validatord: /v1/validate + /v1/iterate HTTP surface (port 3221)
Closes the last "Go primary" backlog item in docs/ARCHITECTURE_COMPARISON.md. Go now owns the entire validator path end-to-end — no Rust dep for staffing safety net. Architecture: cmd/validatord on :3221 hosts both endpoints. Calls chatd directly for the iterate loop's LLM hop (no gateway self-loopback like the Rust shape). Gateway proxies /v1/validate + /v1/iterate to validatord. What's in: - internal/validator/playbook.go — 3rd validator kind (PRD checks: fill: prefix, endorsed_names ≤ target_count×2, fingerprint required) - internal/validator/lookup_jsonl.go — JSONL roster loader (Parquet deferred; producer one-liner documented in package comment) - internal/validator/iterate.go — ExtractJSON helper + Iterate orchestrator with ChatCaller seam for unit tests - cmd/validatord/main.go — HTTP routes, roster load, chat client - internal/shared/config.go — ValidatordConfig + gateway URL field - lakehouse.toml — [validatord] section - cmd/gateway/main.go — proxy routes for /v1/validate + /v1/iterate Smoke: 5/5 PASS through gateway :3110: ✓ playbook happy path ✓ playbook missing fingerprint → 422 schema/fingerprint ✓ phantom candidate W-PHANTOM → 422 consistency ✓ unknown kind → 400 ✓ roster loaded with 3 records go test ./... green across 33 packages. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -30,6 +30,7 @@ type Config struct {
|
||||
Matrixd MatrixdConfig `toml:"matrixd"`
|
||||
Observerd ObserverdConfig `toml:"observerd"`
|
||||
Chatd ChatdConfig `toml:"chatd"`
|
||||
Validatord ValidatordConfig `toml:"validatord"`
|
||||
S3 S3Config `toml:"s3"`
|
||||
Models ModelsConfig `toml:"models"`
|
||||
Log LogConfig `toml:"log"`
|
||||
@@ -70,6 +71,7 @@ type GatewayConfig struct {
|
||||
MatrixdURL string `toml:"matrixd_url"`
|
||||
ObserverdURL string `toml:"observerd_url"`
|
||||
ChatdURL string `toml:"chatd_url"`
|
||||
ValidatordURL string `toml:"validatord_url"`
|
||||
}
|
||||
|
||||
// EmbeddConfig drives the embed service. ProviderURL points at the
|
||||
@@ -143,6 +145,28 @@ type ChatdConfig struct {
|
||||
TimeoutSecs int `toml:"timeout_secs"`
|
||||
}
|
||||
|
||||
// ValidatordConfig drives the validator service (cmd/validatord).
|
||||
// Hosts /validate (FillValidator + EmailValidator + PlaybookValidator)
|
||||
// and /iterate (generate→validate→correct loop). Routes to chatd via
|
||||
// ChatdURL for the iteration loop's LLM hops.
|
||||
//
|
||||
// RosterPath points at a JSONL roster (one WorkerRecord per line) that
|
||||
// FillValidator and EmailValidator use for worker-existence checks.
|
||||
// Empty disables the roster — worker-existence checks all fail
|
||||
// Consistency, which is the correct behavior when the roster isn't
|
||||
// configured. Production sets a stable path under /var/lib/lakehouse/.
|
||||
type ValidatordConfig struct {
|
||||
Bind string `toml:"bind"`
|
||||
ChatdURL string `toml:"chatd_url"`
|
||||
RosterPath string `toml:"roster_path"`
|
||||
// Per-call cap on the iteration loop. 0 = 3 (Phase 43 default).
|
||||
DefaultMaxIterations int `toml:"default_max_iterations"`
|
||||
// Per-call cap on chat hop max_tokens. 0 = 4096.
|
||||
DefaultMaxTokens int `toml:"default_max_tokens"`
|
||||
// Per-call timeout for the chat hop in seconds. 0 = 240s.
|
||||
ChatTimeoutSecs int `toml:"chat_timeout_secs"`
|
||||
}
|
||||
|
||||
// ObserverdConfig drives the observer service (cmd/observerd).
|
||||
// PersistPath: file path to the JSONL ops log; empty = in-memory
|
||||
// only (test/dev). Production sets a stable path under
|
||||
@@ -328,6 +352,7 @@ func DefaultConfig() Config {
|
||||
MatrixdURL: "http://127.0.0.1:3218",
|
||||
ObserverdURL: "http://127.0.0.1:3219",
|
||||
ChatdURL: "http://127.0.0.1:3220",
|
||||
ValidatordURL: "http://127.0.0.1:3221",
|
||||
},
|
||||
Storaged: ServiceConfig{Bind: "127.0.0.1:3211"},
|
||||
Catalogd: CatalogConfig{Bind: "127.0.0.1:3212", StoragedURL: "http://127.0.0.1:3211"},
|
||||
@@ -361,6 +386,14 @@ func DefaultConfig() Config {
|
||||
Bind: "127.0.0.1:3219",
|
||||
// PersistPath empty by default = in-memory only.
|
||||
},
|
||||
Validatord: ValidatordConfig{
|
||||
Bind: "127.0.0.1:3221",
|
||||
ChatdURL: "http://127.0.0.1:3220",
|
||||
RosterPath: "", // empty = no roster; worker-existence checks fail Consistency
|
||||
DefaultMaxIterations: 3,
|
||||
DefaultMaxTokens: 4096,
|
||||
ChatTimeoutSecs: 240,
|
||||
},
|
||||
Chatd: ChatdConfig{
|
||||
Bind: "127.0.0.1:3220",
|
||||
OllamaURL: "http://localhost:11434",
|
||||
|
||||
@@ -0,0 +1,237 @@
|
||||
package validator
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"strings"
|
||||
)
|
||||
|
||||
// IterateRequest is the input to Iterate. Mirrors Rust's
|
||||
// IterateRequest in `crates/gateway/src/v1/iterate.rs` so JSONL
|
||||
// captured from one runtime parses on the other.
|
||||
type IterateRequest struct {
|
||||
Kind string `json:"kind"`
|
||||
Prompt string `json:"prompt"`
|
||||
Provider string `json:"provider"`
|
||||
Model string `json:"model"`
|
||||
System string `json:"system,omitempty"`
|
||||
Context map[string]any `json:"context,omitempty"`
|
||||
MaxIterations int `json:"max_iterations,omitempty"`
|
||||
Temperature *float64 `json:"temperature,omitempty"`
|
||||
MaxTokens int `json:"max_tokens,omitempty"`
|
||||
}
|
||||
|
||||
// IterateAttempt is one row in the history. raw is capped at 2000
|
||||
// chars on the wire to keep responses bounded.
|
||||
type IterateAttempt struct {
|
||||
Iteration int `json:"iteration"`
|
||||
Raw string `json:"raw"`
|
||||
Status AttemptStatus `json:"status"`
|
||||
}
|
||||
|
||||
// AttemptStatus is the per-attempt verdict. Tagged JSON so consumers
|
||||
// can switch on `kind` without trying to parse the optional error.
|
||||
type AttemptStatus struct {
|
||||
Kind string `json:"kind"` // "no_json" | "validation_failed" | "accepted"
|
||||
Error string `json:"error,omitempty"`
|
||||
}
|
||||
|
||||
// IterateResponse is the success payload (200 + Report + accepted artifact).
|
||||
type IterateResponse struct {
|
||||
Artifact map[string]any `json:"artifact"`
|
||||
Validation Report `json:"validation"`
|
||||
Iterations int `json:"iterations"`
|
||||
History []IterateAttempt `json:"history"`
|
||||
}
|
||||
|
||||
// IterateFailure is the max-iter-exhausted payload (422 + history).
|
||||
type IterateFailure struct {
|
||||
Error string `json:"error"`
|
||||
Iterations int `json:"iterations"`
|
||||
History []IterateAttempt `json:"history"`
|
||||
}
|
||||
|
||||
// ChatCaller is the seam Iterate uses to invoke an LLM. Tests inject
|
||||
// scripted callers; production wires this to the chatd /v1/chat HTTP
|
||||
// endpoint. Implementations must return the model's textual content
|
||||
// (no choices wrapper, no message envelope).
|
||||
type ChatCaller func(ctx context.Context, system, user, provider, model string, temperature *float64, maxTokens int) (string, error)
|
||||
|
||||
// IterateConfig threads daemon-level settings into the orchestrator.
|
||||
type IterateConfig struct {
|
||||
DefaultMaxIterations int
|
||||
DefaultMaxTokens int
|
||||
DefaultTemperature float64
|
||||
}
|
||||
|
||||
const (
|
||||
defaultMaxIterations = 3
|
||||
defaultMaxTokens = 4096
|
||||
defaultTemperature = 0.2
|
||||
)
|
||||
|
||||
// Iterate runs the generate→validate→correct loop. Returns
|
||||
// IterateResponse on success (with full history) or IterateFailure
|
||||
// on max-iter exhaustion. Infrastructure errors (chat hop fails)
|
||||
// surface as Go errors so the HTTP layer can return 502.
|
||||
func Iterate(ctx context.Context, req IterateRequest, cfg IterateConfig, chat ChatCaller, validate func(string, map[string]any) (Report, error)) (*IterateResponse, *IterateFailure, error) {
|
||||
maxIter := req.MaxIterations
|
||||
if maxIter <= 0 {
|
||||
maxIter = cfg.DefaultMaxIterations
|
||||
}
|
||||
if maxIter <= 0 {
|
||||
maxIter = defaultMaxIterations
|
||||
}
|
||||
maxTokens := req.MaxTokens
|
||||
if maxTokens <= 0 {
|
||||
maxTokens = cfg.DefaultMaxTokens
|
||||
}
|
||||
if maxTokens <= 0 {
|
||||
maxTokens = defaultMaxTokens
|
||||
}
|
||||
temp := req.Temperature
|
||||
if temp == nil {
|
||||
t := cfg.DefaultTemperature
|
||||
if t == 0 {
|
||||
t = defaultTemperature
|
||||
}
|
||||
temp = &t
|
||||
}
|
||||
|
||||
currentPrompt := req.Prompt
|
||||
history := make([]IterateAttempt, 0, maxIter)
|
||||
|
||||
for i := 0; i < maxIter; i++ {
|
||||
raw, err := chat(ctx, req.System, currentPrompt, req.Provider, req.Model, temp, maxTokens)
|
||||
if err != nil {
|
||||
return nil, nil, fmt.Errorf("/v1/chat hop failed at iter %d: %w", i, err)
|
||||
}
|
||||
|
||||
artifact := ExtractJSON(raw)
|
||||
if artifact == nil {
|
||||
history = append(history, IterateAttempt{
|
||||
Iteration: i,
|
||||
Raw: trim(raw, 2000),
|
||||
Status: AttemptStatus{Kind: "no_json"},
|
||||
})
|
||||
currentPrompt = req.Prompt + "\n\nYour previous attempt did not contain a JSON object. Reply with ONLY a valid JSON object matching the requested artifact shape."
|
||||
continue
|
||||
}
|
||||
|
||||
report, vErr := validate(req.Kind, artifact)
|
||||
if vErr == nil {
|
||||
history = append(history, IterateAttempt{
|
||||
Iteration: i,
|
||||
Raw: trim(raw, 2000),
|
||||
Status: AttemptStatus{Kind: "accepted"},
|
||||
})
|
||||
return &IterateResponse{
|
||||
Artifact: artifact,
|
||||
Validation: report,
|
||||
Iterations: i + 1,
|
||||
History: history,
|
||||
}, nil, nil
|
||||
}
|
||||
|
||||
// Validation failed — append error to prompt for next iter.
|
||||
// The model sees concrete failure mode + retries with corrective
|
||||
// context. Same "validator IS the observer" shape as Phase 43.
|
||||
errSummary := vErr.Error()
|
||||
history = append(history, IterateAttempt{
|
||||
Iteration: i,
|
||||
Raw: trim(raw, 2000),
|
||||
Status: AttemptStatus{Kind: "validation_failed", Error: errSummary},
|
||||
})
|
||||
currentPrompt = req.Prompt + "\n\nPrior attempt failed validation:\n" + errSummary + "\n\nFix the specific issue above and respond with a corrected JSON object."
|
||||
}
|
||||
|
||||
return nil, &IterateFailure{
|
||||
Error: fmt.Sprintf("max iterations reached (%d) without passing validation", maxIter),
|
||||
Iterations: maxIter,
|
||||
History: history,
|
||||
}, nil
|
||||
}
|
||||
|
||||
// ExtractJSON pulls the first JSON object from a model's output.
|
||||
// Handles fenced code blocks (```json ... ```), bare braces, and
|
||||
// stray prose around the JSON. Returns nil on no extractable object.
|
||||
//
|
||||
// Same algorithm shape as Rust's extract_json so a model producing
|
||||
// output that one runtime accepts will be accepted by the other.
|
||||
func ExtractJSON(raw string) map[string]any {
|
||||
// Try fenced first.
|
||||
for _, c := range fencedCandidates(raw) {
|
||||
if v, ok := parseObject(c); ok {
|
||||
return v
|
||||
}
|
||||
}
|
||||
// Fall back to outermost {...} balance.
|
||||
bytes := []byte(raw)
|
||||
depth := 0
|
||||
start := -1
|
||||
for i, b := range bytes {
|
||||
switch b {
|
||||
case '{':
|
||||
if start < 0 {
|
||||
start = i
|
||||
}
|
||||
depth++
|
||||
case '}':
|
||||
depth--
|
||||
if depth == 0 && start >= 0 {
|
||||
if v, ok := parseObject(raw[start : i+1]); ok {
|
||||
return v
|
||||
}
|
||||
start = -1
|
||||
}
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// fencedCandidates returns the bodies of every ``` fenced block in
|
||||
// `raw`. Skips an optional language tag on the opening fence (e.g.
|
||||
// ```json).
|
||||
func fencedCandidates(raw string) []string {
|
||||
var out []string
|
||||
s := raw
|
||||
for {
|
||||
idx := strings.Index(s, "```")
|
||||
if idx < 0 {
|
||||
break
|
||||
}
|
||||
after := s[idx+3:]
|
||||
// Skip optional language tag up to the first newline.
|
||||
bodyStart := strings.Index(after, "\n")
|
||||
if bodyStart < 0 {
|
||||
bodyStart = 0
|
||||
} else {
|
||||
bodyStart++
|
||||
}
|
||||
body := after[bodyStart:]
|
||||
end := strings.Index(body, "```")
|
||||
if end < 0 {
|
||||
break
|
||||
}
|
||||
out = append(out, strings.TrimSpace(body[:end]))
|
||||
s = body[end+3:]
|
||||
}
|
||||
return out
|
||||
}
|
||||
|
||||
func parseObject(s string) (map[string]any, bool) {
|
||||
var v any
|
||||
if err := json.Unmarshal([]byte(s), &v); err != nil {
|
||||
return nil, false
|
||||
}
|
||||
obj, ok := v.(map[string]any)
|
||||
return obj, ok
|
||||
}
|
||||
|
||||
func trim(s string, n int) string {
|
||||
if len(s) <= n {
|
||||
return s
|
||||
}
|
||||
return s[:n]
|
||||
}
|
||||
@@ -0,0 +1,189 @@
|
||||
package validator
|
||||
|
||||
import (
|
||||
"context"
|
||||
"errors"
|
||||
"testing"
|
||||
)
|
||||
|
||||
func TestExtractJSON_FromFencedBlock(t *testing.T) {
|
||||
raw := "Here's my answer:\n```json\n{\"fills\": [{\"candidate_id\": \"W-1\"}]}\n```\nDone."
|
||||
v := ExtractJSON(raw)
|
||||
if v == nil {
|
||||
t.Fatal("expected match in fenced block")
|
||||
}
|
||||
if _, ok := v["fills"]; !ok {
|
||||
t.Errorf("missing fills key: %+v", v)
|
||||
}
|
||||
}
|
||||
|
||||
func TestExtractJSON_FromBareBraces(t *testing.T) {
|
||||
raw := "Here you go: {\"fills\": [{\"candidate_id\": \"W-2\"}]}"
|
||||
v := ExtractJSON(raw)
|
||||
if v == nil {
|
||||
t.Fatal("expected match in bare braces")
|
||||
}
|
||||
}
|
||||
|
||||
func TestExtractJSON_ReturnsNilOnNoObject(t *testing.T) {
|
||||
if v := ExtractJSON("just prose, no json"); v != nil {
|
||||
t.Errorf("expected nil, got %+v", v)
|
||||
}
|
||||
}
|
||||
|
||||
func TestExtractJSON_PicksFirstBalancedObject(t *testing.T) {
|
||||
v := ExtractJSON(`{"a":1} then {"b":2}`)
|
||||
if v == nil {
|
||||
t.Fatal("expected match")
|
||||
}
|
||||
if v["a"].(float64) != 1 {
|
||||
t.Errorf("expected first object, got %+v", v)
|
||||
}
|
||||
}
|
||||
|
||||
func TestExtractJSON_NestedBalancedObjects(t *testing.T) {
|
||||
v := ExtractJSON(`prefix {"outer": {"inner": [1,2,3]}, "x": "y"} suffix`)
|
||||
if v == nil {
|
||||
t.Fatal("expected match on balanced nested object")
|
||||
}
|
||||
if outer, ok := v["outer"].(map[string]any); !ok || outer["inner"] == nil {
|
||||
t.Errorf("nested structure lost: %+v", v)
|
||||
}
|
||||
}
|
||||
|
||||
func TestExtractJSON_TopLevelArrayReturnsFirstInnerObject(t *testing.T) {
|
||||
// Both Rust and Go runtimes accept the first balanced {...} as a
|
||||
// successful match — for `[{"a":1},{"b":2}]` that's the first
|
||||
// inner object. Documenting this so the contract stays consistent
|
||||
// across runtimes.
|
||||
v := ExtractJSON(`[{"a":1},{"b":2}]`)
|
||||
if v == nil {
|
||||
t.Fatal("expected first inner object to be returned")
|
||||
}
|
||||
if v["a"].(float64) != 1 {
|
||||
t.Errorf("expected first object {a:1}, got %+v", v)
|
||||
}
|
||||
}
|
||||
|
||||
// ─── Iterate orchestrator tests with scripted ChatCaller ────────────
|
||||
|
||||
func scriptedChat(responses ...string) (ChatCaller, *int) {
|
||||
idx := 0
|
||||
return func(_ context.Context, _, _ string, _, _ string, _ *float64, _ int) (string, error) {
|
||||
if idx >= len(responses) {
|
||||
return "", errors.New("scripted chat exhausted")
|
||||
}
|
||||
r := responses[idx]
|
||||
idx++
|
||||
return r, nil
|
||||
}, &idx
|
||||
}
|
||||
|
||||
func TestIterate_AcceptsFirstValidArtifact(t *testing.T) {
|
||||
chat, calls := scriptedChat(`{"endorsed_names":["W-1"]}`)
|
||||
validate := func(_ string, _ map[string]any) (Report, error) {
|
||||
return Report{ElapsedMs: 1}, nil
|
||||
}
|
||||
resp, fail, err := Iterate(context.Background(),
|
||||
IterateRequest{Kind: "playbook", Prompt: "produce X", Provider: "ollama", Model: "qwen3.5:latest"},
|
||||
IterateConfig{}, chat, validate)
|
||||
if err != nil || fail != nil {
|
||||
t.Fatalf("expected success, got err=%v fail=%+v", err, fail)
|
||||
}
|
||||
if resp.Iterations != 1 {
|
||||
t.Errorf("iterations = %d, want 1", resp.Iterations)
|
||||
}
|
||||
if len(resp.History) != 1 || resp.History[0].Status.Kind != "accepted" {
|
||||
t.Errorf("history: %+v", resp.History)
|
||||
}
|
||||
if *calls != 1 {
|
||||
t.Errorf("expected 1 chat call, got %d", *calls)
|
||||
}
|
||||
}
|
||||
|
||||
func TestIterate_RetriesOnNoJsonThenSucceeds(t *testing.T) {
|
||||
chat, _ := scriptedChat(
|
||||
"sorry I cannot do that",
|
||||
`{"endorsed_names":["W-1"]}`,
|
||||
)
|
||||
validate := func(_ string, _ map[string]any) (Report, error) {
|
||||
return Report{}, nil
|
||||
}
|
||||
resp, _, err := Iterate(context.Background(),
|
||||
IterateRequest{Kind: "playbook", Prompt: "produce X", Provider: "ollama", Model: "x"},
|
||||
IterateConfig{}, chat, validate)
|
||||
if err != nil || resp == nil {
|
||||
t.Fatalf("expected success, err=%v", err)
|
||||
}
|
||||
if resp.Iterations != 2 {
|
||||
t.Errorf("iterations = %d, want 2", resp.Iterations)
|
||||
}
|
||||
if resp.History[0].Status.Kind != "no_json" {
|
||||
t.Errorf("first history status: %+v", resp.History[0].Status)
|
||||
}
|
||||
}
|
||||
|
||||
func TestIterate_RetriesOnValidationFailureThenSucceeds(t *testing.T) {
|
||||
chat, _ := scriptedChat(
|
||||
`{"bad":"shape"}`,
|
||||
`{"good":"shape"}`,
|
||||
)
|
||||
calls := 0
|
||||
validate := func(_ string, body map[string]any) (Report, error) {
|
||||
calls++
|
||||
if _, ok := body["good"]; ok {
|
||||
return Report{}, nil
|
||||
}
|
||||
return Report{}, &ValidationError{Kind: ErrSchema, Field: "x", Reason: "missing good"}
|
||||
}
|
||||
resp, _, err := Iterate(context.Background(),
|
||||
IterateRequest{Kind: "playbook", Prompt: "produce X", Provider: "ollama", Model: "x"},
|
||||
IterateConfig{}, chat, validate)
|
||||
if err != nil || resp == nil {
|
||||
t.Fatalf("expected success, err=%v", err)
|
||||
}
|
||||
if calls != 2 {
|
||||
t.Errorf("validate calls = %d, want 2", calls)
|
||||
}
|
||||
if resp.History[0].Status.Kind != "validation_failed" {
|
||||
t.Errorf("first history status: %+v", resp.History[0].Status)
|
||||
}
|
||||
if resp.History[0].Status.Error == "" {
|
||||
t.Errorf("validation_failed entry must carry error string")
|
||||
}
|
||||
}
|
||||
|
||||
func TestIterate_MaxIterationsExhaustedReturnsFailure(t *testing.T) {
|
||||
chat, _ := scriptedChat(`{}`, `{}`, `{}`)
|
||||
validate := func(_ string, _ map[string]any) (Report, error) {
|
||||
return Report{}, &ValidationError{Kind: ErrCompleteness, Reason: "always wrong"}
|
||||
}
|
||||
resp, fail, err := Iterate(context.Background(),
|
||||
IterateRequest{Kind: "playbook", Prompt: "X", Provider: "ollama", Model: "x", MaxIterations: 3},
|
||||
IterateConfig{}, chat, validate)
|
||||
if err != nil {
|
||||
t.Fatalf("infrastructure error unexpected: %v", err)
|
||||
}
|
||||
if resp != nil {
|
||||
t.Fatalf("expected failure, got %+v", resp)
|
||||
}
|
||||
if fail.Iterations != 3 {
|
||||
t.Errorf("iterations = %d, want 3", fail.Iterations)
|
||||
}
|
||||
if len(fail.History) != 3 {
|
||||
t.Errorf("history length = %d, want 3", len(fail.History))
|
||||
}
|
||||
}
|
||||
|
||||
func TestIterate_PropagatesChatInfraError(t *testing.T) {
|
||||
chat := func(_ context.Context, _, _ string, _, _ string, _ *float64, _ int) (string, error) {
|
||||
return "", errors.New("connection refused")
|
||||
}
|
||||
validate := func(_ string, _ map[string]any) (Report, error) { return Report{}, nil }
|
||||
_, _, err := Iterate(context.Background(),
|
||||
IterateRequest{Kind: "playbook", Prompt: "X", Provider: "ollama", Model: "x"},
|
||||
IterateConfig{}, chat, validate)
|
||||
if err == nil {
|
||||
t.Fatal("expected infrastructure error to surface")
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,86 @@
|
||||
package validator
|
||||
|
||||
import (
|
||||
"bufio"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"os"
|
||||
"strings"
|
||||
)
|
||||
|
||||
// rosterRow is the on-disk shape of one line in a roster JSONL.
|
||||
// Fields are tolerant — string-valued city/state/role become *string
|
||||
// on WorkerRecord; absent or null fields stay nil so the validators
|
||||
// know "we don't know" vs "we know it's empty."
|
||||
//
|
||||
// Mirrors the projection used in the Rust ParquetWorkerLookup so
|
||||
// JSONL exported from `workers_500k.parquet` (or a synthetic dataset)
|
||||
// loads here without translation. Producer:
|
||||
//
|
||||
// duckdb -c "COPY (SELECT candidate_id, name, status, city, state,
|
||||
// role, blacklisted_clients FROM workers) TO 'roster.jsonl'
|
||||
// (FORMAT JSON, ARRAY false)"
|
||||
type rosterRow struct {
|
||||
CandidateID string `json:"candidate_id"`
|
||||
Name string `json:"name"`
|
||||
Status string `json:"status"`
|
||||
City *string `json:"city"`
|
||||
State *string `json:"state"`
|
||||
Role *string `json:"role"`
|
||||
BlacklistedClients []string `json:"blacklisted_clients"`
|
||||
}
|
||||
|
||||
// LoadJSONLRoster reads a roster JSONL file and returns an
|
||||
// InMemoryWorkerLookup. The validators accept any WorkerLookup, so
|
||||
// callers that need a different backing store (e.g. queryd-backed
|
||||
// lookup against the live Parquet view) can plug in their own
|
||||
// implementation without changing this function.
|
||||
//
|
||||
// Parse errors on individual lines are skipped, not fatal — the
|
||||
// roster is operator-supplied and a corrupted line shouldn't
|
||||
// disable the whole validator surface. The return error is for
|
||||
// I/O failures (path missing, unreadable).
|
||||
//
|
||||
// Empty path returns an empty lookup + nil — gives the daemon a
|
||||
// "no roster configured" mode where worker-existence checks fail
|
||||
// Consistency. Matches the Rust gateway's default.
|
||||
func LoadJSONLRoster(path string) (*InMemoryWorkerLookup, error) {
|
||||
if path == "" {
|
||||
return NewInMemoryWorkerLookup(nil), nil
|
||||
}
|
||||
f, err := os.Open(path)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("open roster: %w", err)
|
||||
}
|
||||
defer f.Close()
|
||||
|
||||
var records []WorkerRecord
|
||||
scanner := bufio.NewScanner(f)
|
||||
scanner.Buffer(make([]byte, 0, 1<<16), 1<<24)
|
||||
for scanner.Scan() {
|
||||
line := scanner.Bytes()
|
||||
if len(line) == 0 {
|
||||
continue
|
||||
}
|
||||
var row rosterRow
|
||||
if err := json.Unmarshal(line, &row); err != nil {
|
||||
continue // tolerate malformed lines
|
||||
}
|
||||
if strings.TrimSpace(row.CandidateID) == "" {
|
||||
continue
|
||||
}
|
||||
records = append(records, WorkerRecord{
|
||||
CandidateID: row.CandidateID,
|
||||
Name: row.Name,
|
||||
Status: row.Status,
|
||||
City: row.City,
|
||||
State: row.State,
|
||||
Role: row.Role,
|
||||
BlacklistedClients: row.BlacklistedClients,
|
||||
})
|
||||
}
|
||||
if err := scanner.Err(); err != nil {
|
||||
return nil, fmt.Errorf("scan roster: %w", err)
|
||||
}
|
||||
return NewInMemoryWorkerLookup(records), nil
|
||||
}
|
||||
@@ -0,0 +1,64 @@
|
||||
package validator
|
||||
|
||||
import (
|
||||
"os"
|
||||
"path/filepath"
|
||||
"testing"
|
||||
)
|
||||
|
||||
func TestLoadJSONLRoster_RoundTripFields(t *testing.T) {
|
||||
dir := t.TempDir()
|
||||
path := filepath.Join(dir, "roster.jsonl")
|
||||
body := `{"candidate_id":"W-1","name":"Ada","status":"active","city":"Toledo","state":"OH","role":"Welder","blacklisted_clients":["C-1"]}
|
||||
{"candidate_id":"W-2","name":"Bea","status":"inactive","city":null,"state":null,"role":null,"blacklisted_clients":[]}
|
||||
malformed line that should be skipped
|
||||
{"candidate_id":"","name":"empty id","status":"active"}
|
||||
`
|
||||
if err := os.WriteFile(path, []byte(body), 0o644); err != nil {
|
||||
t.Fatalf("write fixture: %v", err)
|
||||
}
|
||||
|
||||
l, err := LoadJSONLRoster(path)
|
||||
if err != nil {
|
||||
t.Fatalf("load: %v", err)
|
||||
}
|
||||
if l.Len() != 2 {
|
||||
t.Fatalf("expected 2 records (skip malformed + empty id), got %d", l.Len())
|
||||
}
|
||||
|
||||
w1, ok := l.Find("W-1")
|
||||
if !ok {
|
||||
t.Fatal("missing W-1")
|
||||
}
|
||||
if w1.City == nil || *w1.City != "Toledo" || w1.Role == nil || *w1.Role != "Welder" {
|
||||
t.Errorf("W-1 fields: %+v", w1)
|
||||
}
|
||||
if len(w1.BlacklistedClients) != 1 || w1.BlacklistedClients[0] != "C-1" {
|
||||
t.Errorf("W-1 blacklist: %+v", w1.BlacklistedClients)
|
||||
}
|
||||
|
||||
w2, ok := l.Find("w-2") // case-insensitive
|
||||
if !ok {
|
||||
t.Fatal("missing W-2 (case-insensitive)")
|
||||
}
|
||||
if w2.City != nil || w2.State != nil || w2.Role != nil {
|
||||
t.Errorf("W-2 should have nil pointers for missing fields: %+v", w2)
|
||||
}
|
||||
}
|
||||
|
||||
func TestLoadJSONLRoster_EmptyPathReturnsEmptyLookup(t *testing.T) {
|
||||
l, err := LoadJSONLRoster("")
|
||||
if err != nil {
|
||||
t.Fatalf("empty path should not error: %v", err)
|
||||
}
|
||||
if l.Len() != 0 {
|
||||
t.Errorf("expected empty lookup, got len=%d", l.Len())
|
||||
}
|
||||
}
|
||||
|
||||
func TestLoadJSONLRoster_MissingFileErrors(t *testing.T) {
|
||||
_, err := LoadJSONLRoster("/nonexistent/path/roster.jsonl")
|
||||
if err == nil {
|
||||
t.Fatal("expected error for missing path")
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,132 @@
|
||||
package validator
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"strings"
|
||||
"time"
|
||||
)
|
||||
|
||||
// PlaybookValidator is the Go port of Rust's
|
||||
// `crates/validator/src/staffing/playbook.rs`. Sealed playbook
|
||||
// validation per Phase 25:
|
||||
//
|
||||
// - Operation must be a non-empty string starting with `fill:`
|
||||
// - endorsed_names must be a non-empty array, ≤ target_count × 2
|
||||
// - fingerprint must be non-empty (validity-window requirement)
|
||||
//
|
||||
// PlaybookValidator is stateless — no WorkerLookup dependency, unlike
|
||||
// FillValidator and EmailValidator. The whole validation runs on the
|
||||
// artifact body alone.
|
||||
type PlaybookValidator struct{}
|
||||
|
||||
// NewPlaybookValidator returns a zero-deps validator. Constructor for
|
||||
// symmetry with the other two; not strictly required.
|
||||
func NewPlaybookValidator() *PlaybookValidator { return &PlaybookValidator{} }
|
||||
|
||||
// Name satisfies Validator. Matches Rust's "staffing.playbook" so
|
||||
// audit-log scrapes work across runtimes.
|
||||
func (PlaybookValidator) Name() string { return "staffing.playbook" }
|
||||
|
||||
// Validate runs the four PRD checks. Errors abort the run; warnings
|
||||
// (none today) would attach to a passing Report.
|
||||
func (v PlaybookValidator) Validate(a Artifact) (Report, error) {
|
||||
started := time.Now()
|
||||
if a.Playbook == nil {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrSchema,
|
||||
Field: "artifact",
|
||||
Reason: fmt.Sprintf("PlaybookValidator expects Playbook, got %s", a.Kind()),
|
||||
}
|
||||
}
|
||||
body := a.Playbook
|
||||
|
||||
op, ok := stringField(body, "operation")
|
||||
if !ok {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrSchema,
|
||||
Field: "operation",
|
||||
Reason: "missing or not a string",
|
||||
}
|
||||
}
|
||||
if !strings.HasPrefix(op, "fill:") {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrSchema,
|
||||
Field: "operation",
|
||||
Reason: fmt.Sprintf("expected `fill: ...` prefix, got %q", op),
|
||||
}
|
||||
}
|
||||
|
||||
endorsed, ok := body["endorsed_names"].([]any)
|
||||
if !ok {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrSchema,
|
||||
Field: "endorsed_names",
|
||||
Reason: "missing or not an array",
|
||||
}
|
||||
}
|
||||
if len(endorsed) == 0 {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrCompleteness,
|
||||
Reason: "endorsed_names must be non-empty",
|
||||
}
|
||||
}
|
||||
|
||||
if target, ok := uintField(body, "target_count"); ok {
|
||||
max := target * 2
|
||||
if uint64(len(endorsed)) > max {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrCompleteness,
|
||||
Reason: fmt.Sprintf("endorsed_names (%d) exceeds target_count × 2 (%d)", len(endorsed), max),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if fp, _ := stringField(body, "fingerprint"); fp == "" {
|
||||
return Report{}, &ValidationError{
|
||||
Kind: ErrSchema,
|
||||
Field: "fingerprint",
|
||||
Reason: "missing — required for Phase 25 validity window",
|
||||
}
|
||||
}
|
||||
|
||||
return Report{Findings: []Finding{}, ElapsedMs: elapsed(started)}, nil
|
||||
}
|
||||
|
||||
// stringField returns (val, true) if body[key] is a string, else
|
||||
// ("", false). Matches Rust's serde_json::Value::as_str() shape.
|
||||
func stringField(body map[string]any, key string) (string, bool) {
|
||||
v, ok := body[key]
|
||||
if !ok {
|
||||
return "", false
|
||||
}
|
||||
s, ok := v.(string)
|
||||
return s, ok
|
||||
}
|
||||
|
||||
// uintField returns (val, true) if body[key] is a non-negative whole
|
||||
// number; matches Rust as_u64. JSON numbers come in as float64, which
|
||||
// is why we do the conversion explicitly.
|
||||
func uintField(body map[string]any, key string) (uint64, bool) {
|
||||
v, ok := body[key]
|
||||
if !ok || v == nil {
|
||||
return 0, false
|
||||
}
|
||||
switch t := v.(type) {
|
||||
case float64:
|
||||
if t < 0 {
|
||||
return 0, false
|
||||
}
|
||||
return uint64(t), true
|
||||
case int:
|
||||
if t < 0 {
|
||||
return 0, false
|
||||
}
|
||||
return uint64(t), true
|
||||
case int64:
|
||||
if t < 0 {
|
||||
return 0, false
|
||||
}
|
||||
return uint64(t), true
|
||||
}
|
||||
return 0, false
|
||||
}
|
||||
@@ -0,0 +1,77 @@
|
||||
package validator
|
||||
|
||||
import (
|
||||
"errors"
|
||||
"testing"
|
||||
)
|
||||
|
||||
func TestPlaybook_WellFormedPasses(t *testing.T) {
|
||||
r, err := PlaybookValidator{}.Validate(Artifact{Playbook: map[string]any{
|
||||
"operation": "fill: Welder x2 in Toledo, OH",
|
||||
"endorsed_names": []any{"W-123", "W-456"},
|
||||
"target_count": 2.0,
|
||||
"fingerprint": "abc123",
|
||||
}})
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %v", err)
|
||||
}
|
||||
if r.ElapsedMs < 0 {
|
||||
t.Errorf("elapsed_ms negative: %d", r.ElapsedMs)
|
||||
}
|
||||
}
|
||||
|
||||
func TestPlaybook_EmptyEndorsedNamesFailsCompleteness(t *testing.T) {
|
||||
_, err := PlaybookValidator{}.Validate(Artifact{Playbook: map[string]any{
|
||||
"operation": "fill: Welder x2 in Toledo, OH",
|
||||
"endorsed_names": []any{},
|
||||
"fingerprint": "abc",
|
||||
}})
|
||||
var ve *ValidationError
|
||||
if !errors.As(err, &ve) || ve.Kind != ErrCompleteness {
|
||||
t.Fatalf("expected Completeness, got %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestPlaybook_OverfullEndorsedNamesFailsCompleteness(t *testing.T) {
|
||||
_, err := PlaybookValidator{}.Validate(Artifact{Playbook: map[string]any{
|
||||
"operation": "fill: Welder x1 in Toledo, OH",
|
||||
"endorsed_names": []any{"a", "b", "c"},
|
||||
"target_count": 1.0,
|
||||
"fingerprint": "abc",
|
||||
}})
|
||||
var ve *ValidationError
|
||||
if !errors.As(err, &ve) || ve.Kind != ErrCompleteness {
|
||||
t.Fatalf("expected Completeness, got %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestPlaybook_MissingFingerprintFailsSchema(t *testing.T) {
|
||||
_, err := PlaybookValidator{}.Validate(Artifact{Playbook: map[string]any{
|
||||
"operation": "fill: X x1 in A, B",
|
||||
"endorsed_names": []any{"a"},
|
||||
}})
|
||||
var ve *ValidationError
|
||||
if !errors.As(err, &ve) || ve.Kind != ErrSchema || ve.Field != "fingerprint" {
|
||||
t.Fatalf("expected Schema/fingerprint, got %+v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestPlaybook_WrongOperationPrefixFailsSchema(t *testing.T) {
|
||||
_, err := PlaybookValidator{}.Validate(Artifact{Playbook: map[string]any{
|
||||
"operation": "sms_draft: hello",
|
||||
"endorsed_names": []any{"a"},
|
||||
"fingerprint": "x",
|
||||
}})
|
||||
var ve *ValidationError
|
||||
if !errors.As(err, &ve) || ve.Kind != ErrSchema {
|
||||
t.Fatalf("expected Schema, got %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestPlaybook_WrongArtifactKindFailsSchema(t *testing.T) {
|
||||
_, err := PlaybookValidator{}.Validate(Artifact{FillProposal: map[string]any{}})
|
||||
var ve *ValidationError
|
||||
if !errors.As(err, &ve) || ve.Kind != ErrSchema || ve.Field != "artifact" {
|
||||
t.Fatalf("expected Schema/artifact, got %+v", err)
|
||||
}
|
||||
}
|
||||
@@ -29,6 +29,8 @@ type Artifact struct {
|
||||
FillProposal map[string]any
|
||||
// EmailDraft: {to, body, subject?, kind?, _context?: {candidate_id?}}
|
||||
EmailDraft map[string]any
|
||||
// Playbook: {operation, endorsed_names, target_count?, fingerprint}
|
||||
Playbook map[string]any
|
||||
}
|
||||
|
||||
// Kind returns a short string for error messages — mirrors the
|
||||
@@ -39,6 +41,8 @@ func (a Artifact) Kind() string {
|
||||
return "FillProposal"
|
||||
case a.EmailDraft != nil:
|
||||
return "EmailDraft"
|
||||
case a.Playbook != nil:
|
||||
return "Playbook"
|
||||
default:
|
||||
return "Unknown"
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user