ClaudeDo.Worker.Planning;
using Microsoft.EntityFrameworkCore;
using TaskStatus = ClaudeDo.Data.Models.TaskStatus;
namespace ClaudeDo.Worker.State;
public sealed class TaskStateService : ITaskStateService
{
private readonly IDbContextFactory<ClaudeDoDbContext> _dbFactory;
private readonly HubBroadcaster _broadcaster;
private readonly Action _wakeQueue;
private readonly PlanningChainCoordinator _chain;
private readonly ILogger<TaskStateService> _logger;
public TaskStateService(
IDbContextFactory<ClaudeDoDbContext> dbFactory,
HubBroadcaster broadcaster,
Action wakeQueue,
PlanningChainCoordinator chain,
ILogger<TaskStateService> logger)
{
_dbFactory = dbFactory;
_broadcaster = broadcaster;
_wakeQueue = wakeQueue;
_chain = chain;
_logger = logger;
}
public async Task<TransitionResult> EnqueueAsync(string taskId, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == taskId && t.Status != TaskStatus.Running)
.ExecuteUpdateAsync(s => s.SetProperty(t => t.Status, TaskStatus.Queued), ct);
if (affected == 0)
return new TransitionResult(false, "Task not found or already running.");
_wakeQueue();
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> StartRunningAsync(string taskId, DateTime startedAt, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == taskId && t.Status != TaskStatus.Running)
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Running)
.SetProperty(t => t.StartedAt, startedAt), ct);
if (affected == 0)
return new TransitionResult(false, "Task already running or not found.");
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> CompleteAsync(string taskId, DateTime finishedAt, string? result, CancellationToken ct)
{
await using (var ctx = await _dbFactory.CreateDbContextAsync(ct))
{
var affected = await ctx.Tasks
.Where(t => t.Id == taskId && t.Status == TaskStatus.Running)
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Done)
.SetProperty(t => t.FinishedAt, finishedAt)
.SetProperty(t => t.Result, result), ct);
if (affected == 0)
return new TransitionResult(false, "Task not running; cannot complete.");
}
await OnChildTerminalAsync(taskId, TaskStatus.Done);
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> FailAsync(string taskId, DateTime finishedAt, string? error, CancellationToken ct)
{
await using (var ctx = await _dbFactory.CreateDbContextAsync(ct))
{
var affected = await ctx.Tasks
.Where(t => t.Id == taskId && t.Status != TaskStatus.Done)
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Failed)
.SetProperty(t => t.FinishedAt, finishedAt)
.SetProperty(t => t.Result, error), ct);
if (affected == 0)
return new TransitionResult(false, "Task already done; cannot fail.");
}
await OnChildTerminalAsync(taskId, TaskStatus.Failed);
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> CancelAsync(string taskId, DateTime finishedAt, CancellationToken ct)
{
await using (var ctx = await _dbFactory.CreateDbContextAsync(ct))
{
var affected = await ctx.Tasks
.Where(t => t.Id == taskId &&
(t.Status == TaskStatus.Running || t.Status == TaskStatus.Queued))
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Cancelled)
.SetProperty(t => t.FinishedAt, finishedAt), ct);
if (affected == 0)
return new TransitionResult(false, "Task not in cancellable state.");
}
await OnChildTerminalAsync(taskId, TaskStatus.Cancelled);
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> ResetToIdleAsync(string taskId, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == taskId && t.Status != TaskStatus.Running)
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Idle)
.SetProperty(t => t.StartedAt, (DateTime?)null)
.SetProperty(t => t.FinishedAt, (DateTime?)null)
.SetProperty(t => t.Result, (string?)null), ct);
if (affected == 0)
return new TransitionResult(false, "Task is running; cannot reset.");
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> StartPlanningAsync(string parentId, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == parentId &&
(t.Status == TaskStatus.Manual || t.Status == TaskStatus.Idle))
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Planning)
.SetProperty(t => t.PlanningPhase, PlanningPhase.Active), ct);
if (affected == 0)
return new TransitionResult(false, "Task not in plannable state.");
await _broadcaster.TaskUpdated(parentId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> FinalizePlanningAsync(string parentId, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == parentId && t.PlanningPhase == PlanningPhase.Active)
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.PlanningPhase, PlanningPhase.Finalized)
.SetProperty(t => t.PlanningFinalizedAt, DateTime.UtcNow), ct);
if (affected == 0)
return new TransitionResult(false, "No active planning session.");
await _broadcaster.TaskUpdated(parentId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> BlockOnAsync(string taskId, string predecessorTaskId, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == taskId)
.ExecuteUpdateAsync(s => s.SetProperty(t => t.BlockedByTaskId, predecessorTaskId), ct);
if (affected == 0)
return new TransitionResult(false, "Task not found.");
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<TransitionResult> UnblockAsync(string taskId, CancellationToken ct)
{
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
var affected = await ctx.Tasks
.Where(t => t.Id == taskId)
.ExecuteUpdateAsync(s => s.SetProperty(t => t.BlockedByTaskId, (string?)null), ct);
if (affected == 0)
return new TransitionResult(false, "Task not found.");
// Bridge to legacy chain layout: a Waiting predecessor-blocked sibling becomes Queued
// when its predecessor finishes. New layout (post-Slice 4) stores siblings as
// Status=Queued + BlockedByTaskId set, so this is a no-op for them.
await ctx.Tasks
.Where(t => t.Id == taskId && t.Status == TaskStatus.Waiting)
.ExecuteUpdateAsync(s => s.SetProperty(t => t.Status, TaskStatus.Queued), ct);
_wakeQueue();
await _broadcaster.TaskUpdated(taskId);
return new TransitionResult(true, null);
}
public async Task<int> RecoverStaleRunningAsync(string reason, CancellationToken ct)
{
var resultText = "[stale] " + reason;
var now = DateTime.UtcNow;
await using var ctx = await _dbFactory.CreateDbContextAsync(ct);
return await ctx.Tasks
.Where(t => t.Status == TaskStatus.Running)
.ExecuteUpdateAsync(s => s
.SetProperty(t => t.Status, TaskStatus.Failed)
.SetProperty(t => t.FinishedAt, now)
.SetProperty(t => t.Result, resultText), ct);
}
private async Task OnChildTerminalAsync(string taskId, TaskStatus finalStatus)
{
// Terminal child writes are best-effort and use CancellationToken.None so the
// task lifecycle is never left partially completed because a caller cancelled.
string? parentId;
await using (var ctx = await _dbFactory.CreateDbContextAsync(CancellationToken.None))
{
parentId = await ctx.Tasks
.AsNoTracking()
.Where(t => t.Id == taskId)
.Select(t => t.ParentTaskId)
.FirstOrDefaultAsync(CancellationToken.None);
}
if (parentId is null) return;
try
{
await _chain.OnChildFinishedAsync(taskId, finalStatus, CancellationToken.None);
}
catch (Exception ex)
{
_logger.LogWarning(ex, "PlanningChain advance failed for {TaskId}", taskId);
}
try
{
await using var ctx = await _dbFactory.CreateDbContextAsync(CancellationToken.None);
await new TaskRepository(ctx).TryCompleteParentAsync(parentId, CancellationToken.None);
}
catch (Exception ex)
{
_logger.LogWarning(ex, "TryCompleteParent failed for {ParentId}", parentId);
}
}
}