mirror of
https://github.com/infiniflow/ragflow.git
synced 2026-07-16 20:57:21 +08:00
## Summary
Aligns the **Go agent runtime/canvas/components/tools** behavior with
the **Python `agent/` implementation** so the same stored canvas DSL
produces the same execution result on either side. Every component,
tool, and runtime primitive in `internal/agent/` is now driven by the
same semantics as its Python counterpart — variable resolution, template
substitution, control flow, error reporting, retry/cancel, and stream
event shapes.
The **retrieval component is the one explicit exception** in this PR. It
is being reworked in a separate change and is excluded from this
alignment pass; the wrapper slot (`universe_a_wrappers.go →
newRetrievalComponent`) is preserved.
## Scope of alignment
### Components (all aligned with `agent/component/`)
`Begin` · `Message` · `LLM` (incl. ChatTemplateKwargs,
MessageHistoryWindowSize, VisualFiles, Cite, OutputStructure,
JSONOutput, TopP, MaxRetries, DelayAfterError, credentials) · `Agent`
(react + tool artifact capture + `Reset()` interface-assert) · `Switch`
(12/12 operators, Python-equivalent semantics) · `Categorize` · `Invoke`
· `Iteration` · `Loop` (macro-expansion through `workflowx.AddLoopNode`)
· `UserFillUp` (Python-equivalent interrupt/resume via eino
`compose.Interrupt`/`ResumeWithData`) · `FillUp` · `DataOperations` ·
`ListOperations` · `StringTransform` · `VariableAggregator` ·
`VariableAssigner` · `Browser` (full stagehand runtime parity) ·
`DocsGenerator` · `ExcelProcessor`.
### Tools (all aligned with `agent/tools/`)
`Retrieval` (wrapper slot only — logic out of scope) · `MCPToolAdapter`
(streamable-HTTP) · `CodeExec` (sandbox bridge with
`code_exec_contract.go` matching Python contract) · `AkShare` · `ArXiv`
· `Crawler` · `DeepL` · `DuckDuckGo` · `Email` · `ExeSQL` · `GitHub` ·
`Google` · `GoogleScholar` · `Jin10` · `PubMed` · `QWeather` · `SearXNG`
· `Tavily` · `Tushare` · `Wencai` · `Wikipedia` · `YahooFinance` —
uniform `eino tool.InvokableTool` interface, SSRF protection, shared
HTTP client.
### Canvas execution engine (`internal/agent/canvas/`)
Aligned with Python's `agent/canvas.py`:
- **Scheduler** (`scheduler.go`): state pre/post handlers, node lambdas,
per-component timeout resolver (4-level: per-class env → per-class table
→ uniform env → 600s fallback), `legacyNoOpNames`.
- **Loop subgraph** (`loop_subgraph.go`): Python-equivalent
`AddLoopNode` macro expansion + condition translation.
- **Multibranch** (`multibranch.go`): `Switch` / `Categorize` routing
via `compose.NewGraphMultiBranch` — same branch selection semantics as
Python.
- **Parallel subgraph** (`parallel_subgraph.go`): matches Python's
parallel fan-out contract.
- **Interrupt/Resume** (`interrupt_resume.go`): `UserFillUpNodeBody` /
`IsInterruptError` / `ExtractInterruptContexts` — replaces the
deprecated Python sentinel chain with eino's native interrupt API,
preserving the same external behavior.
- **Checkpoint** (`checkpoint_store.go`): `RedisCheckPointStore`
Get/Set/Delete, with business metadata (status / canvas_id /
parent_run_id) on a parallel Redis Hash.
- **RunTracker** (`run_tracker.go`): Start / MarkSucceeded / MarkFailed
/ MarkCancelled / AttachCheckpoint — same lifecycle as the Python run
record.
- **Cancel** (`cancel.go`): Redis pub/sub watch.
- **Stream** (`stream.go`): SSE channel with `messages` / `waiting` /
`errors` / `done` events, same shape as Python's `agent.canvas.RunEvent`
payload.
### DSL bridge (`internal/agent/dsl/`)
- `normalize.go`: v1↔v2 collapsed into a single wire format — Python and
Go consume the same stored JSON.
- `reset.go`: per-run state reset matches Python's `Canvas.reset()`
semantics.
- Testdata mirrors Python's `agent_msg.json` / `all.json` / etc.
### Runtime (`internal/agent/runtime/`)
- `CanvasState` / `NewCanvasState` / `GetVar` / `SetVar` / `ReadVars`:
same `{{cpn_id@param}}` resolution model.
- `ResolveTemplate` (regex fast path + gonja fallback) — Python
Jinja-style semantics.
- `selector.go`, `metrics.go`, `component.go`: shared runtime contracts.
## Out of scope (intentionally)
- **`Retrieval` component logic** — wrapped only; full parity lands in a
follow-up PR.
- **Frontend** — only minor dsl-bridge / canvas UX fixes ride along.
- **CLI / admin / model registry** — orthogonal to agent behavior.
## How alignment is verified
`internal/service/agent_run_e2e_test.go` exercises the **full production
chain** against real Python-shaped DSL fixtures:
```
loadCanvasForUser → versionDAO.GetLatest → decodeCanvasFromDSL →
canvas.Compile → cc.Workflow.Invoke → answer extraction
```
using in-memory SQLite + miniredis (no Docker). Covers:
- `TestRunAgent_RealCanvas_BeginMessage` — happy path, `{{sys.query}}`
resolution
- `TestRunAgent_RealCanvas_WaitForUserResume` — two-run resume cycle
(Python-equivalent)
- `TestRunAgent_RealCanvas_CompileFails` — unknown component name →
sanitized error (Python-equivalent)
- `TestRunAgent_RealCanvas_InvokeFails` — unresolvable template ref
(Python-equivalent)
- `TestRunAgent_RunTracker_AttachCheckpoint_CallSequence` —
Start→AttachCheckpoint→MarkSucceeded lifecycle
`internal/handler/agent_test.go` — SSE streaming parity (`Content-Type:
text/event-stream`, `data: {…}\n\n`, trailing `data: [DONE]\n\n`,
OpenAI-compatible non-stream `choices`).
`internal/agent/canvas/fixture_compile_test.go` + per-component tests
pin the Python-equivalent outputs.
```
go test -count=1 -v -run 'TestRunAgent_RealCanvas|TestRunAgent_RunTracker' ./internal/service/
```
## Design reference
`docs/develop/agent-go-port-design.md` (1329 lines, last cross-checked
2026-06-17) — module layout, per-component / per-tool inventory,
corner-case catalogue, and the actionable backlog (Section 14, including
the retrieval alignment follow-up).
---------
Co-authored-by: Claude <noreply@anthropic.com>
87 lines
3.6 KiB
Go
87 lines
3.6 KiB
Go
// Package canvas implements the RAGFlow agent canvas Go port.
|
|
//
|
|
// Shared runtime contracts (CanvasState, Component, ComponentFactory,
|
|
// state context plumbing, template helpers) live in
|
|
// internal/agent/runtime. Canvas re-exports them through thin aliases
|
|
// so existing call sites keep working while breaking the historic
|
|
// canvas <-> component import cycle.
|
|
package canvas
|
|
|
|
import (
|
|
"ragflow/internal/agent/runtime"
|
|
)
|
|
|
|
// legacyNoOpNames is the set of component names that the Go port
|
|
// recognises for DSL v1 compatibility but does not ship a real
|
|
// implementation for. Encountering one of these in a DSL is mapped to
|
|
// the same no-op echo lambda used for placeholder bodies by the
|
|
// BuildWorkflow in scheduler.go. New DSLs should not use these names —
|
|
// they exist only so v1 DSLs that reference Python-era sentinel
|
|
// components ("ExitLoop") still compile and run in the Go port.
|
|
//
|
|
// Membership semantics inside a Loop's sub-graph: legacy names that
|
|
// appear as descendants of a Loop are absorbed as no-op members of the
|
|
// sub-graph; they do not contribute to loop control. Termination is
|
|
// driven by the Loop's loop_termination_condition predicate, not by
|
|
// reaching an ExitLoop node.
|
|
var legacyNoOpNames = map[string]bool{
|
|
"exitloop": true,
|
|
}
|
|
|
|
// CanvasState aliases runtime.CanvasState so existing canvas callers
|
|
// (and component tests that still import the canvas package) keep
|
|
// compiling without changes. The canonical definition lives in
|
|
// internal/agent/runtime/state.go.
|
|
type CanvasState = runtime.CanvasState
|
|
|
|
// NewCanvasState re-exports runtime.NewCanvasState.
|
|
func NewCanvasState(runID, taskID string) *CanvasState {
|
|
return runtime.NewCanvasState(runID, taskID)
|
|
}
|
|
|
|
// Canvas is the in-memory DSL representation loaded from a user_canvas row.
|
|
// It is the input to compile.go which builds the eino Workflow.
|
|
type Canvas struct {
|
|
Components map[string]CanvasComponent `json:"components"`
|
|
Path []string `json:"path"`
|
|
History []map[string]any `json:"history,omitempty"`
|
|
Retrieval map[string]any `json:"retrieval,omitempty"`
|
|
Globals map[string]any `json:"globals,omitempty"`
|
|
// NodeParents preserves the front-end graph's grouping metadata
|
|
// (graph.nodes[*].parentId) for runtime-only subgraph expansion.
|
|
// The backend treats the incoming DSL as read-only; this is a
|
|
// decoder-side mirror used only to decide which nodes belong to a
|
|
// Loop / Parallel body during compilation.
|
|
NodeParents map[string]string `json:"-"`
|
|
}
|
|
|
|
// CanvasComponent is the in-memory DSL node. The Obj.ComponentName
|
|
// matches agent/component/<name>.py's class name (case-insensitive,
|
|
// per Python v1 DSL semantics).
|
|
type CanvasComponent struct {
|
|
Obj CanvasComponentObj `json:"obj"`
|
|
Downstream []string `json:"downstream"`
|
|
Upstream []string `json:"upstream"`
|
|
}
|
|
|
|
type CanvasComponentObj struct {
|
|
ComponentName string `json:"component_name"`
|
|
Params map[string]any `json:"params"`
|
|
}
|
|
|
|
// Component is an alias for runtime.Component — the minimal runtime
|
|
// surface BuildWorkflow needs at sub-graph build time. The canonical
|
|
// definition (and the SetDefaultFactory / DefaultFactory plumbing)
|
|
// lives in internal/agent/runtime/component.go.
|
|
type Component = runtime.Component
|
|
|
|
// ComponentFactory aliases runtime.ComponentFactory.
|
|
type ComponentFactory = runtime.ComponentFactory
|
|
|
|
// SetDefaultFactory re-exports runtime.SetDefaultFactory. The
|
|
// orchestrator's main.go can call either entry point; new code
|
|
// should prefer the runtime package directly.
|
|
func SetDefaultFactory(f ComponentFactory) {
|
|
runtime.SetDefaultFactory(f)
|
|
}
|