Add Scriptorium configuration contract

This commit is contained in:
2026-05-06 21:12:22 +00:00
parent ce01ca7619
commit 84b0f6fe0e
6 changed files with 344 additions and 1 deletions

View File

@@ -8,7 +8,7 @@ This repository currently contains a **working scaffold** with strict config loa
`narratio` expects two YAML files:
- `pipeline.yml`: pipeline/workspace settings (`workspace`, `storage`, `whisperx`, `seriatim`, `audita`, `analyzer`, `notification`)
- `pipeline.yml`: pipeline/workspace settings (`workspace`, `storage`, `whisperx`, `seriatim`, `audita`, optional `scriptorium`, `analyzer`, `notification`)
- `session.yml`: per-session settings (`session_id`, `inputs`, optional metadata)
WhisperX config contract in `pipeline.yml`:
@@ -43,6 +43,24 @@ Audita runtime note:
- Narratio currently passes primary LLM concurrency via `AUDITA_LLM_CONCURRENCY` subprocess environment override, not a `--llm-concurrency` flag.
Scriptorium config contract in `pipeline.yml` (optional until `analyze` is implemented):
- optional section: `scriptorium` (when omitted, current pipeline behavior is unchanged)
- required when `scriptorium` is present: `scriptorium.binary` (name on PATH or full path)
- optional: `scriptorium.config_path` (if set, it must be non-empty; file existence is validated at execution time, not basic config validation time)
- defaulted when omitted inside present `scriptorium` section: `scriptorium.timeout` (`10m`)
- optional: `scriptorium.render_debug` (`false` by default through YAML zero-value behavior)
- artifact definitions live under `scriptorium.artifacts.<artifact_name>` and support multiple configured artifacts
- enabled artifacts require `prompt_id` and `output_path`
- artifact `timeout` values must be valid Go durations when provided
- input definitions live under `scriptorium.artifacts.<artifact_name>.inputs.<input_name>` and require non-empty `source`
- `vars` values currently accept booleans and strings only
- Narratio does not store secrets in Scriptorium config; keep credentials in environment management, not YAML
Scriptorium initial artifact note:
- this pass does not auto-create `session_recap`; configure `scriptorium.artifacts.session_recap` explicitly in `pipeline.yml`
`speakers.yml` note:
- use Seriatims documented `match:` format (not the legacy direct mapping style used by older scripts/scaffolds)

View File

@@ -43,6 +43,34 @@ audita:
validation_llm_concurrency: 1
report: true
scriptorium:
binary: "scriptorium"
config_path: "/etc/scriptorium/config.yml"
timeout: "10m"
render_debug: false
artifacts:
session_recap:
enabled: true
prompt_id: "dnd.session_recap"
profile_id: "local-quality"
output_path: "artifacts/session_recap.md"
timeout: "10m"
inputs:
transcript:
source: "processed_transcript"
required: true
previous_recap:
source: "previous_session_artifact"
artifact: "session_recap"
path: ""
required: false
vars:
session_id: true
session_date: true
campaign_name: true
previous_session_id: true
output_kind: "session_recap"
analyzer:
timeout: 20m
artifacts:

View File

@@ -15,6 +15,7 @@ type PipelineConfig struct {
WhisperX WhisperXConfig `yaml:"whisperx"`
Seriatim SeriatimConfig `yaml:"seriatim"`
Audita AuditaConfig `yaml:"audita"`
Scriptorium *ScriptoriumConfig `yaml:"scriptorium"`
Analyzer AnalyzerConfig `yaml:"analyzer"`
Notification NotificationConfig `yaml:"notification"`
}
@@ -82,6 +83,34 @@ type AuditaConfig struct {
Report *bool `yaml:"report"`
}
// ScriptoriumConfig configures Scriptorium-backed artifact generation.
type ScriptoriumConfig struct {
Binary string `yaml:"binary"`
ConfigPath string `yaml:"config_path"`
Timeout string `yaml:"timeout"`
RenderDebug bool `yaml:"render_debug"`
Artifacts map[string]ScriptoriumArtifactConfig `yaml:"artifacts"`
}
// ScriptoriumArtifactConfig configures one named output artifact workflow.
type ScriptoriumArtifactConfig struct {
Enabled bool `yaml:"enabled"`
PromptID string `yaml:"prompt_id"`
ProfileID string `yaml:"profile_id"`
OutputPath string `yaml:"output_path"`
Timeout string `yaml:"timeout"`
Inputs map[string]ScriptoriumInputConfig `yaml:"inputs"`
Vars map[string]any `yaml:"vars"`
}
// ScriptoriumInputConfig configures one named prompt input source.
type ScriptoriumInputConfig struct {
Source string `yaml:"source"`
Artifact string `yaml:"artifact"`
Path string `yaml:"path"`
Required bool `yaml:"required"`
}
// AnalyzerConfig configures analyzer adapter settings.
type AnalyzerConfig struct {
BinaryPath string `yaml:"binary_path"`

View File

@@ -84,6 +84,7 @@ func applyPipelineDefaults(cfg *PipelineConfig) {
applyWhisperXDefaults(&cfg.WhisperX)
applySeriatimDefaults(&cfg.Seriatim)
applyAuditaDefaults(&cfg.Audita)
applyScriptoriumDefaults(cfg.Scriptorium)
}
func applyWhisperXDefaults(cfg *WhisperXConfig) {
@@ -165,6 +166,15 @@ func applyAuditaDefaults(cfg *AuditaConfig) {
}
}
func applyScriptoriumDefaults(cfg *ScriptoriumConfig) {
if cfg == nil {
return
}
if cfg.Timeout == "" {
cfg.Timeout = "10m"
}
}
func float64Ptr(v float64) *float64 {
p := v
return &p

View File

@@ -0,0 +1,203 @@
package config
import (
"strings"
"testing"
)
func TestScriptoriumLoadAndValidate(t *testing.T) {
tests := []struct {
name string
scriptoriumYAML string
wantLoadErr string
wantValidateErr string
assert func(t *testing.T, cfg *Config)
}{
{
name: "valid minimal scriptorium config",
scriptoriumYAML: `scriptorium:
binary: scriptorium
artifacts:
session_recap:
enabled: true
prompt_id: dnd.session_recap
output_path: artifacts/session_recap.md
inputs:
transcript:
source: processed_transcript
required: true
vars:
session_id: true
output_kind: session_recap
`,
assert: func(t *testing.T, cfg *Config) {
t.Helper()
if cfg.Pipeline.Scriptorium == nil {
t.Fatal("scriptorium config should be present")
}
if cfg.Pipeline.Scriptorium.Timeout != "10m" {
t.Fatalf("scriptorium.timeout default = %q, want 10m", cfg.Pipeline.Scriptorium.Timeout)
}
},
},
{
name: "unknown scriptorium field fails",
scriptoriumYAML: `scriptorium:
binary: scriptorium
bogus: true
`,
wantLoadErr: "strict decode failed",
},
{
name: "missing binary fails when section present",
scriptoriumYAML: `scriptorium:
timeout: 10m
`,
wantValidateErr: "pipeline.scriptorium.binary is required",
},
{
name: "enabled artifact missing prompt id fails",
scriptoriumYAML: `scriptorium:
binary: scriptorium
artifacts:
session_recap:
enabled: true
output_path: artifacts/session_recap.md
`,
wantValidateErr: "pipeline.scriptorium.artifacts.session_recap.prompt_id is required when enabled",
},
{
name: "enabled artifact missing output path fails",
scriptoriumYAML: `scriptorium:
binary: scriptorium
artifacts:
session_recap:
enabled: true
prompt_id: dnd.session_recap
`,
wantValidateErr: "pipeline.scriptorium.artifacts.session_recap.output_path is required when enabled",
},
{
name: "invalid timeout fails",
scriptoriumYAML: `scriptorium:
binary: scriptorium
timeout: definitely-not-a-duration
`,
wantValidateErr: "pipeline.scriptorium.timeout must be a valid duration",
},
{
name: "optional previous recap input is accepted",
scriptoriumYAML: `scriptorium:
binary: scriptorium
artifacts:
session_recap:
enabled: true
prompt_id: dnd.session_recap
output_path: artifacts/session_recap.md
inputs:
transcript:
source: processed_transcript
required: true
previous_recap:
source: previous_session_artifact
artifact: session_recap
path: ""
required: false
vars:
session_id: true
output_kind: session_recap
`,
},
{
name: "multiple artifact definitions can be decoded",
scriptoriumYAML: `scriptorium:
binary: scriptorium
artifacts:
session_recap:
enabled: true
prompt_id: dnd.session_recap
output_path: artifacts/session_recap.md
inputs:
transcript:
source: processed_transcript
required: true
player_summary:
enabled: true
prompt_id: dnd.player_summary
profile_id: local-fast
output_path: artifacts/player_summary.md
timeout: 3m
inputs:
transcript:
source: processed_transcript
required: true
`,
assert: func(t *testing.T, cfg *Config) {
t.Helper()
if cfg.Pipeline.Scriptorium == nil {
t.Fatal("scriptorium config should be present")
}
if len(cfg.Pipeline.Scriptorium.Artifacts) != 2 {
t.Fatalf("scriptorium.artifacts count = %d, want 2", len(cfg.Pipeline.Scriptorium.Artifacts))
}
},
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
pipelineYAML := testPipelineBaseYAML + "\n" + tt.scriptoriumYAML
pipelinePath, sessionPath := writeConfigFiles(t, pipelineYAML, testSessionBaseYAML)
cfg, err := Load(pipelinePath, sessionPath)
if tt.wantLoadErr != "" {
if err == nil {
t.Fatalf("expected load error containing %q, got nil", tt.wantLoadErr)
}
if !strings.Contains(err.Error(), tt.wantLoadErr) {
t.Fatalf("load error = %q, want to contain %q", err.Error(), tt.wantLoadErr)
}
return
}
if err != nil {
t.Fatalf("Load() error = %v", err)
}
if tt.assert != nil {
tt.assert(t, cfg)
}
err = Validate(cfg)
if tt.wantValidateErr != "" {
if err == nil {
t.Fatalf("expected validation error containing %q, got nil", tt.wantValidateErr)
}
if !strings.Contains(err.Error(), tt.wantValidateErr) {
t.Fatalf("validation error = %q, want to contain %q", err.Error(), tt.wantValidateErr)
}
return
}
if err != nil {
t.Fatalf("Validate() error = %v", err)
}
})
}
}
const testPipelineBaseYAML = `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
audita:
binary: audita
`
const testSessionBaseYAML = `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`

View File

@@ -42,6 +42,9 @@ func validatePipeline(cfg *PipelineConfig) error {
if err := validateAudita(cfg.Audita); err != nil {
return err
}
if err := validateScriptorium(cfg.Scriptorium); err != nil {
return err
}
if err := validateDuration("pipeline.analyzer.timeout", cfg.Analyzer.Timeout); err != nil {
return err
}
@@ -170,6 +173,58 @@ func validateAudita(cfg AuditaConfig) error {
return nil
}
func validateScriptorium(cfg *ScriptoriumConfig) error {
if cfg == nil {
return nil
}
if strings.TrimSpace(cfg.Binary) == "" {
return fmt.Errorf("pipeline.scriptorium.binary is required when pipeline.scriptorium is configured")
}
if cfg.ConfigPath != "" && strings.TrimSpace(cfg.ConfigPath) == "" {
return fmt.Errorf("pipeline.scriptorium.config_path must be non-empty when provided")
}
if err := validateDuration("pipeline.scriptorium.timeout", cfg.Timeout); err != nil {
return err
}
for artifactName, artifactCfg := range cfg.Artifacts {
trimmedArtifactName := strings.TrimSpace(artifactName)
if trimmedArtifactName == "" {
return fmt.Errorf("pipeline.scriptorium.artifacts keys must be non-empty")
}
if artifactCfg.Enabled && strings.TrimSpace(artifactCfg.PromptID) == "" {
return fmt.Errorf("pipeline.scriptorium.artifacts.%s.prompt_id is required when enabled", artifactName)
}
if artifactCfg.Enabled && strings.TrimSpace(artifactCfg.OutputPath) == "" {
return fmt.Errorf("pipeline.scriptorium.artifacts.%s.output_path is required when enabled", artifactName)
}
if err := validateDuration("pipeline.scriptorium.artifacts."+artifactName+".timeout", artifactCfg.Timeout); err != nil {
return err
}
for inputName, inputCfg := range artifactCfg.Inputs {
trimmedInputName := strings.TrimSpace(inputName)
if trimmedInputName == "" {
return fmt.Errorf("pipeline.scriptorium.artifacts.%s.inputs keys must be non-empty", artifactName)
}
if strings.TrimSpace(inputCfg.Source) == "" {
return fmt.Errorf("pipeline.scriptorium.artifacts.%s.inputs.%s.source is required", artifactName, inputName)
}
}
for varName, varValue := range artifactCfg.Vars {
if strings.TrimSpace(varName) == "" {
return fmt.Errorf("pipeline.scriptorium.artifacts.%s.vars keys must be non-empty", artifactName)
}
switch varValue.(type) {
case bool, string:
default:
return fmt.Errorf("pipeline.scriptorium.artifacts.%s.vars.%s must be a string or boolean", artifactName, varName)
}
}
}
return nil
}
func validateSession(cfg *SessionConfig) error {
if strings.TrimSpace(cfg.SessionID) == "" {
return fmt.Errorf("session.session_id is required")