Files
agent-framework/dotnet/src/Microsoft.Agents.AI.Workflows/InProc/InProcessRunnerContext.cs
T
8ed2159c4b .NET: Workflow Outputs Overhaul: Support Tagging, Filtering Agent Outputs (#6045)
* test: reshuffle .NET Workflow tests in preparation for Outputs overhaul

Phase 1 of the .NET Workflows outputs overhaul (see
working/implementation-plan.md). Pure moves/renames in
dotnet/tests/Microsoft.Agents.AI.Workflows.UnitTests; no production code
changes, no new test cases. The split keeps each orchestration mode in
its own source file so the upcoming tag-aware and orchestration-default
test additions land on clean diffs.

Renames:
* WorkflowBuilderSmokeTests.cs -> WorkflowBuilderTests.cs (with class
  rename to match). The scope is no longer "smoke"-only once subsequent
  phases add tag-aware builder tests.
* InputWaiterAndOutputFilterTests.cs -> InputWaiterTests.cs +
  OutputFilterTests.cs. The file already declared the two test classes
  separately; this split simply gives each its own file so the
  output-filter cases have a dedicated home for tag-aware additions.

Split of AgentWorkflowBuilderTests.cs:
* AgentWorkflowBuilderTests.cs is now the outer
  `public static partial class AgentWorkflowBuilderTests` holding the
  shared test helpers (DoubleEchoAgent + session + WithBarrier variant,
  WorkflowRunResult, RunWorkflow* methods) bumped from `private` to
  `internal` so the new top-level GroupChatWorkflowBuilderTests in the
  same assembly can reach them.
* AgentWorkflowBuilder.SequentialTests.cs (nested SequentialTests):
  BuildSequential_InvalidArguments_Throws,
  BuildSequential_AgentsRunInOrderAsync.
* AgentWorkflowBuilder.ConcurrentTests.cs (nested ConcurrentTests):
  BuildConcurrent_InvalidArguments_Throws,
  BuildConcurrent_AgentsRunInParallelAsync.

Sequential and Concurrent are kept as nested classes because they're
modes of the same `AgentWorkflowBuilder` static factory and do not
produce dedicated builder types.

New file:
* GroupChatWorkflowBuilderTests.cs (top-level): the existing
  BuildGroupChat_* and GroupChatManager_* cases moved out of the old
  AgentWorkflowBuilderTests file. They exercise the
  `GroupChatWorkflowBuilder` type (returned by
  `AgentWorkflowBuilder.CreateGroupChatBuilderWith`), so a dedicated
  top-level test class - matching the convention reserved by the plan
  for HandoffWorkflowBuilderTests / MagenticWorkflowBuilderTests - is
  the right home. Cross-class helper references qualify with
  `AgentWorkflowBuilderTests.DoubleEchoAgent` and
  `AgentWorkflowBuilderTests.RunWorkflowAsync`.

The outer partial class is `static` (and nested classes carry the
instance test methods) because the outer holds only static helpers;
this satisfies CA1052 without suppressions and is invisible to xUnit
discovery, which finds tests on the nested classes as
`AgentWorkflowBuilderTests.SequentialTests.*` etc.

Validation: `dotnet build` clean on both target frameworks; all 547
tests in Microsoft.Agents.AI.Workflows.UnitTests pass on net10.0.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* feat: introduce OutputTag, Futures, and tag-aware WorkflowBuilder API

Phase 2 of the .NET Workflows outputs overhaul. Additive code change
only - no observable runtime behavior change. The runner still uses the
legacy bypass for AgentResponse / AgentResponseUpdate payloads, and the
new `Futures.EnableAgentResponseOutputTaggingAndFiltering` flag defaults
to false. Phase 3 will wire the flag into the runner; this commit only
introduces the types and the builder API.

New public surface:
* `OutputTag` (readonly struct): wraps a string Value with ordinal
  equality (IEquatable, GetHashCode, == / !=) so it can participate as a
  HashSet element. Internal ctor closes the set. One public singleton:
  `OutputTag.Intermediate`. Terminal / regular outputs carry no tag
  (empty Tags set). JSON-serialized as a bare string via
  [JsonConverter(typeof(OutputTagJsonConverter))], with the converter
  rehydrating to the well-known singleton on read.
* `Futures` (static class): hosts opt-in pre-GA behavior switches.
  First flag is `EnableAgentResponseOutputTaggingAndFiltering`; XML doc
  captures the v2.0.0 obsoletion / v3.0.0 removal lifecycle.
* `WorkflowOutputEvent.Tags`: `HashSet<OutputTag>` exposed directly
  (concrete collection, matches the JSON-serialization convention used
  for `WorkflowInfo.OutputExecutorIds`). Never null; empty for legacy /
  terminal events. New ctors take a single `OutputTag` or
  `IEnumerable<OutputTag>?`; the existing (data, executorId) ctor
  remains and produces an untagged event. `HasTag(OutputTag)` helper.
  `AgentResponseEvent` and `AgentResponseUpdateEvent` gain matching
  tag-accepting ctors forwarding to the base.
* `WorkflowOutputEventExtensions.IsIntermediate(this WorkflowOutputEvent)`:
  extension method returning `evt.HasTag(OutputTag.Intermediate)`. The
  preferred way to ask "is this an intermediate output?" without
  reaching into the Tags set.
* `WorkflowBuilder.WithOutputFrom(IEnumerable<ExecutorBinding>, OutputTag)`
  and `WorkflowBuilder.WithOutputFrom(ExecutorBinding, OutputTag)`:
  forward-looking tagged overloads. The IEnumerable form is the primary
  tagged surface; the single-executor form is a convenience for the
  common one-executor case. Currently usable for the
  `OutputTag.Intermediate` singleton; will become the primary surface
  once the `OutputTag` constructor is opened to user-defined tags in
  a future release. Callers in this release should prefer the
  intent-specific `WithIntermediateOutputFrom` extension for the
  intermediate case. Tags accumulate across repeated calls; same tag
  repeated dedupes via the HashSet.
* `WorkflowBuilderExtensions.WithIntermediateOutputFrom(this WorkflowBuilder, IEnumerable<ExecutorBinding>)`:
  helper that forwards to `WithOutputFrom(executors, OutputTag.Intermediate)`.
  Takes an IEnumerable (matching the tagged WithOutputFrom shape) -
  callers pass collection literals: `builder.WithIntermediateOutputFrom([a, b])`.
  XML doc remarks call out the Futures-flag interaction and the
  AIAgent-payload forwarding contract.

Internal shape changes:
* `WorkflowBuilder._outputExecutors`: HashSet<string> -> Dictionary<
  string, HashSet<OutputTag>>. The value set is empty for executors
  designated only via the untagged WithOutputFrom; contains Intermediate
  (and possibly future tags) otherwise.
* `Workflow.OutputExecutors`: HashSet<string> -> Dictionary<string,
  HashSet<OutputTag>>.
* `OutputFilter.CanOutput`: `Contains(id)` -> `ContainsKey(id)`.
* `WorkflowInfo.OutputExecutorIds`: HashSet<string> -> Dictionary<
  string, HashSet<OutputTag>>, with a custom JsonConverter that reads
  both the new map shape (`{id: ["intermediate", ...]}`) and the legacy
  array shape (`[id1, id2]`, where each id is treated as an untagged
  output). Always writes the map shape. IsMatch updated to compare
  per-id tag sets.

Tests landing in this commit (per the test-with-feature principle):
* `OutputTagTests.cs` (6 tests): KnownValues, EqualityIsOrdinalOnValue,
  DefaultStructValueIsDistinct (default(OutputTag) does not collide
  with the Intermediate singleton in a HashSet),
  GetHashCodeMatchesEquals, JsonConverter_RoundtripsValueAsString,
  ConstructorIsInternal (reflection-based assertion that the (string)
  ctor is `internal`).
* `WorkflowBuilderTests.cs` adds 7 new tests pinning the builder
  API contract: RegistersWithEmptyTagSet, AddsIntermediateTag,
  MultipleExecutorsAllUntagged, ThenIntermediate_AccumulatesTags,
  RepeatedDedupes, OnlyRegistersWithoutPriorWithOutputFrom,
  TracksExecutorBinding.
* `BackwardsCompatibility/JsonCheckpointSerializationTests.cs`
  (new folder + file, 5 tests): event-level ctor contract tests
  (single-tag, no-tag, multi-tag — the last with a custom tag);
  IsIntermediate() asserted; load-bearing JSON BC tests for
  `WorkflowInfo.OutputExecutorIds` -
  `WorkflowOutputExecutorsReadsLegacyArrayShape` (legacy ids map to
  empty tag sets) and `WorkflowOutputExecutorsWritesMapShape`.

The plan's three JSON round-trip tests for `WorkflowOutputEvent.Tags`
were dropped: `WorkflowEvent` is not currently a serialized checkpoint
shape (see the comment in WorkflowsJsonUtilities.cs about events not
being persisted), so there is no real back-compat surface to pin
through JSON. They are substituted with in-process ctor/property
round-trip tests that exercise the `Tags` / `HasTag` / `IsIntermediate`
contract.

Validation: full `Microsoft.Agents.AI.Workflows.UnitTests` suite runs
green on net10.0 (565 passing, 0 failing). Core library builds clean
on net472, netstandard2.0, net8.0, net9.0, and net10.0. Test project
builds clean on net472 + net10.0.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* feat: route AgentResponse(Update) through the output filter under a Futures flag

`InProcessRunnerContext.YieldOutputAsync` historically special-cased AgentResponse and
AgentResponseUpdate payloads: it built the typed event subclass and emitted it directly,
bypassing the output filter. Rewrites the method so that:

- When `Futures.EnableAgentResponseOutputTaggingAndFiltering` is `false` (the current
  default), AgentResponse(Update) keep the legacy bypass — emitted as
  AgentResponseEvent / AgentResponseUpdateEvent with no tags. Existing callers see no
  behavior change.
- When the flag is `true`, AIAgent payloads flow through the output filter just like
  every other payload type: undesignated sources are dropped, and the emitted event
  carries the source's tag set (empty for terminal `WithOutputFrom`, `{Intermediate}`
  for `WithIntermediateOutputFrom`, the set union when both designations apply).

Non-AIAgent (POCO) outputs also now carry the source's tag set on the emitted
WorkflowOutputEvent unconditionally — additive, since no existing assertion inspected
Tags. Subclass events (`AgentResponseEvent` / `AgentResponseUpdateEvent`) continue to
be emitted under both modes so `switch (evt) { case AgentResponseEvent: ... }`
consumer code keeps matching.

Adds `OutputFilter.TryGetTags` as the tag-aware lookup used by the runner.
`OutputFilter.CanOutput` is kept (still used by the existing sync tests in
`OutputFilterTests.cs`).

Tests
-----
- `Futures/Futures.AgentResponseOutputFilteringAndTaggingTests.cs` (new): the F1–F13
  matrix from the plan, covering every combination of `(flag on/off) × (designation)
  × (payload shape)`. Uses a `FuturesScope` IDisposable + a `FuturesSerial` xUnit
  collection (DisableParallelization = true) to keep the process-global flag from
  leaking across parallel tests.
- `OutputFilterTests.cs`: four new `Test_OutputFilter_…` cases for the `TryGetTags`
  surface (empty-tag-set for terminal designation, `{Intermediate}` for intermediate
  designation, union for accumulated designation, `false` for unregistered).

582/582 unit tests pass on net10.0 (565 baseline + 17 new).

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* feat: tag-aware defaults and designation API on orchestration builders

Aligns the .NET orchestration builders with Python's output / intermediate-output
distinction. Each builder either applies a Python-aligned default designation set or
replays the user's explicit `WithOutputFrom` / `WithIntermediateOutputFrom` calls,
never both.

Static `AgentWorkflowBuilder.BuildSequential` / `BuildConcurrent` apply defaults
unconditionally (no user-facing fluent surface to take control through):

- Sequential: terminal `end` + every agent designated intermediate.
- Concurrent: terminal `end` + every agent and per-agent accumulator designated
  intermediate.

The three fluent instance builders memoize agent-typed designation calls in a
`Dictionary<AIAgent, HashSet<OutputTag>>` (empty set = terminal-only, non-empty =
intermediate tag(s)) so repeated calls dedupe naturally. They replay the entries
at `Build()` time, suppressing defaults when any call has been made:

- `HandoffWorkflowBuilder` / `HandoffWorkflowBuilderCore<TBuilder>` (also picked up
  by the obsolete `HandoffsWorkflowBuilder` via inheritance).
  Default: terminal `HandoffEnd` + every handoff agent intermediate.
  (Bug fix: legacy code relied on `WithOutputFrom(end)` to bind `HandoffEnd`. The
  new explicit-designation path bypasses that, so `Build()` now calls
  `BindExecutor(end)` unconditionally to keep validation happy.)
- `GroupChatWorkflowBuilder` — default: terminal host + every participant intermediate.
- `MagenticWorkflowBuilder` — default: terminal orchestrator + every team member
  intermediate.

Designating a non-participant agent throws `InvalidOperationException`.

The bare `WorkflowBuilder` default is unchanged — only the orchestration-style
builders gain implicit defaults, matching the plan's non-goal.

Tests
-----
- `AgentWorkflowBuilder.SequentialTests` / `.ConcurrentTests`: one default-spec
  assertion each.
- `GroupChatWorkflowBuilderTests`: defaults-match-spec, explicit-replaces-defaults,
  non-participant throws.
- `HandoffWorkflowBuilderTests` (new file): same three.
- `MagenticWorkflowBuilderTests` (new file): same three.

593/593 unit tests pass on net10.0 (582 baseline + 11 new).

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* feat: WorkflowHostAgent forwards AgentResponseEvent unconditionally under Futures-on

Aligns the .NET Workflow-as-Agent surface with Python `as_agent`. Under
`Futures.EnableAgentResponseOutputTaggingAndFiltering = true`,
`WorkflowSession.InvokeStageAsync` now forwards `AgentResponseEvent`
unconditionally — joining `AgentResponseUpdateEvent` in ignoring the host's
`includeWorkflowOutputsInResponse` switch. That switch keeps governing the
generic `WorkflowOutputEvent` path for non-AIAgent payloads, where it is
further short-circuited by an `IsIntermediate()` check (tagged intermediate
outputs always surface).

Under Futures-off the legacy asymmetry is preserved: `AgentResponseUpdateEvent`
always forwarded, `AgentResponseEvent` gated by `includeWorkflowOutputsInResponse`.

Back-compat: with `Futures.EnableAgentResponseOutputTaggingAndFiltering` left at
its default `false`, observable behavior is identical to before.

`Futures` documentation gains a remark explaining the `Workflow.AsAIAgent()`
interaction in both flag states.

Runner fix
----------
`InProcessRunnerContext.YieldOutputAsync` now skips `Executor.CanOutput` for
AgentResponse-shaped payloads under both Futures branches. `AIAgentHostExecutor`
doesn't declare AgentResponse(Update) in its `Yields` set, so the historical
legacy bypass had silently skipped the check; Phase 3's Futures-on path was
running it and would reject AIAgent payloads. AIAgent-shaped payloads are now
always a valid output shape, matching the legacy bypass semantics.

Phase 4 follow-on
-----------------
Switched the three orchestration-builder designation-replay loops to iterate
`Dictionary.Keys` with a value lookup instead of constructing/destructuring
`KeyValuePair<,>`. Cleaner shape and avoids the netstandard2.0 / net472
`KeyValuePair<,>.Deconstruct` unavailability that surfaced when this branch
multi-TFM-built.

Tests
-----
`WorkflowHostSmokeTests.IntermediateForwarding` (new nested class, 6 tests):
- intermediate AgentResponse forwarded past the include-outputs gate (Futures on)
- terminal AgentResponse forwarded unconditionally (Futures on)
- terminal AgentResponse gated by include flag (Futures off, legacy)
- undesignated AIAgent executor emits no AgentResponseEvent under Futures-on
- legacy bypass still emits AgentResponseEvent under Futures-off
- intermediate tag is observable via `update.RawRepresentation`

The class joins the `FuturesSerial` xUnit collection so the process-global flag
is serialized against other Futures-toggling tests.

599/599 unit tests pass on net10.0 (593 baseline + 6 new).

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* feat: SequentialWorkflowBuilder and ConcurrentWorkflowBuilder, OrchestrationBuilderBase

Promotes the Sequential and Concurrent orchestration shapes to first-class fluent
builder classes, matching Handoff / GroupChat / Magentic. Users can call
`WithOutputFrom(agents)` / `WithIntermediateOutputFrom(agents)` to control which
agents are designated output / intermediate sources; when no designation call is
made, the Python-aligned defaults apply (terminal aggregator output + every agent
intermediate; Concurrent also tags per-agent accumulators).

`AgentWorkflowBuilder.BuildSequential(...)` and `BuildConcurrent(...)` are kept
and now delegate to the new builders; observable behavior unchanged. Five static
factories now mirror each other:

- `AgentWorkflowBuilder.CreateSequentialBuilderWith(params IEnumerable<AIAgent>)`
- `AgentWorkflowBuilder.CreateConcurrentBuilderWith(params IEnumerable<AIAgent>)`
- `AgentWorkflowBuilder.CreateHandoffBuilderWith(AIAgent)`        (already existed)
- `AgentWorkflowBuilder.CreateGroupChatBuilderWith(Func<...>)`    (already existed)
- `AgentWorkflowBuilder.CreateMagenticBuilderWith(AIAgent)`       (new)

OrchestrationBuilderBase
------------------------
New abstract `OrchestrationBuilderBase<TBuilder>` unifies the shared fluent
surface across all five orchestration builders: `WithName`, `WithDescription`,
`WithOutputFrom`, `WithIntermediateOutputFrom`, and the
`ApplyOutputDesignations(builder, agentMap, kind, applyDefaults)` helper that
either replays the user's designations or invokes the orchestration-specific
defaults.

Removes ~150 LOC of duplicated designation-management code from the four
non-Handoff builders, plus the equivalent from `HandoffWorkflowBuilderCore`.

Tests
-----
- New `SequentialWorkflowBuilderTests.cs` / `ConcurrentWorkflowBuilderTests.cs`
  (replace the old `AgentWorkflowBuilder.{Sequential,Concurrent}Tests.cs`
  nested-class files). Method names normalized to
  `Test_<BuilderType>_<Scenario>[Async]`.
- Shared helpers (`DoubleEchoAgent`, `DoubleEchoAgentWithBarrier`,
  `WorkflowRunResult`, `RunWorkflow*`) moved from the old
  `AgentWorkflowBuilderTests` partial class into a new
  `OrchestrationTestHelpers` static class in `OrchestrationTestHelpers.cs`.
  Downstream test files (Group Chat, Handoff, Sequential, Concurrent) updated
  to qualify with `OrchestrationTestHelpers.*`.
- A new `AgentWorkflowBuilderTests.cs` covers the static surface directly:
  `BuildSequential` / `BuildConcurrent` invariants and aggregator wiring, plus
  null-rejection + round-trip checks for every `Create*BuilderWith` factory.
- New AsAgent intermediate-suppression tests on a nested `AsAgentForwarding`
  class for each of Sequential and Concurrent: build with only the terminal
  agent designated via `WithOutputFrom`, run via `AsAIAgent(...)`, assert via
  `AgentResponseUpdate.AuthorName` that intermediate agents do not surface.
  Both join the `FuturesSerial` collection.
- New `Test_<Builder>_WithDescriptionPropagatesToWorkflow` smoke tests on
  Sequential and Concurrent (newly available via the base class).

625/625 unit tests pass on net10.0.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* chore: dotnet format

* fixup: encoding

* fixup: charset

* fixup: Updates for PR feedback

* fixup: format

* fixup: merge issue

* Fix intermediate filtering on .AsAgent()

* fix filter logic

* fix: Revert logic change and add comments

---------

Co-authored-by: Jacob Alber <jalber@lokitoth.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
2026-05-28 21:26:31 +00:00

524 lines
22 KiB
C#

// Copyright (c) Microsoft. All rights reserved.
using System;
using System.Collections.Concurrent;
using System.Collections.Generic;
using System.Diagnostics;
using System.Diagnostics.CodeAnalysis;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;
using Microsoft.Agents.AI.Workflows.Checkpointing;
using Microsoft.Agents.AI.Workflows.Execution;
using Microsoft.Agents.AI.Workflows.Observability;
using Microsoft.Agents.AI.Workflows.Specialized;
using Microsoft.Extensions.Logging;
using Microsoft.Shared.Diagnostics;
using OpenTelemetry;
using OpenTelemetry.Context.Propagation;
namespace Microsoft.Agents.AI.Workflows.InProc;
internal sealed class InProcessRunnerContext : IRunnerContext
{
private int _runEnded;
private readonly string _sessionId;
private readonly Workflow _workflow;
private readonly object? _previousOwnership;
private bool _ownsWorkflow;
private readonly EdgeMap _edgeMap;
private readonly OutputFilter _outputFilter;
private StepContext _nextStep = new();
private readonly ConcurrentDictionary<string, Task<Executor>> _executors = new();
private readonly ConcurrentQueue<Func<ValueTask>> _queuedExternalDeliveries = new();
private readonly ConcurrentDictionary<string, ISuperStepRunner> _joinedSubworkflowRunners = new();
private readonly ConcurrentDictionary<string, ExternalRequest> _externalRequests = new();
public InProcessRunnerContext(
Workflow workflow,
string sessionId,
bool checkpointingEnabled,
IEventSink outgoingEvents,
IStepTracer? stepTracer,
object? existingOwnershipSignoff = null,
bool subworkflow = false,
bool enableConcurrentRuns = false,
ILogger? logger = null)
{
if (enableConcurrentRuns)
{
workflow.CheckOwnership(existingOwnershipSignoff: existingOwnershipSignoff);
}
else
{
workflow.TakeOwnership(this, existingOwnershipSignoff: existingOwnershipSignoff);
this._previousOwnership = existingOwnershipSignoff;
this._ownsWorkflow = true;
}
this._workflow = workflow;
this._sessionId = sessionId;
this._edgeMap = new(this, this._workflow, stepTracer);
this._outputFilter = new(workflow);
this.IsCheckpointingEnabled = checkpointingEnabled;
this.ConcurrentRunsEnabled = enableConcurrentRuns;
this.OutgoingEvents = outgoingEvents;
}
public WorkflowTelemetryContext TelemetryContext => this._workflow.TelemetryContext;
public IExternalRequestSink RegisterPort(string executorId, RequestPort port)
{
if (!this._edgeMap.TryRegisterPort(this, executorId, port))
{
throw new InvalidOperationException($"A port with ID {port.Id} already exists.");
}
return this;
}
public async ValueTask<Executor> EnsureExecutorAsync(string executorId, IStepTracer? tracer, CancellationToken cancellationToken = default)
{
this.CheckEnded();
Task<Executor> executorTask = this._executors.GetOrAdd(executorId, CreateExecutorAsync);
async Task<Executor> CreateExecutorAsync(string id)
{
if (!this._workflow.ExecutorBindings.TryGetValue(executorId, out var registration))
{
throw new InvalidOperationException($"Executor with ID '{executorId}' is not registered.");
}
Executor executor = await registration.CreateInstanceAsync(this._sessionId).ConfigureAwait(false);
executor.AttachRequestContext(this.BindExternalRequestContext(executorId));
await executor.InitializeAsync(this.BindWorkflowContext(executorId), cancellationToken: cancellationToken)
.ConfigureAwait(false);
tracer?.TraceActivated(executorId);
if (executor is RequestInfoExecutor requestInputExecutor)
{
requestInputExecutor.AttachRequestSink(this);
}
if (executor is WorkflowHostExecutor workflowHostExecutor)
{
await workflowHostExecutor.AttachSuperStepContextAsync(this).ConfigureAwait(false);
}
return executor;
}
return await executorTask.ConfigureAwait(false);
}
public async ValueTask<IEnumerable<Type>> GetStartingExecutorInputTypesAsync(CancellationToken cancellationToken = default)
{
Executor startingExecutor = await this.EnsureExecutorAsync(this._workflow.StartExecutorId, tracer: null, cancellationToken)
.ConfigureAwait(false);
return startingExecutor.InputTypes;
}
public ValueTask AddExternalMessageAsync(object message, Type declaredType)
{
this.CheckEnded();
Throw.IfNull(message);
this._queuedExternalDeliveries.Enqueue(PrepareExternalDeliveryAsync);
return default;
async ValueTask PrepareExternalDeliveryAsync()
{
DeliveryMapping? maybeMapping =
await this._edgeMap.PrepareDeliveryForInputAsync(new(message, ExecutorIdentity.None, declaredType))
.ConfigureAwait(false);
maybeMapping?.MapInto(this._nextStep);
}
}
public ValueTask AddExternalResponseAsync(ExternalResponse response)
{
this.CheckEnded();
Throw.IfNull(response);
this._queuedExternalDeliveries.Enqueue(PrepareExternalDeliveryAsync);
return default;
async ValueTask PrepareExternalDeliveryAsync()
{
if (!this.CompleteRequest(response.RequestId))
{
throw new InvalidOperationException($"No pending request with ID {response.RequestId} found in the workflow context.");
}
DeliveryMapping? maybeMapping =
await this._edgeMap.PrepareDeliveryForResponseAsync(response)
.ConfigureAwait(false);
maybeMapping?.MapInto(this._nextStep);
}
}
public bool HasQueuedExternalDeliveries => !this._queuedExternalDeliveries.IsEmpty;
public bool JoinedRunnersHaveActions => this._joinedSubworkflowRunners.Values.Any(runner => runner.HasUnprocessedMessages);
public bool NextStepHasActions => this._nextStep.HasMessages ||
this.HasQueuedExternalDeliveries ||
this.JoinedRunnersHaveActions;
public bool HasUnservicedRequests => !this._externalRequests.IsEmpty ||
this._joinedSubworkflowRunners.Values.Any(runner => runner.HasUnservicedRequests);
public async ValueTask<StepContext> AdvanceAsync(CancellationToken cancellationToken = default)
{
this.CheckEnded();
while (this._queuedExternalDeliveries.TryDequeue(out var deliveryPrep))
{
// It's important we do not try to run these in parallel, because they may be modifying
// inner edge state, etc.
await deliveryPrep().ConfigureAwait(false);
}
return Interlocked.Exchange(ref this._nextStep, new StepContext());
}
public ValueTask AddEventAsync(WorkflowEvent workflowEvent, CancellationToken cancellationToken = default)
{
this.CheckEnded();
return this.OutgoingEvents.EnqueueAsync(workflowEvent);
}
public async ValueTask SendMessageAsync(string sourceId, object message, string? targetId = null, CancellationToken cancellationToken = default)
{
using Activity? activity = this._workflow.TelemetryContext.StartMessageSendActivity(sourceId, targetId, message);
// Create a carrier for trace context propagation
var traceContext = activity is null ? null : new Dictionary<string, string>();
if (traceContext is not null)
{
// Inject the current activity context into the carrier
Propagators.DefaultTextMapPropagator.Inject(
new PropagationContext(activity?.Context ?? default, Baggage.Current),
traceContext,
(carrier, key, value) => carrier[key] = value);
}
this.CheckEnded();
Debug.Assert(this._executors.ContainsKey(sourceId));
Executor source = await this.EnsureExecutorAsync(sourceId, tracer: null, cancellationToken).ConfigureAwait(false);
TypeId? declaredType = source.Protocol.SendTypeTranslator.GetDeclaredType(message.GetType());
if (declaredType is null)
{
throw new InvalidOperationException($"Executor '{sourceId}' cannot send messages of type '{message.GetType().FullName}'.");
}
MessageEnvelope envelope = new(message, sourceId, declaredType, targetId: targetId, traceContext: traceContext);
if (this._workflow.Edges.TryGetValue(sourceId, out HashSet<Edge>? edges))
{
foreach (Edge edge in edges)
{
DeliveryMapping? maybeMapping =
await this._edgeMap.PrepareDeliveryForEdgeAsync(edge, envelope, cancellationToken)
.ConfigureAwait(false);
maybeMapping?.MapInto(this._nextStep);
}
}
}
private async ValueTask YieldOutputAsync(string sourceId, object output, CancellationToken cancellationToken = default)
{
this.CheckEnded();
Throw.IfNull(output);
bool isAgentResponseShaped = output is AgentResponse or AgentResponseUpdate;
if (isAgentResponseShaped && !Futures.EnableAgentResponseOutputTaggingAndFiltering)
{
// Legacy bypass: AgentResponse/AgentResponseUpdate skip the output filter and are
// emitted as their typed event subclasses with no tags. Preserved verbatim for
// back-compat; once Futures.EnableAgentResponseOutputTaggingAndFiltering becomes the
// default in v2.0.0, this branch goes away.
WorkflowEvent typedEvent = output switch
{
AgentResponseUpdate u => new AgentResponseUpdateEvent(sourceId, u),
AgentResponse r => new AgentResponseEvent(sourceId, r),
_ => throw new InvalidOperationException("Unexpected AIAgent-shaped payload type."),
};
await this.AddEventAsync(typedEvent, cancellationToken).ConfigureAwait(false);
return;
}
Executor sourceExecutor = await this.EnsureExecutorAsync(sourceId, tracer: null, cancellationToken).ConfigureAwait(false);
if (!isAgentResponseShaped && !sourceExecutor.CanOutput(output.GetType()))
{
// AIAgent-shaped payloads bypass the per-executor declared-yield check (matching the
// legacy bypass branch above). The AIAgent host executor relays the agent's output
// without declaring AgentResponse(Update) in its Yields set, so a CanOutput probe
// here would always reject — but those payloads are always a valid output shape.
throw new InvalidOperationException($"Cannot output object of type {output.GetType().Name}. Expecting one of [{string.Join(", ", sourceExecutor.OutputTypes)}].");
}
if (!this._outputFilter.TryGetTags(sourceId, out HashSet<OutputTag>? tags))
{
// Not designated as an output source — drop silently.
return;
}
WorkflowOutputEvent evt = output switch
{
AgentResponseUpdate u => new AgentResponseUpdateEvent(sourceId, u, tags),
AgentResponse r => new AgentResponseEvent(sourceId, r, tags),
_ => new WorkflowOutputEvent(output, sourceId, tags),
};
await this.AddEventAsync(evt, cancellationToken).ConfigureAwait(false);
}
public IExternalRequestContext BindExternalRequestContext(string executorId)
{
this.CheckEnded();
return new BoundExternalRequestContext(this, executorId);
}
public IWorkflowContext BindWorkflowContext(string executorId, Dictionary<string, string>? traceContext = null)
{
this.CheckEnded();
return new BoundWorkflowContext(this, executorId, traceContext);
}
public ValueTask PostAsync(ExternalRequest request)
{
this.CheckEnded();
if (!this._externalRequests.TryAdd(request.RequestId, request))
{
throw new ArgumentException($"Pending request with id '{request.RequestId}' already exists.");
}
return this.AddEventAsync(new RequestInfoEvent(request));
}
public bool CompleteRequest(string requestId)
{
this.CheckEnded();
return this._externalRequests.TryRemove(requestId, out _);
}
internal bool TryGetResponsePortExecutorId(string portId, [NotNullWhen(true)] out string? executorId)
=> this._edgeMap.TryGetResponsePortExecutorId(portId, out executorId);
private IEventSink OutgoingEvents { get; }
internal StateManager StateManager { get; } = new();
private sealed class BoundExternalRequestContext(
InProcessRunnerContext RunnerContext,
string ExecutorId) : IExternalRequestContext
{
public IExternalRequestSink RegisterPort(RequestPort port)
{
return RunnerContext.RegisterPort(ExecutorId, port);
}
}
private sealed class BoundWorkflowContext(
InProcessRunnerContext RunnerContext,
string ExecutorId,
Dictionary<string, string>? traceContext) : IWorkflowContext
{
public ValueTask AddEventAsync(WorkflowEvent workflowEvent, CancellationToken cancellationToken = default) => RunnerContext.AddEventAsync(workflowEvent, cancellationToken);
public ValueTask SendMessageAsync(object message, string? targetId = null, CancellationToken cancellationToken = default)
{
return RunnerContext.SendMessageAsync(ExecutorId, Throw.IfNull(message), targetId, cancellationToken);
}
public ValueTask YieldOutputAsync(object output, CancellationToken cancellationToken = default)
{
return RunnerContext.YieldOutputAsync(ExecutorId, Throw.IfNull(output), cancellationToken);
}
public ValueTask RequestHaltAsync() => this.AddEventAsync(new RequestHaltEvent());
public ValueTask<T?> ReadStateAsync<T>(string key, string? scopeName = null, CancellationToken cancellationToken = default)
=> RunnerContext.StateManager.ReadStateAsync<T>(ExecutorId, scopeName, key);
[return: NotNull]
public ValueTask<T> ReadOrInitStateAsync<T>(string key, Func<T> initialStateFactory, string? scopeName = null, CancellationToken cancellationToken = default)
=> RunnerContext.StateManager.ReadOrInitStateAsync(ExecutorId, scopeName, key, initialStateFactory);
public ValueTask<HashSet<string>> ReadStateKeysAsync(string? scopeName = null, CancellationToken cancellationToken = default)
=> RunnerContext.StateManager.ReadKeysAsync(ExecutorId, scopeName);
public ValueTask QueueStateUpdateAsync<T>(string key, T? value, string? scopeName = null, CancellationToken cancellationToken = default)
=> RunnerContext.StateManager.WriteStateAsync(ExecutorId, scopeName, key, value);
public ValueTask QueueClearScopeAsync(string? scopeName = null, CancellationToken cancellationToken = default)
=> RunnerContext.StateManager.ClearStateAsync(ExecutorId, scopeName);
public IReadOnlyDictionary<string, string>? TraceContext => traceContext;
public bool ConcurrentRunsEnabled => RunnerContext.ConcurrentRunsEnabled;
}
public bool IsCheckpointingEnabled { get; }
public bool ConcurrentRunsEnabled { get; }
internal Task PrepareForCheckpointAsync(CancellationToken cancellationToken = default)
{
this.CheckEnded();
return Task.WhenAll(this._executors.Values.Select(InvokeCheckpointingAsync));
async Task InvokeCheckpointingAsync(Task<Executor> executorTask)
{
Executor executor = await executorTask.ConfigureAwait(false);
await executor.OnCheckpointingAsync(this.BindWorkflowContext(executor.Id), cancellationToken).ConfigureAwait(false);
}
}
internal Task NotifyCheckpointLoadedAsync(CancellationToken cancellationToken = default)
{
this.CheckEnded();
return Task.WhenAll(this._executors.Values.Select(InvokeCheckpointRestoredAsync));
async Task InvokeCheckpointRestoredAsync(Task<Executor> executorTask)
{
Executor executor = await executorTask.ConfigureAwait(false);
await executor.OnCheckpointRestoredAsync(this.BindWorkflowContext(executor.Id), cancellationToken).ConfigureAwait(false);
}
}
internal ValueTask<RunnerStateData> ExportStateAsync()
{
this.CheckEnded();
Dictionary<string, List<PortableMessageEnvelope>> queuedMessages = this._nextStep.ExportMessages();
RunnerStateData result = new(instantiatedExecutors: [.. this._executors.Keys],
queuedMessages,
outstandingRequests: [.. this._externalRequests.Values]);
return new(result);
}
internal async ValueTask RepublishUnservicedRequestsAsync(CancellationToken cancellationToken = default)
{
this.CheckEnded();
if (this.HasUnservicedRequests)
{
foreach (string requestId in this._externalRequests.Keys)
{
await this.AddEventAsync(new RequestInfoEvent(this._externalRequests[requestId]), cancellationToken)
.ConfigureAwait(false);
}
}
}
internal async ValueTask ImportStateAsync(Checkpoint checkpoint)
{
this.CheckEnded();
RunnerStateData importedState = checkpoint.RunnerData;
Task<Executor>[] executorTasks = importedState.InstantiatedExecutors
.Where(id => !this._executors.ContainsKey(id))
.Select(id => this.EnsureExecutorAsync(id, tracer: null).AsTask())
.ToArray();
// Discard queued external deliveries from the superseded timeline so a runtime
// restore cannot apply stale responses after importing the checkpoint state.
while (this._queuedExternalDeliveries.TryDequeue(out _))
{
}
this._nextStep = new StepContext();
this._nextStep.ImportMessages(importedState.QueuedMessages);
this._externalRequests.Clear();
foreach (ExternalRequest request in importedState.OutstandingRequests)
{
// TODO: Reduce the amount of data we need to store in the checkpoint by not storing the entire request object.
// For example, the Port object is not needed - we should be able to reconstruct it from the ID and the workflow
// definition.
this._externalRequests[request.RequestId] = request;
}
await Task.WhenAll(executorTasks).ConfigureAwait(false);
}
[SuppressMessage("Maintainability", "CA1513:Use ObjectDisposedException throw helper",
Justification = "Does not exist in NetFx 4.7.2")]
internal void CheckEnded()
{
if (Volatile.Read(ref this._runEnded) == 1)
{
throw new InvalidOperationException($"Workflow run for session '{this._sessionId}' has been ended. Please start a new Run or StreamingRun.");
}
}
public async ValueTask EndRunAsync()
{
if (Interlocked.Exchange(ref this._runEnded, 1) == 0)
{
foreach (string executorId in this._executors.Keys)
{
Task<Executor> executorTask = this._executors[executorId];
Executor executor = await executorTask.ConfigureAwait(false);
if (executor is IAsyncDisposable asyncDisposable)
{
await asyncDisposable.DisposeAsync().ConfigureAwait(false);
}
else if (executor is IDisposable disposable)
{
disposable.Dispose();
}
}
if (this._ownsWorkflow)
{
await this._workflow.ReleaseOwnershipAsync(this, this._previousOwnership).ConfigureAwait(false);
this._ownsWorkflow = false;
}
}
}
public IEnumerable<ISuperStepRunner> JoinedSubworkflowRunners => this._joinedSubworkflowRunners.Values;
public ValueTask<string> AttachSuperstepAsync(ISuperStepRunner superStepRunner, CancellationToken cancellationToken = default)
{
// This needs to be a thread-safe ordered collection because we can potentially instantiate executors
// in parallel, which means multiple sub-workflows could be attaching at the same time.
string joinId;
do
{
joinId = Guid.NewGuid().ToString("N");
} while (!this._joinedSubworkflowRunners.TryAdd(joinId, superStepRunner));
return default;
}
public ValueTask<bool> DetachSuperstepAsync(string joinId) => new(this._joinedSubworkflowRunners.TryRemove(joinId, out _));
ValueTask ISuperStepJoinContext.ForwardWorkflowEventAsync(WorkflowEvent workflowEvent, CancellationToken cancellationToken)
=> this.AddEventAsync(workflowEvent, cancellationToken);
ValueTask ISuperStepJoinContext.SendMessageAsync<TMessage>(string senderId, [DisallowNull] TMessage message, CancellationToken cancellationToken)
=> this.SendMessageAsync(senderId, Throw.IfNull(message), cancellationToken: cancellationToken);
ValueTask ISuperStepJoinContext.YieldOutputAsync<TOutput>(string senderId, [DisallowNull] TOutput output, CancellationToken cancellationToken)
=> this.YieldOutputAsync(senderId, Throw.IfNull(output), cancellationToken);
}