Add archive storage path configuration

This commit is contained in:
2026-05-16 14:11:59 +00:00
parent 58c6ab2d54
commit 0454296c81
23 changed files with 950 additions and 22 deletions

View File

@@ -12,6 +12,8 @@ type Config struct {
type PipelineConfig struct {
Workspace WorkspaceConfig `yaml:"workspace"`
Storage StorageConfig `yaml:"storage"`
Spool SpoolConfig `yaml:"spool"`
Archive *ArchiveConfig `yaml:"archive"`
Secrets *SecretsConfig `yaml:"secrets"`
WhisperX WhisperXConfig `yaml:"whisperx"`
Seriatim SeriatimConfig `yaml:"seriatim"`
@@ -44,9 +46,39 @@ type SecretsConfig struct {
// StorageConfig configures storage backends and related parameters.
type StorageConfig struct {
Backend string `yaml:"backend"`
Bucket string `yaml:"bucket"`
Prefix string `yaml:"prefix"`
Backend string `yaml:"backend"`
Bucket string `yaml:"bucket"`
Prefix string `yaml:"prefix"`
S3 *StorageS3Config `yaml:"s3"`
}
// StorageS3Config configures S3 storage coordinates.
type StorageS3Config struct {
Bucket string `yaml:"bucket"`
RootPrefix string `yaml:"root_prefix"`
Region string `yaml:"region"`
Endpoint string `yaml:"endpoint"`
ForcePathStyle bool `yaml:"force_path_style"`
}
// SpoolConfig configures local spool storage for staged data.
type SpoolConfig struct {
Root string `yaml:"root"`
DeleteAudioAfterArchive bool `yaml:"delete_audio_after_archive"`
}
// ArchiveConfig configures archive behavior and artifact promotions.
type ArchiveConfig struct {
Enabled *bool `yaml:"enabled"`
UploadRun *bool `yaml:"upload_run"`
PromoteArtifacts []ArchivePromotionRule `yaml:"promote_artifacts"`
}
// ArchivePromotionRule configures one artifact promotion mapping.
type ArchivePromotionRule struct {
From string `yaml:"from"`
To string `yaml:"to"`
Required *bool `yaml:"required"`
}
// WhisperXConfig configures WhisperX adapter settings.
@@ -180,9 +212,15 @@ type ArtifactSettings struct {
// SessionInputsConfig contains per-session input references.
type SessionInputsConfig struct {
AudioDir string `yaml:"audio_dir"`
AudioFiles []string `yaml:"audio_files"`
SpeakersFile string `yaml:"speakers_file"`
AutocorrectFile string `yaml:"autocorrect_file"`
GlossaryFile string `yaml:"glossary_file"`
AudioDir string `yaml:"audio_dir"`
AudioFiles []string `yaml:"audio_files"`
AudioS3 *SessionAudioS3Input `yaml:"audio_s3"`
SpeakersFile string `yaml:"speakers_file"`
AutocorrectFile string `yaml:"autocorrect_file"`
GlossaryFile string `yaml:"glossary_file"`
}
// SessionAudioS3Input configures S3 session-audio input discovery.
type SessionAudioS3Input struct {
Prefix string `yaml:"prefix"`
}

View File

@@ -81,6 +81,9 @@ func applyPipelineDefaults(cfg *PipelineConfig) {
if cfg == nil {
return
}
applyStorageDefaults(&cfg.Storage)
applySpoolDefaults(&cfg.Spool)
applyArchiveDefaults(&cfg.Archive)
applyWhisperXDefaults(&cfg.WhisperX)
applySeriatimDefaults(&cfg.Seriatim)
applyAuditaDefaults(&cfg.Audita)
@@ -92,6 +95,54 @@ func applyPipelineDefaults(cfg *PipelineConfig) {
applyScriptoriumDefaults(cfg.Scriptorium)
}
func applyStorageDefaults(cfg *StorageConfig) {
if cfg == nil {
return
}
if cfg.S3 == nil {
cfg.S3 = &StorageS3Config{}
}
if cfg.S3.RootPrefix == "" {
cfg.S3.RootPrefix = "dnd"
}
}
func applySpoolDefaults(cfg *SpoolConfig) {
if cfg == nil {
return
}
if cfg.Root == "" {
cfg.Root = "/var/spool/narratio"
}
}
func applyArchiveDefaults(cfg **ArchiveConfig) {
if cfg == nil {
return
}
if *cfg == nil {
*cfg = &ArchiveConfig{}
}
if (*cfg).Enabled == nil {
(*cfg).Enabled = boolPtr(true)
}
if (*cfg).UploadRun == nil {
(*cfg).UploadRun = boolPtr(true)
}
if len((*cfg).PromoteArtifacts) == 0 {
(*cfg).PromoteArtifacts = []ArchivePromotionRule{
{From: "transcripts/trimmed.json", To: "transcripts/trimmed.json", Required: boolPtr(true)},
{From: "artifacts/session_recap.md", To: "artifacts/session_recap.md", Required: boolPtr(true)},
}
}
for i := range (*cfg).PromoteArtifacts {
if (*cfg).PromoteArtifacts[i].Required == nil {
(*cfg).PromoteArtifacts[i].Required = boolPtr(true)
}
}
}
func applyWhisperXDefaults(cfg *WhisperXConfig) {
if cfg == nil {
return

View File

@@ -820,6 +820,7 @@ func TestValidateMissingAudioSource(t *testing.T) {
},
Session: &SessionConfig{
SessionID: "2026-05-03",
Campaign: "sample-campaign",
Inputs: SessionInputsConfig{
SpeakersFile: "speakers.yml",
AutocorrectFile: "autocorrect.yml",
@@ -832,7 +833,7 @@ func TestValidateMissingAudioSource(t *testing.T) {
if err == nil {
t.Fatal("expected validation error, got nil")
}
if !strings.Contains(err.Error(), "audio_dir or at least one audio_files") {
if !strings.Contains(err.Error(), "audio_dir, at least one audio_files entry, or audio_s3") {
t.Fatalf("error = %q, want audio source guidance", err.Error())
}
if !strings.Contains(err.Error(), "session config") {
@@ -861,6 +862,12 @@ func writeConfigFiles(t *testing.T, pipelineYAML, sessionYAML string) (string, s
}
pipelineYAML += "audita:\n binary: audita\n"
}
if !strings.Contains(sessionYAML, "\ncampaign:") && !strings.HasPrefix(sessionYAML, "campaign:") {
if !strings.HasSuffix(sessionYAML, "\n") {
sessionYAML += "\n"
}
sessionYAML += "campaign: sample-campaign\n"
}
dir := t.TempDir()
pipelinePath := filepath.Join(dir, "pipeline.yml")

View File

@@ -0,0 +1,235 @@
package config
import (
"strings"
"testing"
)
func TestStorageS3DefaultsAndValidation(t *testing.T) {
pipelineYAML := testPipelineBaseYAML + `
storage:
backend: s3
s3:
bucket: my-dnd-archive
`
pipelinePath, sessionPath := writeConfigFiles(t, pipelineYAML, testSessionBaseYAML)
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load() error = %v", err)
}
if cfg.Pipeline.Storage.S3 == nil {
t.Fatal("storage.s3 should be initialized")
}
if cfg.Pipeline.Storage.S3.RootPrefix != "dnd" {
t.Fatalf("storage.s3.root_prefix = %q, want dnd", cfg.Pipeline.Storage.S3.RootPrefix)
}
if cfg.Pipeline.Storage.S3.ForcePathStyle {
t.Fatalf("storage.s3.force_path_style = true, want false default")
}
if err := Validate(cfg); err != nil {
t.Fatalf("Validate() error = %v", err)
}
}
func TestSpoolAndArchiveDefaults(t *testing.T) {
pipelinePath, sessionPath := writeConfigFiles(t, testPipelineBaseYAML, testSessionBaseYAML)
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load() error = %v", err)
}
if cfg.Pipeline.Spool.Root != "/var/spool/narratio" {
t.Fatalf("spool.root = %q, want /var/spool/narratio", cfg.Pipeline.Spool.Root)
}
if cfg.Pipeline.Spool.DeleteAudioAfterArchive {
t.Fatalf("spool.delete_audio_after_archive = true, want false")
}
if cfg.Pipeline.Archive == nil {
t.Fatal("archive should be initialized by defaults")
}
if cfg.Pipeline.Archive.Enabled == nil || !*cfg.Pipeline.Archive.Enabled {
t.Fatalf("archive.enabled = %#v, want true", cfg.Pipeline.Archive.Enabled)
}
if cfg.Pipeline.Archive.UploadRun == nil || !*cfg.Pipeline.Archive.UploadRun {
t.Fatalf("archive.upload_run = %#v, want true", cfg.Pipeline.Archive.UploadRun)
}
if len(cfg.Pipeline.Archive.PromoteArtifacts) != 2 {
t.Fatalf("archive.promote_artifacts len = %d, want 2 defaults", len(cfg.Pipeline.Archive.PromoteArtifacts))
}
for i, item := range cfg.Pipeline.Archive.PromoteArtifacts {
if item.Required == nil || !*item.Required {
t.Fatalf("archive.promote_artifacts[%d].required = %#v, want true", i, item.Required)
}
}
}
func TestArchivePromotionPathValidation(t *testing.T) {
tests := []struct {
name string
ruleYML string
wantErr string
}{
{
name: "absolute from path rejected",
ruleYML: `archive:
promote_artifacts:
- from: "/transcripts/trimmed.json"
to: "transcripts/trimmed.json"
`,
wantErr: "must be a relative path",
},
{
name: "traversal to path rejected",
ruleYML: `archive:
promote_artifacts:
- from: "transcripts/trimmed.json"
to: "../trimmed.json"
`,
wantErr: "must not contain path traversal",
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
pipelineYAML := testPipelineBaseYAML + "\n" + tt.ruleYML
pipelinePath, sessionPath := writeConfigFiles(t, pipelineYAML, testSessionBaseYAML)
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load() error = %v", err)
}
err = Validate(cfg)
if err == nil || !strings.Contains(err.Error(), tt.wantErr) {
t.Fatalf("Validate() error = %v, want to contain %q", err, tt.wantErr)
}
})
}
}
func TestSessionAudioS3Validation(t *testing.T) {
tests := []struct {
name string
sessionYAML string
wantErr string
}{
{
name: "valid audio_s3 prefix",
sessionYAML: `session_id: 2026-05-03
campaign: forsaken
inputs:
audio_s3:
prefix: audio/
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
},
{
name: "invalid audio_s3 absolute prefix",
sessionYAML: `session_id: 2026-05-03
campaign: forsaken
inputs:
audio_s3:
prefix: /audio/
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantErr: "session.inputs.audio_s3.prefix must be a relative path",
},
{
name: "invalid audio_s3 traversal prefix",
sessionYAML: `session_id: 2026-05-03
campaign: forsaken
inputs:
audio_s3:
prefix: ../audio/
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantErr: "session.inputs.audio_s3.prefix must not contain path traversal",
},
{
name: "local and s3 audio conflict",
sessionYAML: `session_id: 2026-05-03
campaign: forsaken
inputs:
audio_dir: ./audio
audio_s3:
prefix: audio/
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantErr: "mutually exclusive",
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
pipelineYAML := testPipelineBaseYAML + `
storage:
backend: s3
s3:
bucket: my-dnd-archive
`
pipelinePath, sessionPath := writeConfigFiles(t, pipelineYAML, tt.sessionYAML)
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load() error = %v", err)
}
err = Validate(cfg)
if tt.wantErr != "" {
if err == nil || !strings.Contains(err.Error(), tt.wantErr) {
t.Fatalf("Validate() error = %v, want to contain %q", err, tt.wantErr)
}
return
}
if err != nil {
t.Fatalf("Validate() error = %v", err)
}
})
}
}
func TestStorageS3BucketRequiredWhenS3DependentFeatureEnabled(t *testing.T) {
pipelineYAML := testPipelineBaseYAML + `
storage:
backend: s3
`
sessionYAML := `session_id: 2026-05-03
campaign: forsaken
inputs:
audio_s3:
prefix: audio/
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`
pipelinePath, sessionPath := writeConfigFiles(t, pipelineYAML, sessionYAML)
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load() error = %v", err)
}
err = Validate(cfg)
if err == nil || !strings.Contains(err.Error(), "pipeline.storage.s3.bucket is required") {
t.Fatalf("Validate() error = %v, want bucket requirement", err)
}
}
func TestLocalAudioConfigStillValid(t *testing.T) {
pipelinePath, sessionPath := writeConfigFiles(t, testPipelineBaseYAML, testSessionBaseYAML)
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load() error = %v", err)
}
if err := Validate(cfg); err != nil {
t.Fatalf("Validate() error = %v", err)
}
}

View File

@@ -3,6 +3,8 @@ package config
import (
"fmt"
"net/url"
"path/filepath"
"regexp"
"strings"
"time"
)
@@ -25,6 +27,9 @@ func Validate(cfg *Config) error {
if err := validateSession(cfg.Session); err != nil {
return fmt.Errorf("session config %q invalid: %w", shortName(cfg.SessionPath, "session.yml"), err)
}
if err := validateCrossConfig(cfg.Pipeline, cfg.Session); err != nil {
return fmt.Errorf("pipeline/session config invalid: %w", err)
}
return nil
}
@@ -36,6 +41,15 @@ func validatePipeline(cfg *PipelineConfig) error {
if err := validateSecrets(cfg.Secrets); err != nil {
return err
}
if err := validateStorage(cfg.Storage); err != nil {
return err
}
if err := validateSpool(cfg.Spool); err != nil {
return err
}
if err := validateArchive(cfg.Archive); err != nil {
return err
}
if err := validateWhisperX(cfg.WhisperX); err != nil {
return err
}
@@ -64,6 +78,48 @@ func validatePipeline(cfg *PipelineConfig) error {
return nil
}
func validateStorage(cfg StorageConfig) error {
if cfg.S3 == nil {
return nil
}
if strings.TrimSpace(cfg.S3.RootPrefix) == "" {
return fmt.Errorf("pipeline.storage.s3.root_prefix must be non-empty")
}
if err := validateRelativeSafePath("pipeline.storage.s3.root_prefix", cfg.S3.RootPrefix); err != nil {
return err
}
if cfg.S3.Endpoint != "" && strings.TrimSpace(cfg.S3.Endpoint) == "" {
return fmt.Errorf("pipeline.storage.s3.endpoint must be non-empty when provided")
}
return nil
}
func validateSpool(cfg SpoolConfig) error {
return nil
}
func validateArchive(cfg *ArchiveConfig) error {
if cfg == nil {
return nil
}
for i, item := range cfg.PromoteArtifacts {
prefix := fmt.Sprintf("pipeline.archive.promote_artifacts[%d]", i)
if strings.TrimSpace(item.From) == "" {
return fmt.Errorf("%s.from is required", prefix)
}
if strings.TrimSpace(item.To) == "" {
return fmt.Errorf("%s.to is required", prefix)
}
if err := validateRelativeSafePath(prefix+".from", item.From); err != nil {
return err
}
if err := validateRelativeSafePath(prefix+".to", item.To); err != nil {
return err
}
}
return nil
}
func validateSecrets(cfg *SecretsConfig) error {
if cfg == nil {
return nil
@@ -307,6 +363,9 @@ func validateSession(cfg *SessionConfig) error {
if strings.TrimSpace(cfg.SessionID) == "" {
return fmt.Errorf("session.session_id is required")
}
if strings.TrimSpace(cfg.Campaign) == "" {
return fmt.Errorf("session.campaign is required")
}
if strings.TrimSpace(cfg.Inputs.SpeakersFile) == "" {
return fmt.Errorf("session.inputs.speakers_file is required")
@@ -320,13 +379,79 @@ func validateSession(cfg *SessionConfig) error {
hasAudioDir := strings.TrimSpace(cfg.Inputs.AudioDir) != ""
hasAudioFiles := len(cfg.Inputs.AudioFiles) > 0
if !hasAudioDir && !hasAudioFiles {
return fmt.Errorf("session.inputs requires audio_dir or at least one audio_files entry")
hasAudioS3 := cfg.Inputs.AudioS3 != nil
if hasAudioS3 {
if strings.TrimSpace(cfg.Inputs.AudioS3.Prefix) == "" {
return fmt.Errorf("session.inputs.audio_s3.prefix is required when session.inputs.audio_s3 is configured")
}
if err := validateRelativeSafePath("session.inputs.audio_s3.prefix", cfg.Inputs.AudioS3.Prefix); err != nil {
return err
}
}
if hasAudioS3 && (hasAudioDir || hasAudioFiles) {
return fmt.Errorf("session.inputs.audio_dir/audio_files and session.inputs.audio_s3 are mutually exclusive")
}
if !hasAudioDir && !hasAudioFiles && !hasAudioS3 {
return fmt.Errorf("session.inputs requires audio_dir, at least one audio_files entry, or audio_s3")
}
return nil
}
func validateCrossConfig(pipeline *PipelineConfig, session *SessionConfig) error {
if pipeline == nil || session == nil {
return nil
}
if pipeline.Storage.S3 == nil {
return nil
}
audioS3Enabled := session.Inputs.AudioS3 != nil
archiveUploadEnabled := archiveUploadConfiguredForS3(pipeline)
if (audioS3Enabled || archiveUploadEnabled) && strings.TrimSpace(pipeline.Storage.S3.Bucket) == "" {
return fmt.Errorf("pipeline.storage.s3.bucket is required when S3 session audio or archive upload is enabled")
}
return nil
}
func archiveUploadConfiguredForS3(pipeline *PipelineConfig) bool {
if pipeline == nil || pipeline.Archive == nil {
return false
}
if !strings.EqualFold(strings.TrimSpace(pipeline.Storage.Backend), "s3") {
return false
}
enabled := true
if pipeline.Archive.Enabled != nil {
enabled = *pipeline.Archive.Enabled
}
upload := true
if pipeline.Archive.UploadRun != nil {
upload = *pipeline.Archive.UploadRun
}
return enabled && upload
}
var windowsAbsPathRE = regexp.MustCompile(`^[A-Za-z]:[\\/].*`)
func validateRelativeSafePath(fieldName, value string) error {
trimmed := strings.TrimSpace(value)
if trimmed == "" {
return fmt.Errorf("%s must be non-empty", fieldName)
}
if filepath.IsAbs(trimmed) || strings.HasPrefix(trimmed, "/") || strings.HasPrefix(trimmed, "\\") || windowsAbsPathRE.MatchString(trimmed) {
return fmt.Errorf("%s must be a relative path", fieldName)
}
normalized := strings.ReplaceAll(trimmed, "\\", "/")
for _, segment := range strings.Split(normalized, "/") {
if segment == ".." {
return fmt.Errorf("%s must not contain path traversal", fieldName)
}
}
return nil
}
func validateDuration(fieldName, value string) error {
trimmed := strings.TrimSpace(value)
if trimmed == "" {