Files
narratio/internal/config/load_validate_test.go

502 lines
14 KiB
Go

package config
import (
"os"
"path/filepath"
"strings"
"testing"
)
func TestLoadAndValidate(t *testing.T) {
tests := []struct {
name string
pipelineYAML string
sessionYAML string
wantLoadErr string
wantValidate string
checkDefault bool
}{
{
name: "valid minimal config",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
audita:
timeout: 1h
analyzer:
timeout: 20m
notification:
timeout: 15s
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
checkDefault: true,
},
{
name: "unknown pipeline field fails",
pipelineYAML: `workspace:
root: /tmp/narratio
bogus: true
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantLoadErr: "pipeline file",
},
{
name: "unknown whisperx field fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
bogus: true
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantLoadErr: "strict decode failed",
},
{
name: "unknown session field fails",
pipelineYAML: `workspace:
root: /tmp/narratio
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
unknown_field: true
`,
wantLoadErr: "session file",
},
{
name: "missing required field fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: ""
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "session config \"session.yml\" invalid: session.session_id is required",
},
{
name: "missing transcribe_url fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.transcribe_url is required",
},
{
name: "invalid transcribe_url fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: ://
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.transcribe_url must be a valid URL",
},
{
name: "invalid timeout fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
timeout: definitely-not-a-duration
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.timeout must be a valid duration",
},
{
name: "invalid retry_delay fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
retry_delay: not-a-duration
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.retry_delay must be a valid duration",
},
{
name: "negative retries fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
retries: -1
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.retries must be >= 0",
},
{
name: "invalid concurrency fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
concurrency: 0
seriatim:
binary: seriatim
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.concurrency must be > 0",
},
{
name: "unknown seriatim field fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
bogus: true
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantLoadErr: "strict decode failed",
},
{
name: "unknown seriatim env field fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
env:
unknown_knob: 1.0
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantLoadErr: "strict decode failed",
},
{
name: "missing seriatim binary fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
timeout: 10m
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.binary is required",
},
{
name: "invalid seriatim timeout fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
timeout: not-a-duration
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.timeout must be a valid duration",
},
{
name: "invalid seriatim output schema fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
output_schema: invalid-schema
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.output_schema must be one of: seriatim-minimal, seriatim-intermediate, seriatim-full",
},
{
name: "negative seriatim coalesce gap fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
coalesce_gap: -0.1
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.coalesce_gap must be >= 0",
},
{
name: "invalid seriatim env float fails",
pipelineYAML: `workspace:
root: /tmp/narratio
whisperx:
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
seriatim:
binary: seriatim
env:
overlap_word_run_gap: 0
`,
sessionYAML: `session_id: 2026-05-03
inputs:
audio_dir: ./audio
speakers_file: ./speakers.yml
autocorrect_file: ./autocorrect.yml
glossary_file: ./glossary.yml
`,
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.env.overlap_word_run_gap must be > 0 when provided",
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
pipelinePath, sessionPath := writeConfigFiles(t, tt.pipelineYAML, tt.sessionYAML)
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)
}
if !strings.Contains(err.Error(), "strict decode failed") {
t.Fatalf("load error = %q, want strict decode context", err.Error())
}
return
}
if err != nil {
t.Fatalf("Load() error = %v", err)
}
if cfg.PipelinePath != pipelinePath {
t.Fatalf("PipelinePath = %q, want %q", cfg.PipelinePath, pipelinePath)
}
if cfg.SessionPath != sessionPath {
t.Fatalf("SessionPath = %q, want %q", cfg.SessionPath, sessionPath)
}
if tt.checkDefault {
if cfg.Pipeline.WhisperX.Language != "en" {
t.Fatalf("whisperx.language = %q, want %q", cfg.Pipeline.WhisperX.Language, "en")
}
if cfg.Pipeline.WhisperX.Timeout != "30m" {
t.Fatalf("whisperx.timeout = %q, want %q", cfg.Pipeline.WhisperX.Timeout, "30m")
}
if cfg.Pipeline.WhisperX.RetryDelay != "2s" {
t.Fatalf("whisperx.retry_delay = %q, want %q", cfg.Pipeline.WhisperX.RetryDelay, "2s")
}
if cfg.Pipeline.WhisperX.Retries == nil || *cfg.Pipeline.WhisperX.Retries != 3 {
t.Fatalf("whisperx.retries = %v, want 3", cfg.Pipeline.WhisperX.Retries)
}
if cfg.Pipeline.WhisperX.Concurrency == nil || *cfg.Pipeline.WhisperX.Concurrency != 2 {
t.Fatalf("whisperx.concurrency = %v, want 2", cfg.Pipeline.WhisperX.Concurrency)
}
if cfg.Pipeline.Seriatim.Timeout != "10m" {
t.Fatalf("seriatim.timeout = %q, want %q", cfg.Pipeline.Seriatim.Timeout, "10m")
}
if cfg.Pipeline.Seriatim.OutputSchema != "seriatim-intermediate" {
t.Fatalf("seriatim.output_schema = %q, want %q", cfg.Pipeline.Seriatim.OutputSchema, "seriatim-intermediate")
}
if cfg.Pipeline.Seriatim.CoalesceGap == nil || *cfg.Pipeline.Seriatim.CoalesceGap != 3.0 {
t.Fatalf("seriatim.coalesce_gap = %v, want 3.0", cfg.Pipeline.Seriatim.CoalesceGap)
}
if cfg.Pipeline.Seriatim.Report == nil || *cfg.Pipeline.Seriatim.Report != true {
t.Fatalf("seriatim.report = %v, want true", cfg.Pipeline.Seriatim.Report)
}
}
err = Validate(cfg)
if tt.wantValidate != "" {
if err == nil {
t.Fatalf("expected validation error containing %q, got nil", tt.wantValidate)
}
if !strings.Contains(err.Error(), tt.wantValidate) {
t.Fatalf("validation error = %q, want to contain %q", err.Error(), tt.wantValidate)
}
return
}
if err != nil {
t.Fatalf("Validate() error = %v", err)
}
})
}
}
func TestValidateMissingAudioSource(t *testing.T) {
retries := 3
concurrency := 2
cfg := &Config{
Pipeline: &PipelineConfig{
Workspace: WorkspaceConfig{Root: "/tmp/narratio"},
WhisperX: WhisperXConfig{
TranscribeURL: "https://transcription.ai.rakestrawhome.com/transcribe",
Language: "en",
Timeout: "30m",
Retries: &retries,
RetryDelay: "2s",
Concurrency: &concurrency,
},
Seriatim: SeriatimConfig{
Binary: "seriatim",
Timeout: "10m",
OutputSchema: "seriatim-intermediate",
CoalesceGap: float64Ptr(3.0),
Report: boolPtr(true),
},
},
Session: &SessionConfig{
SessionID: "2026-05-03",
Inputs: SessionInputsConfig{
SpeakersFile: "speakers.yml",
AutocorrectFile: "autocorrect.yml",
GlossaryFile: "glossary.yml",
},
},
}
err := Validate(cfg)
if err == nil {
t.Fatal("expected validation error, got nil")
}
if !strings.Contains(err.Error(), "audio_dir or at least one audio_files") {
t.Fatalf("error = %q, want audio source guidance", err.Error())
}
if !strings.Contains(err.Error(), "session config") {
t.Fatalf("error = %q, want session config context", err.Error())
}
}
func TestExamplesLoadAndValidate(t *testing.T) {
pipelinePath := filepath.Join("..", "..", "examples", "pipeline.minimal.yml")
sessionPath := filepath.Join("..", "..", "examples", "session.minimal.yml")
cfg, err := Load(pipelinePath, sessionPath)
if err != nil {
t.Fatalf("Load(examples) error = %v", err)
}
if err := Validate(cfg); err != nil {
t.Fatalf("Validate(examples) error = %v", err)
}
}
func writeConfigFiles(t *testing.T, pipelineYAML, sessionYAML string) (string, string) {
t.Helper()
dir := t.TempDir()
pipelinePath := filepath.Join(dir, "pipeline.yml")
sessionPath := filepath.Join(dir, "session.yml")
if err := os.WriteFile(pipelinePath, []byte(pipelineYAML), 0o644); err != nil {
t.Fatalf("write pipeline.yml: %v", err)
}
if err := os.WriteFile(sessionPath, []byte(sessionYAML), 0o644); err != nil {
t.Fatalf("write session.yml: %v", err)
}
return pipelinePath, sessionPath
}