mirror of
https://github.com/infiniflow/ragflow.git
synced 2026-06-30 16:01:58 +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>
206 lines
5.5 KiB
Go
206 lines
5.5 KiB
Go
// Package canvas — variable resolver unit tests.
|
|
//
|
|
// Scope: tests the 3 reference forms documented in
|
|
// docs/develop/agent-go-port-design.md appendix D:
|
|
// - cpn_id@param (e.g. "llm_0@content", "begin_0@query")
|
|
// - sys.<name> (e.g. "sys.query", "sys.user_id")
|
|
// - env.<name> (e.g. "env.max_tokens")
|
|
//
|
|
// Additional supported aliases:
|
|
// - {{item}} / {{index}} iteration aliases
|
|
//
|
|
// Out of scope:
|
|
// - nested dot paths (cpn_0@result.answer) — base.py:400-410 does this
|
|
// in canvas.get_value_with_variable AFTER the regex match succeeds.
|
|
// - list indexing (xs.0) — same nested-path machinery.
|
|
//
|
|
// Cpn IDs in tests use underscores (e.g. "llm_0") which is the real
|
|
// RAGFlow naming convention; the original documented regex
|
|
// `[a-zA-Z:0-9]+` did not allow underscores — see variable.go
|
|
// VarRefPattern comment.
|
|
package canvas
|
|
|
|
import (
|
|
"reflect"
|
|
"testing"
|
|
)
|
|
|
|
func TestVariableResolver(t *testing.T) {
|
|
mkState := func() *CanvasState {
|
|
s := NewCanvasState("run-1", "task-1")
|
|
s.SetVar("llm_0", "content", "hello world")
|
|
s.SetVar("begin_0", "query", "ragflow go port")
|
|
s.Sys["query"] = "what is ragflow"
|
|
s.Sys["user_id"] = "tenant-1"
|
|
s.Env["max_tokens"] = 1024
|
|
return s
|
|
}
|
|
|
|
type tcase struct {
|
|
name string
|
|
template string
|
|
setup func(s *CanvasState)
|
|
want string
|
|
wantErr bool
|
|
}
|
|
|
|
cases := []tcase{
|
|
{
|
|
name: "single cpn ref",
|
|
template: "{{llm_0@content}}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "hello world",
|
|
},
|
|
{
|
|
name: "triple-brace (Python allows extra braces)",
|
|
template: "{{{llm_0@content}}}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "hello world",
|
|
},
|
|
{
|
|
name: "single brace (Python allows)",
|
|
template: "{llm_0@content}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "hello world",
|
|
},
|
|
{
|
|
name: "embedded in text",
|
|
template: "Refined: {{llm_0@content}} done",
|
|
setup: func(s *CanvasState) {},
|
|
want: "Refined: hello world done",
|
|
},
|
|
{
|
|
name: "sys ref",
|
|
template: "Q: {{sys.query}}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "Q: what is ragflow",
|
|
},
|
|
{
|
|
name: "env ref",
|
|
template: "limit {{env.max_tokens}}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "limit 1024",
|
|
},
|
|
{
|
|
name: "multiple refs in one template",
|
|
template: "{{sys.query}} :: {{llm_0@content}} :: {{env.max_tokens}}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "what is ragflow :: hello world :: 1024",
|
|
},
|
|
{
|
|
name: "no ref returns input as-is",
|
|
template: "plain text only",
|
|
setup: func(s *CanvasState) {},
|
|
want: "plain text only",
|
|
},
|
|
{
|
|
// Go behavior: ResolveTemplate returns an error on
|
|
// unresolved refs (loud-fail; see variable.go
|
|
// ResolveTemplate doc). Python's canvas.py:177-178
|
|
// silently returns "" — the Go port trades Python's
|
|
// silent soft-fail for a Go-idiomatic error return so
|
|
// parameter binding can surface misconfigured canvases
|
|
// early.
|
|
name: "unresolved cpn ref returns error (loud-fail, Go port deviation)",
|
|
template: "x={{missing@thing}}y",
|
|
setup: func(s *CanvasState) {},
|
|
wantErr: true,
|
|
},
|
|
{
|
|
name: "sys ref missing key returns error",
|
|
template: "[{{sys.nope}}]",
|
|
setup: func(s *CanvasState) {},
|
|
wantErr: true,
|
|
},
|
|
{
|
|
name: "iteration item alias resolves from globals",
|
|
template: "{{item}}",
|
|
setup: func(s *CanvasState) {
|
|
s.Globals["__item__"] = "alpha"
|
|
},
|
|
want: "alpha",
|
|
},
|
|
{
|
|
name: "iteration index alias resolves from globals",
|
|
template: "i={{index}}",
|
|
setup: func(s *CanvasState) {
|
|
s.Globals["__index__"] = 3
|
|
},
|
|
want: "i=3",
|
|
},
|
|
{
|
|
name: "garbage ref (no @ or sys/env prefix) passes through unchanged",
|
|
template: "{{garbage}}",
|
|
setup: func(s *CanvasState) {},
|
|
want: "{{garbage}}",
|
|
},
|
|
{
|
|
name: "empty template",
|
|
template: "",
|
|
setup: func(s *CanvasState) {},
|
|
want: "",
|
|
},
|
|
}
|
|
|
|
for _, c := range cases {
|
|
t.Run(c.name, func(t *testing.T) {
|
|
s := mkState()
|
|
c.setup(s)
|
|
got, err := ResolveTemplate(c.template, s)
|
|
if c.wantErr {
|
|
if err == nil {
|
|
t.Fatalf("expected error, got nil (val=%q)", got)
|
|
}
|
|
return
|
|
}
|
|
if err != nil {
|
|
t.Fatalf("unexpected error: %v", err)
|
|
}
|
|
if got != c.want {
|
|
t.Fatalf("got %q want %q", got, c.want)
|
|
}
|
|
})
|
|
}
|
|
}
|
|
|
|
// TestVarRefPattern_MatchesPythonDrift guards against accidental regex
|
|
// changes. If someone edits VarRefPattern, this test demands they also
|
|
// update the Python source (or document the deviation) — preventing
|
|
// silent divergence between Go and Python regex behavior.
|
|
func TestVarRefPattern_MatchesPythonDrift(t *testing.T) {
|
|
positive := []string{
|
|
"{{llm_0@content}}",
|
|
"{{{llm_0@content}}}",
|
|
"{llm_0@content}",
|
|
"{{sys.query}}",
|
|
"{{sys.user_id}}",
|
|
"{{env.max_tokens}}",
|
|
"{{begin_0@query}}",
|
|
"prefix {{llm_0@x}} suffix",
|
|
"{{agent:ThreePathsDecide@content}}", // colon-prefixed cpn id
|
|
}
|
|
for _, s := range positive {
|
|
if !VarRefPattern.MatchString(s) {
|
|
t.Errorf("expected match for %q", s)
|
|
}
|
|
}
|
|
negative := []string{
|
|
"plain text",
|
|
"",
|
|
}
|
|
for _, s := range negative {
|
|
if VarRefPattern.MatchString(s) {
|
|
t.Errorf("expected NO match for %q", s)
|
|
}
|
|
}
|
|
}
|
|
|
|
// TestExtractRefs covers the pure-regex extraction helper.
|
|
func TestExtractRefs(t *testing.T) {
|
|
got := ExtractRefs("{{a@x}} {{b@y}} {{a@x}} {{sys.q}}")
|
|
want := []string{"a@x", "b@y", "sys.q"}
|
|
if !reflect.DeepEqual(got, want) {
|
|
t.Fatalf("ExtractRefs: got %v want %v", got, want)
|
|
}
|
|
}
|