Files
ragflow/internal/agent/component/llm_credentials.go
Zhichang Yu 3f805a64f1 feat(agent): align Go agent behavior with Python (except retrieval component) (#16225)
## 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>
2026-06-22 11:58:29 +08:00

163 lines
5.6 KiB
Go

package component
import (
"context"
"encoding/json"
"log"
"strings"
"ragflow/internal/agent/runtime"
"ragflow/internal/dao"
"ragflow/internal/entity"
)
// resolveTenantLLMConfig fills tenant-scoped API credentials for the supplied
// driver/model pair when the canvas DSL omitted them. It first checks the old
// tenant_llm table, then falls back to tenant_model_provider +
// tenant_model_instance when the composite llm_id carries an instance name.
func resolveTenantLLMConfig(ctx context.Context, driver, modelID, apiKey, baseURL, originalModelID string) (string, string) {
if apiKey != "" || driver == "" || modelID == "" {
return apiKey, baseURL
}
state, _, err := runtime.GetStateFromContext[*runtime.CanvasState](ctx)
if err != nil || state == nil {
log.Printf("DEBUG llm credentials: no canvas state in ctx")
return apiKey, baseURL
}
tid, _ := state.Sys["tenant_id"].(string)
if tid == "" {
log.Printf("DEBUG llm credentials: state.Sys has no tenant_id")
return apiKey, baseURL
}
if resolvedKey, resolvedBaseURL, ok := resolveTenantLLMCredentials(tid, driver, modelID, baseURL); ok {
return resolvedKey, resolvedBaseURL
}
if originalModelID == "" {
return apiKey, baseURL
}
if resolvedKey, resolvedBaseURL, ok := resolveTenantModelInstanceCredentials(tid, originalModelID, baseURL); ok {
return resolvedKey, resolvedBaseURL
}
return apiKey, baseURL
}
// resolveTenantLLMCredentials looks up the old tenant_llm table for the given
// tenant / factory / model. Returns true when credentials were found.
func resolveTenantLLMCredentials(tid, driver, modelID, baseURL string) (string, string, bool) {
log.Printf("DEBUG llm credentials: tenant_llm lookup tid=%q factory=%q model=%q", tid, driver, modelID)
row, err := dao.NewTenantLLMDAO().GetByTenantFactoryAndModelName(tid, driver, modelID)
if err != nil {
log.Printf("DEBUG llm credentials: tenant_llm lookup err=%v", err)
return "", baseURL, false
}
if row == nil {
log.Printf("DEBUG llm credentials: tenant_llm lookup: no row")
return "", baseURL, false
}
apiKey := ""
if row.APIKey != nil {
apiKey = *row.APIKey
}
if baseURL == "" && row.APIBase != nil {
baseURL = *row.APIBase
}
log.Printf("DEBUG llm credentials: tenant_llm OK api_key=%q base_url=%q", apiKey, baseURL)
return apiKey, baseURL, apiKey != ""
}
// resolveTenantModelInstanceCredentials attempts to resolve llm credentials
// through tenant_model_provider + tenant_model_instance using the original
// composite llm_id (which still carries the instance name).
func resolveTenantModelInstanceCredentials(tid, compositeLLMID, baseURL string) (string, string, bool) {
modelName, instanceName, providerName := parseLLMIDParts(compositeLLMID)
if instanceName == "" {
log.Printf("DEBUG llm credentials: new-table fallback skipped: no instance name in %q", compositeLLMID)
return "", baseURL, false
}
log.Printf("DEBUG llm credentials: new-table fallback tid=%q provider=%q model=%q instance=%q",
tid, providerName, modelName, instanceName)
provider, err := dao.NewTenantModelProviderDAO().GetByTenantIDAndProviderName(tid, providerName)
if err != nil || provider == nil {
log.Printf("DEBUG llm credentials: new-table fallback: provider %q not found (err=%v)", providerName, err)
return "", baseURL, false
}
instance, err := dao.NewTenantModelInstanceDAO().GetByProviderIDAndInstanceName(provider.ID, instanceName)
if err != nil || instance == nil {
if instanceName == "default" {
if fallback := findSoleActiveProviderInstance(provider.ID); fallback != nil {
log.Printf("DEBUG llm credentials: new-table fallback: remapped default instance to sole active instance %q for provider %q",
fallback.InstanceName, providerName)
instance = fallback
err = nil
}
}
}
if err != nil || instance == nil {
log.Printf("DEBUG llm credentials: new-table fallback: instance %q not found for provider %q (err=%v)",
instanceName, providerName, err)
return "", baseURL, false
}
apiKey := instance.APIKey
if instance.Extra != "" && baseURL == "" {
var extra map[string]string
if err := json.Unmarshal([]byte(instance.Extra), &extra); err == nil {
if u := extra["base_url"]; u != "" {
baseURL = u
}
}
}
log.Printf("DEBUG llm credentials: new-table OK provider=%q instance=%q api_key=%q base_url=%q",
providerName, instance.InstanceName, apiKey, baseURL)
return apiKey, baseURL, apiKey != ""
}
func findSoleActiveProviderInstance(providerID string) *entity.TenantModelInstance {
instances, err := dao.NewTenantModelInstanceDAO().GetAllInstancesByProviderID(providerID)
if err != nil {
log.Printf("DEBUG llm credentials: list provider instances err=%v", err)
return nil
}
active := make([]*entity.TenantModelInstance, 0, len(instances))
for _, inst := range instances {
if inst == nil {
continue
}
if strings.EqualFold(strings.TrimSpace(inst.Status), "inactive") {
continue
}
active = append(active, inst)
}
if len(active) != 1 {
return nil
}
return active[0]
}
// parseLLMIDParts splits a composite llm_id into model, instance, and
// provider segments.
//
// "model@provider" -> ("model", "default", "provider")
// "model@instance@provider" -> ("model", "instance", "provider")
// 4+ parts -> ("parts[0]", "parts[1]", "parts[2]")
func parseLLMIDParts(s string) (modelName, instanceName, providerName string) {
parts := strings.Split(strings.TrimSpace(s), "@")
switch len(parts) {
case 2:
return parts[0], "default", parts[1]
case 3:
return parts[0], parts[1], parts[2]
default:
if len(parts) >= 4 {
return parts[0], parts[1], parts[2]
}
return s, "", ""
}
}