Music Pipeline Configuration Hardening

README.md

@@ -114,7 +114,7 @@ music:
     max_stem_length: 16
     naming_scheme: nwn_bmu
     output_extension: .bmu
-    convert_extensions: [.mp3, .ogg]
+    convert_extensions: [.flac, .m4a, .mp3, .ogg, .wav]
   datasets:
     westgate:
       source: envi/music/westgate
@@ -259,7 +259,7 @@ music:
     max_stem_length: 16
     naming_scheme: nwn_bmu
     output_extension: .bmu
-    convert_extensions: [.mp3, .ogg]
+    convert_extensions: [.flac, .m4a, .mp3, .ogg, .wav]
 ```
 
 Music datasets define individual music collections with independent sources and
@@ -269,6 +269,12 @@ naming prefixes. `source` is the authored input root under `paths.assets`;
 configured source extensions into generated HAK resources; `package_mode: none`
 scans and writes credits without adding generated HAK resources. The legacy
 `music.prefixes` shorthand is automatically normalized into datasets.
+`convert_extensions` is configuration-driven; the built-in default set is
+`.flac`, `.m4a`, `.mp3`, `.ogg`, and `.wav`.
+Those extensions are discovered from YAML within configured music dataset source
+roots, so setting them in `music.defaults.convert_extensions` or a dataset
+override is sufficient; you do not need a parallel
+`inventory.asset_extensions` override just to make music scanning see them.
 
 Music can also be run directly:
 

internal/music/music.go

@@ -15,9 +15,19 @@ const (
 )
 
 var (
+	defaultConvertExtensions = []string{
+		".flac",
+		".m4a",
+		".mp3",
+		".ogg",
+		".wav",
+	}
 	ConvertExtensions = map[string]struct{}{
-		".mp3": {},
+		".flac": {},
-		".ogg": {},
+		".m4a":  {},
+		".mp3":  {},
+		".ogg":  {},
+		".wav":  {},
 	}
 	PassthroughExtensions = map[string]struct{}{
 		".bmu": {},
@@ -38,6 +48,10 @@ var (
 	VowelRE   = regexp.MustCompile(`[aeiou]`)
 )
 
+func DefaultConvertExtensions() []string {
+	return append([]string(nil), defaultConvertExtensions...)
+}
+
 type Metadata struct {
 	Title     string
 	Artist    string

internal/pipeline/pipeline_test.go

@@ -3447,6 +3447,146 @@ haks:
 	}
 }
 
+func TestBuildHAKsConvertsFLACMusicSourcesByDefault(t *testing.T) {
+	root := t.TempDir()
+	mustMkdir(t, filepath.Join(root, "assets", "audio", "westgate"))
+	mustMkdir(t, filepath.Join(root, "build"))
+
+	ffprobePath := filepath.Join(root, "tools", "ffprobe")
+	ffmpegPath := filepath.Join(root, "tools", "ffmpeg")
+	mustMkdir(t, filepath.Dir(ffprobePath))
+	mustWriteFile(t, ffprobePath, "#!/bin/sh\nprintf '%s\\n' '{\"format\":{\"tags\":{\"artist\":\"FLAC Artist\",\"title\":\"FLAC Title\"}}}'\n")
+	mustWriteFile(t, ffmpegPath, "#!/bin/sh\nfor last; do :; done\nprintf 'dataset-bmu' > \"$last\"\n")
+	if err := os.Chmod(ffprobePath, 0o755); err != nil {
+		t.Fatalf("chmod ffprobe: %v", err)
+	}
+	if err := os.Chmod(ffmpegPath, 0o755); err != nil {
+		t.Fatalf("chmod ffmpeg: %v", err)
+	}
+
+	mustWriteFile(t, filepath.Join(root, "nwn-tool.yaml"), `
+module:
+  name: Test Assets
+  resref: testassets
+paths:
+  assets: assets
+  build: build
+music:
+  tools:
+    ffmpeg: tools/ffmpeg
+    ffprobe: tools/ffprobe
+  datasets:
+    westgate_audio:
+      source: audio/westgate
+      output: generated/music
+      prefix: wg_
+haks:
+  - name: music
+    priority: 1
+    max_bytes: 1048576
+    split: false
+    include:
+      - generated/music/**
+`)
+	mustWriteFile(t, filepath.Join(root, "assets", "audio", "westgate", "Theme Song.flac"), "source-flac")
+
+	p, err := project.Load(root)
+	if err != nil {
+		t.Fatalf("load project: %v", err)
+	}
+	if err := p.ValidateLayout(); err != nil {
+		t.Fatalf("validate layout: %v", err)
+	}
+	if err := p.Scan(); err != nil {
+		t.Fatalf("scan: %v", err)
+	}
+
+	result, err := BuildHAKs(p)
+	if err != nil {
+		t.Fatalf("build haks: %v", err)
+	}
+	if result.CreditsSummary.TrackCount != 1 {
+		t.Fatalf("expected one music track, got %#v", result.CreditsSummary)
+	}
+	manifestRaw, err := os.ReadFile(filepath.Join(root, "build", "haks.json"))
+	if err != nil {
+		t.Fatalf("read manifest: %v", err)
+	}
+	manifestText := string(manifestRaw)
+	if !strings.Contains(manifestText, "generated/music/wg_theme.bmu") {
+		t.Fatalf("expected generated FLAC music in manifest, got:\n%s", manifestText)
+	}
+	if strings.Contains(manifestText, "audio/westgate/Theme Song.flac") {
+		t.Fatalf("did not expect source flac in manifest, got:\n%s", manifestText)
+	}
+}
+
+func TestBuildHAKsUsesYAMLMusicConvertExtensionsForDiscovery(t *testing.T) {
+	root := t.TempDir()
+	mustMkdir(t, filepath.Join(root, "assets", "audio", "westgate"))
+	mustMkdir(t, filepath.Join(root, "build"))
+
+	ffprobePath := filepath.Join(root, "tools", "ffprobe")
+	ffmpegPath := filepath.Join(root, "tools", "ffmpeg")
+	mustMkdir(t, filepath.Dir(ffprobePath))
+	mustWriteFile(t, ffprobePath, "#!/bin/sh\nprintf '%s\\n' '{\"format\":{\"tags\":{\"artist\":\"AAC Artist\",\"title\":\"AAC Title\"}}}'\n")
+	mustWriteFile(t, ffmpegPath, "#!/bin/sh\nfor last; do :; done\nprintf 'dataset-bmu' > \"$last\"\n")
+	if err := os.Chmod(ffprobePath, 0o755); err != nil {
+		t.Fatalf("chmod ffprobe: %v", err)
+	}
+	if err := os.Chmod(ffmpegPath, 0o755); err != nil {
+		t.Fatalf("chmod ffmpeg: %v", err)
+	}
+
+	mustWriteFile(t, filepath.Join(root, "nwn-tool.yaml"), `
+module:
+  name: Test Assets
+  resref: testassets
+paths:
+  assets: assets
+  build: build
+music:
+  defaults:
+    convert_extensions:
+      - .aac
+  tools:
+    ffmpeg: tools/ffmpeg
+    ffprobe: tools/ffprobe
+  datasets:
+    westgate_audio:
+      source: audio/westgate
+      output: generated/music
+      prefix: wg_
+haks:
+  - name: music
+    priority: 1
+    max_bytes: 1048576
+    split: false
+    include:
+      - generated/music/**
+`)
+	mustWriteFile(t, filepath.Join(root, "assets", "audio", "westgate", "Theme Song.aac"), "source-aac")
+
+	p, err := project.Load(root)
+	if err != nil {
+		t.Fatalf("load project: %v", err)
+	}
+	if err := p.ValidateLayout(); err != nil {
+		t.Fatalf("validate layout: %v", err)
+	}
+	if err := p.Scan(); err != nil {
+		t.Fatalf("scan: %v", err)
+	}
+
+	result, err := BuildHAKs(p)
+	if err != nil {
+		t.Fatalf("build haks: %v", err)
+	}
+	if result.CreditsSummary.TrackCount != 1 {
+		t.Fatalf("expected one music track discovered through YAML convert_extensions, got %#v", result.CreditsSummary)
+	}
+}
+
 func TestPlanHAKsDoesNotTranscodeMusicSources(t *testing.T) {
 	root := t.TempDir()
 	mustMkdir(t, filepath.Join(root, "assets", "audio", "westgate"))

internal/project/effective.go

@@ -8,6 +8,8 @@ import (
 	"slices"
 	"strings"
 	"time"
+
+	"gitea.westgate.pw/ShadowsOverWestgate/sow-tools/internal/music"
 )
 
 const (
@@ -249,7 +251,7 @@ func (p *Project) EffectiveConfig() EffectiveConfig {
 	musicMaxStemLength := 16
 	musicNamingScheme := DefaultMusicNamingScheme
 	musicOutputExtension := DefaultMusicOutputExtension
-	musicConvertExtensions := []string{".mp3", ".ogg"}
+	musicConvertExtensions := music.DefaultConvertExtensions()
 	if d := p.Config.Music.Defaults; d != nil {
 		if d.StageRoot != "" {
 			musicStageRoot = expandPathTemplate(d.StageRoot, paths)
@@ -318,7 +320,6 @@ func (p *Project) EffectiveConfig() EffectiveConfig {
 			ConvertExtensions: dsConvertExtensions,
 		}
 	}
-
 	effective := EffectiveConfig{
 		ConfigSource: p.ConfigSource,
 		Module:       p.Config.Module,
@@ -428,7 +429,7 @@ func markMissingDefaults(provenance ConfigProvenance) {
 		"music.max_stem_length":                    "16",
 		"music.naming_scheme":                      DefaultMusicNamingScheme,
 		"music.output_extension":                   DefaultMusicOutputExtension,
-		"music.convert_extensions":                 ".mp3, .ogg",
+		"music.convert_extensions":                 strings.Join(music.DefaultConvertExtensions(), ", "),
 		"topdata.build":                            DefaultTopDataBuild,
 		"topdata.compiled_2da_dir":                 DefaultTopDataCompiled2DADir,
 		"topdata.compiled_tlk":                     DefaultTopDataCompiledTLK,

internal/project/project.go

@@ -602,10 +602,46 @@ func (p *Project) Scan() error {
 	assetDir := p.AssetsDir()
 	var assetFiles []string
 	if assetDir != "" && filepath.Clean(assetDir) != filepath.Clean(p.Root) {
+		musicSourceExts := make(map[string]struct{})
+		for _, ext := range effective.Music.ConvertExtensions {
+			musicSourceExts[ext] = struct{}{}
+		}
+		musicSourceRoots := make(map[string]map[string]struct{})
+		for _, ds := range effective.Music.Datasets {
+			if strings.TrimSpace(ds.Source) == "" {
+				continue
+			}
+			exts := musicSourceRoots[ds.Source]
+			if exts == nil {
+				exts = make(map[string]struct{})
+				musicSourceRoots[ds.Source] = exts
+			}
+			for _, ext := range ds.ConvertExtensions {
+				exts[ext] = struct{}{}
+			}
+			for ext := range musicSourceExts {
+				exts[ext] = struct{}{}
+			}
+		}
 		var err error
 		assetFiles, _, err = scanDir(assetDir, func(path string) bool {
-			ext := strings.ToLower(filepath.Ext(path))
+			rel, err := filepath.Rel(assetDir, path)
-			return slices.Contains(effective.Inventory.AssetExtensions, ext)
+			if err != nil {
+				return false
+			}
+			rel = filepath.ToSlash(rel)
+			ext := strings.ToLower(filepath.Ext(rel))
+			if slices.Contains(effective.Inventory.AssetExtensions, ext) {
+				return true
+			}
+			for root, exts := range musicSourceRoots {
+				if !pathIsUnder(root, rel) {
+					continue
+				}
+				_, ok := exts[ext]
+				return ok
+			}
+			return false
 		})
 		if err != nil {
 			if errors.Is(err, os.ErrNotExist) {
@@ -1393,6 +1429,15 @@ func validateOutputFileName(field, name, expectedExt string) error {
 	return nil
 }
 
+func pathIsUnder(root, rel string) bool {
+	root = strings.Trim(strings.TrimSpace(filepath.ToSlash(root)), "/")
+	rel = strings.Trim(strings.TrimSpace(filepath.ToSlash(rel)), "/")
+	if root == "" || rel == "" {
+		return false
+	}
+	return rel == root || strings.HasPrefix(rel, root+"/")
+}
+
 func validateRelativePath(field, path string) []error {
 	var failures []error
 	trimmed := strings.TrimSpace(path)

internal/project/project_test.go

@@ -81,6 +81,9 @@ paths:
 	if got := strings.Join(effective.Validation.RequiredFields["ifo"], ","); got != "Mod_Name" {
 		t.Fatalf("expected default IFO required fields, got %#v", effective.Validation.RequiredFields)
 	}
+	if got, want := strings.Join(effective.Music.ConvertExtensions, ","), ".flac,.m4a,.mp3,.ogg,.wav"; got != want {
+		t.Fatalf("expected default music convert extensions %q, got %q", want, got)
+	}
 	if prov := effective.Provenance["paths.build"]; prov.Source != "toolkit default" {
 		t.Fatalf("expected paths.build toolkit default provenance, got %#v", prov)
 	}
@@ -120,6 +123,53 @@ validation:
 	}
 }
 
+func TestScanUsesYAMLMusicConvertExtensionsWithinConfiguredDatasetRoots(t *testing.T) {
+	root := t.TempDir()
+	if err := os.MkdirAll(filepath.Join(root, "assets", "audio", "westgate"), 0o755); err != nil {
+		t.Fatalf("mkdir westgate audio: %v", err)
+	}
+	if err := os.MkdirAll(filepath.Join(root, "assets", "audio", "other"), 0o755); err != nil {
+		t.Fatalf("mkdir other audio: %v", err)
+	}
+	writeProjectFile(t, filepath.Join(root, ConfigFile), `
+module:
+  name: Test Module
+  resref: testmod
+paths:
+  source: src
+  assets: assets
+music:
+  defaults:
+    convert_extensions:
+      - .flac
+  datasets:
+    westgate_audio:
+      source: audio/westgate
+      convert_extensions:
+        - .aac
+`)
+	writeProjectFile(t, filepath.Join(root, "assets", "audio", "westgate", "theme.aac"), "aac")
+	writeProjectFile(t, filepath.Join(root, "assets", "audio", "westgate", "theme.flac"), "flac")
+	writeProjectFile(t, filepath.Join(root, "assets", "audio", "other", "ignored.aac"), "aac")
+
+	proj, err := Load(root)
+	if err != nil {
+		t.Fatalf("Load returned error: %v", err)
+	}
+	if err := proj.Scan(); err != nil {
+		t.Fatalf("Scan returned error: %v", err)
+	}
+	if !slices.Contains(proj.Inventory.AssetFiles, "audio/westgate/theme.aac") {
+		t.Fatalf("expected configured dataset AAC file to be discovered, got %#v", proj.Inventory.AssetFiles)
+	}
+	if !slices.Contains(proj.Inventory.AssetFiles, "audio/westgate/theme.flac") {
+		t.Fatalf("expected configured dataset FLAC file to be discovered, got %#v", proj.Inventory.AssetFiles)
+	}
+	if slices.Contains(proj.Inventory.AssetFiles, "audio/other/ignored.aac") {
+		t.Fatalf("did not expect non-dataset AAC file to be discovered, got %#v", proj.Inventory.AssetFiles)
+	}
+}
+
 func TestEffectiveConfigHonorsConfiguredOutputAndCacheFields(t *testing.T) {
 	root := t.TempDir()
 	writeProjectFile(t, filepath.Join(root, ConfigFile), `