feat: Phase 2 — RAG integration in Requirements + DSFA Draft
All checks were successful
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / test-go-ai-compliance (push) Successful in 35s
CI / test-python-backend-compliance (push) Successful in 26s
CI / test-python-document-crawler (push) Successful in 22s
CI / test-python-dsms-gateway (push) Successful in 19s

Add legal context enrichment from Qdrant vector corpus to the two
highest-priority modules (Requirements AI assistant and DSFA drafting
engine).

Go SDK:
- Add SearchCollection() with collection override + whitelist validation
- Refactor Search() to delegate to shared searchInternal()

Python backend:
- New ComplianceRAGClient proxying POST /sdk/v1/rag/search (error-tolerant)
- AI assistant: enrich interpret_requirement() and suggest_controls() with RAG
- Requirements API: add ?include_legal_context=true query parameter

Admin (Next.js):
- Extract shared queryRAG() utility from chat route
- Inject RAG legal context into v1 and v2 draft pipelines

Tests for all three layers (Go, Python, TypeScript shared utility).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-03-02 08:57:39 +01:00
parent 3d9bc285ac
commit 14a99322eb
12 changed files with 739 additions and 36 deletions

View File

@@ -21,9 +21,20 @@ func NewRAGHandlers(corpusVersionStore *ucca.CorpusVersionStore) *RAGHandlers {
}
}
// AllowedCollections is the whitelist of Qdrant collections that can be queried.
var AllowedCollections = map[string]bool{
"bp_compliance_ce": true,
"bp_compliance_recht": true,
"bp_compliance_gesetze": true,
"bp_compliance_datenschutz": true,
"bp_dsfa_corpus": true,
"bp_legal_templates": true,
}
// SearchRequest represents a RAG search request.
type SearchRequest struct {
Query string `json:"query" binding:"required"`
Collection string `json:"collection,omitempty"`
Regulations []string `json:"regulations,omitempty"`
TopK int `json:"top_k,omitempty"`
}
@@ -41,7 +52,15 @@ func (h *RAGHandlers) Search(c *gin.Context) {
req.TopK = 5
}
results, err := h.ragClient.Search(c.Request.Context(), req.Query, req.Regulations, req.TopK)
// Validate collection if specified
if req.Collection != "" {
if !AllowedCollections[req.Collection] {
c.JSON(http.StatusBadRequest, gin.H{"error": "Unknown collection: " + req.Collection + ". Allowed: bp_compliance_ce, bp_compliance_recht, bp_compliance_gesetze, bp_compliance_datenschutz, bp_dsfa_corpus, bp_legal_templates"})
return
}
}
results, err := h.ragClient.SearchCollection(c.Request.Context(), req.Collection, req.Query, req.Regulations, req.TopK)
if err != nil {
c.JSON(http.StatusInternalServerError, gin.H{"error": "RAG search failed: " + err.Error()})
return

View File

@@ -0,0 +1,109 @@
package handlers
import (
"bytes"
"encoding/json"
"net/http"
"net/http/httptest"
"testing"
"github.com/gin-gonic/gin"
)
func TestAllowedCollections(t *testing.T) {
allowed := []string{
"bp_compliance_ce",
"bp_compliance_recht",
"bp_compliance_gesetze",
"bp_compliance_datenschutz",
"bp_dsfa_corpus",
"bp_legal_templates",
}
for _, c := range allowed {
if !AllowedCollections[c] {
t.Errorf("Expected %s to be in AllowedCollections", c)
}
}
disallowed := []string{
"bp_unknown",
"",
"some_random_collection",
}
for _, c := range disallowed {
if AllowedCollections[c] {
t.Errorf("Expected %s to NOT be in AllowedCollections", c)
}
}
}
func TestSearch_InvalidCollection_Returns400(t *testing.T) {
gin.SetMode(gin.TestMode)
handler := &RAGHandlers{}
w := httptest.NewRecorder()
c, _ := gin.CreateTestContext(w)
body := SearchRequest{
Query: "test query",
Collection: "bp_evil_collection",
TopK: 5,
}
bodyBytes, _ := json.Marshal(body)
c.Request, _ = http.NewRequest("POST", "/sdk/v1/rag/search", bytes.NewReader(bodyBytes))
c.Request.Header.Set("Content-Type", "application/json")
handler.Search(c)
if w.Code != http.StatusBadRequest {
t.Errorf("Expected 400, got %d", w.Code)
}
var resp map[string]interface{}
json.Unmarshal(w.Body.Bytes(), &resp)
errMsg, ok := resp["error"].(string)
if !ok || errMsg == "" {
t.Error("Expected error message in response")
}
}
func TestSearch_WithCollectionParam_BindsCorrectly(t *testing.T) {
// Test that the SearchRequest struct correctly binds the collection field
body := `{"query":"DSGVO Art. 35","collection":"bp_compliance_recht","top_k":3}`
var req SearchRequest
err := json.Unmarshal([]byte(body), &req)
if err != nil {
t.Fatalf("Failed to unmarshal: %v", err)
}
if req.Query != "DSGVO Art. 35" {
t.Errorf("Expected query 'DSGVO Art. 35', got '%s'", req.Query)
}
if req.Collection != "bp_compliance_recht" {
t.Errorf("Expected collection 'bp_compliance_recht', got '%s'", req.Collection)
}
if req.TopK != 3 {
t.Errorf("Expected top_k 3, got %d", req.TopK)
}
}
func TestSearch_EmptyCollection_IsAllowed(t *testing.T) {
// Empty collection should be allowed (falls back to default in the handler)
body := `{"query":"test"}`
var req SearchRequest
err := json.Unmarshal([]byte(body), &req)
if err != nil {
t.Fatalf("Failed to unmarshal: %v", err)
}
if req.Collection != "" {
t.Errorf("Expected empty collection, got '%s'", req.Collection)
}
// Empty string is not in AllowedCollections, but the handler
// should skip validation for empty collection
}