configurable scan parallelism based on hdd or sdd

This commit is contained in:
2026-02-19 13:59:02 -05:00
parent 81793975b4
commit d9ebd38382
19 changed files with 1579 additions and 142 deletions
+257 -63
View File
@@ -14,6 +14,7 @@ import (
"strings"
"sync"
"sync/atomic"
"time"
"github.com/wailsapp/wails/v2/pkg/runtime"
"golang.org/x/sync/errgroup"
@@ -22,6 +23,7 @@ import (
"yellowjacket/backend/database/sql/sqlcgen"
"yellowjacket/backend/events"
"yellowjacket/backend/metadata"
"yellowjacket/backend/system"
)
var errLibraryDirNotConfigured = errors.New("library directory not configured")
@@ -148,24 +150,38 @@ func (l *Library) registerEventHandlers() {
}
// Scan syncs the library by adding new files and removing deleted ones.
// Files that exist but have incomplete metadata (recording_id = 0) will be updated.
func (l *Library) Scan() error {
// Files that exist but have incomplete metadata (recording_id = 0)
// will be updated. The returned ScanMetrics contains timing and
// count data for every phase of the scan.
func (l *Library) Scan() (*ScanMetrics, error) {
metrics := newScanMetrics()
scanStart := time.Now()
if len(l.conf.DirectoryPath) == 0 {
return metrics, errLibraryDirNotConfigured
}
workerCount := resolveScanWorkerCount(
l.conf.ScanConcurrency,
string(l.conf.DirectoryPath),
)
l.logger.Info(
"beginning library scan", "workers", scanWorkerCount,
"beginning library scan",
"workers", workerCount,
"concurrencyMode", l.conf.ScanConcurrency,
)
runtime.EventsEmit(l.ctx, events.LibraryScanStarted)
if len(l.conf.DirectoryPath) == 0 {
return errLibraryDirNotConfigured
}
// --- Phase 1: load existing files from DB ---
loadStart := time.Now()
// Load existing file paths from the database into a sync.Map for concurrent access.
// The map tracks path → audioFile; entries are removed as files are "seen" during the walk.
// Any entries remaining after the walk are orphans (files deleted from disk).
existingFiles, err := l.db.Queries.GetAllAudioFiles(l.ctx)
if err != nil {
return fmt.Errorf("could not load existing audio files: %w", err)
return metrics, fmt.Errorf(
"could not load existing audio files: %w", err,
)
}
existingPaths := &sync.Map{}
@@ -173,6 +189,8 @@ func (l *Library) Scan() error {
existingPaths.Store(f.FilePath, f)
}
metrics.LoadExisting = time.Since(loadStart)
l.logger.Debug(
"loaded existing files from database",
"count", len(existingFiles),
@@ -189,16 +207,25 @@ func (l *Library) Scan() error {
var errMu sync.Mutex
// Walker goroutine: traverse directory and send work items to workers
// --- Phase 2: directory walk ---
walkStart := time.Now()
go func() {
defer close(workChan)
defer func() {
metrics.WalkDuration = time.Since(walkStart)
close(workChan)
}()
walkErr := fs.WalkDir(
os.DirFS(basePath),
".",
func(path string, d fs.DirEntry, err error) error {
if err != nil {
l.logger.Error("problem walking directory", "path", path, "err", err)
l.logger.Error(
"problem walking directory",
"path", path, "err", err,
)
return nil // continue walking
}
@@ -207,7 +234,9 @@ func (l *Library) Scan() error {
return nil
}
absoluteFilePath := filepath.Join(basePath, path)
absoluteFilePath := filepath.Join(
basePath, path,
)
fileExt := filepath.Ext(d.Name())
fileType, isSupportedAudioFile := metadata.GetSupportedFileType(fileExt)
@@ -215,13 +244,15 @@ func (l *Library) Scan() error {
return nil
}
// Check if file already exists in database
// Check if file already exists in database.
if existing, exists := existingPaths.LoadAndDelete(absoluteFilePath); exists {
audioFile := existing.(sqlcgen.AudioFile)
// Check if this file needs metadata update (recording_id = 0)
if audioFile.RecordingID == 0 {
l.logger.Debug("file needs metadata update", "path", absoluteFilePath)
l.logger.Debug(
"file needs metadata update",
"path", absoluteFilePath,
)
select {
case workChan <- scanWork{
@@ -248,11 +279,16 @@ func (l *Library) Scan() error {
return nil
}
l.logger.Debug("queueing file for import", "path", absoluteFilePath)
l.logger.Debug(
"queueing file for import",
"path", absoluteFilePath,
)
// Send to workers for processing
select {
case workChan <- scanWork{absolutePath: absoluteFilePath, fileType: fileType}:
case workChan <- scanWork{
absolutePath: absoluteFilePath,
fileType: fileType,
}:
case <-l.ctx.Done():
return l.ctx.Err()
}
@@ -265,15 +301,44 @@ func (l *Library) Scan() error {
errMu.Lock()
scanErr = errors.Join(
scanErr,
fmt.Errorf("problem walking library directory: %w", walkErr),
fmt.Errorf(
"problem walking library directory: %w",
walkErr,
),
)
errMu.Unlock()
}
}()
// DB writer goroutine: serialize all database writes to avoid SQLite
// contention. Results are committed in batches to amortize the cost
// of SQLite's fsync-per-commit.
// --- Thumbnail worker pool (async, decoupled from DB writer) ---
thumbChan := make(chan thumbnailWork, 100)
var thumbWg sync.WaitGroup
for range workerCount {
thumbWg.Add(1)
go func() {
defer thumbWg.Done()
for work := range thumbChan {
if err := l.generateSizedVariantsWithMetrics(
work.imgData,
work.dir,
work.hashStr,
work.metrics,
); err != nil {
l.logger.Warn(
"could not generate thumbnails",
"hash", work.hashStr,
"err", err,
)
}
}
}()
}
// --- Phase 4: DB writer goroutine ---
var dbWg sync.WaitGroup
dbWg.Add(1)
@@ -283,53 +348,79 @@ func (l *Library) Scan() error {
cache := newEntityCache()
var batch []importResult
var (
batch []importResult
dbStarted bool
dbStartVal time.Time
)
flushBatch := func() {
if len(batch) == 0 {
return
}
batchStart := time.Now()
if batchErr := l.commitBatch(
batch, cache, &added, &updated,
batch, cache, metrics,
&added, &updated,
thumbChan,
); batchErr != nil {
errMu.Lock()
scanErr = errors.Join(scanErr, batchErr)
errMu.Unlock()
}
metrics.BatchCommits += time.Since(batchStart)
batch = batch[:0]
}
for result := range resultChan {
if !dbStarted {
dbStartVal = time.Now()
dbStarted = true
}
batch = append(batch, result)
if len(batch) >= scanBatchSize {
flushBatch()
}
}
// Flush any remaining results.
flushBatch()
if dbStarted {
metrics.DBWritesWallClock = time.Since(
dbStartVal,
)
}
}()
// Worker pool: extract metadata concurrently, send results to DB writer
// --- Phase 3: worker pool ---
extractStart := time.Now()
g := new(errgroup.Group)
g.SetLimit(scanWorkerCount)
g.SetLimit(workerCount)
for work := range workChan {
g.Go(func() error {
result, err := l.extractAudioMetadata(work)
result, err := l.extractAudioMetadata(
work, metrics,
)
if err != nil {
l.logger.Warn("failed to extract metadata", "path", work.absolutePath, "err", err)
l.logger.Warn(
"failed to extract metadata",
"path", work.absolutePath,
"err", err,
)
errMu.Lock()
scanErr = errors.Join(scanErr, err)
errMu.Unlock()
return nil // continue processing other files
return nil
}
// Send to DB writer
select {
case resultChan <- result:
case <-l.ctx.Done():
@@ -340,21 +431,40 @@ func (l *Library) Scan() error {
})
}
_ = g.Wait() // Wait for all metadata extraction to complete
_ = g.Wait()
close(resultChan) // Signal DB writer to finish
dbWg.Wait() // Wait for all DB writes to complete
metrics.ExtractionWallClock = time.Since(extractStart)
close(resultChan)
dbWg.Wait()
// Close thumbnail channel and wait for all thumbnail workers
// to finish. The DB writer has stopped sending work at this
// point so it is safe to close.
thumbStart := time.Now()
close(thumbChan)
thumbWg.Wait()
metrics.ThumbnailWallClock = time.Since(thumbStart)
// --- Phase 5: orphan cleanup ---
orphanStart := time.Now()
// Orphan cleanup: any entries remaining in existingPaths are files deleted from disk
var removed atomic.Int64
existingPaths.Range(func(key, value any) bool {
path := key.(string)
audioFile := value.(sqlcgen.AudioFile)
l.logger.Debug("removing orphaned database entry", "path", path, "id", audioFile.ID)
l.logger.Debug(
"removing orphaned database entry",
"path", path, "id", audioFile.ID,
)
if err := l.db.Queries.DeleteAudioFile(l.ctx, audioFile.ID); err != nil {
if err := l.db.Queries.DeleteAudioFile(
l.ctx, audioFile.ID,
); err != nil {
l.logger.Warn(
"failed to delete orphaned audio file",
"path", path,
@@ -370,8 +480,11 @@ func (l *Library) Scan() error {
return true
})
// Generate sized variants for any cover art missing them,
// and migrate legacy _thumb files.
metrics.OrphanCleanup = time.Since(orphanStart)
// --- Phase 6: post-scan variant generation ---
variantStart := time.Now()
if err := l.generateMissingSizedVariants(); err != nil {
l.logger.Warn(
"could not generate missing sized variants",
@@ -379,23 +492,58 @@ func (l *Library) Scan() error {
)
}
metrics.PostScanVariants = time.Since(variantStart)
// --- Finalize ---
metrics.Added = added.Load()
metrics.Updated = updated.Load()
metrics.Skipped = skipped.Load()
metrics.Removed = removed.Load()
metrics.Total = time.Since(scanStart)
l.logger.Info(
"library scan complete",
"added", added.Load(),
"updated", updated.Load(),
"removed", removed.Load(),
"skipped", skipped.Load(),
"added", metrics.Added,
"updated", metrics.Updated,
"removed", metrics.Removed,
"skipped", metrics.Skipped,
"total", metrics.Total,
"library", l.conf.DirectoryPath,
)
runtime.EventsEmit(l.ctx, events.LibraryScanComplete)
runtime.EventsEmit(
l.ctx, events.LibraryScanComplete, metrics,
)
return scanErr
return metrics, scanErr
}
// scanWorkerCount controls the number of concurrent file processors.
// TODO: make configurable via Config.
var scanWorkerCount = goruntime.NumCPU()
// hddWorkerCount is the maximum number of concurrent extraction
// workers when the library resides on a spinning disk.
const hddWorkerCount = 2
// resolveScanWorkerCount returns the number of concurrent
// extraction workers based on the configured concurrency mode
// and the storage type of the library directory.
func resolveScanWorkerCount(
mode ScanConcurrency,
libraryPath string,
) int {
switch mode {
case ScanConcurrencySSD:
return goruntime.NumCPU()
case ScanConcurrencyHDD:
return min(hddWorkerCount, goruntime.NumCPU())
default: // auto
if system.IsRotationalDisk(libraryPath) {
return min(
hddWorkerCount, goruntime.NumCPU(),
)
}
return goruntime.NumCPU()
}
}
// scanWork represents a file to be processed by a worker.
type scanWork struct {
@@ -417,8 +565,12 @@ type importResult struct {
}
// extractAudioMetadata reads and extracts metadata from an audio file.
// It opens the file once, extracting both tags and duration in a single pass.
func (l *Library) extractAudioMetadata(work scanWork) (importResult, error) {
// It opens the file once, extracting both tags and duration in a
// single pass, and records per-file timing in the shared metrics.
func (l *Library) extractAudioMetadata(
work scanWork,
metrics *ScanMetrics,
) (importResult, error) {
result := importResult{
absolutePath: work.absolutePath,
fileType: work.fileType,
@@ -429,9 +581,18 @@ func (l *Library) extractAudioMetadata(work scanWork) (importResult, error) {
// Skip duration decode if we already have it from a previous import.
skipDuration := work.needsUpdate && work.existingLength > 0
tags, lengthMillis, err := metadata.ExtractAllMetadata(
tags, lengthMillis, timing, err := metadata.ExtractAllMetadata(
work.absolutePath, skipDuration,
)
if timing != nil {
metrics.addExtraction(
string(work.fileType),
timing.TagExtraction,
timing.DurationExtraction,
)
}
if err != nil {
return result, fmt.Errorf(
"could not extract metadata for %s: %w",
@@ -454,11 +615,14 @@ func (l *Library) extractAudioMetadata(work scanWork) (importResult, error) {
// commitBatch wraps a slice of import results in a single database
// transaction, creating all related records and audio file entries.
// Individual file failures are logged and accumulated but do not
// abort the entire batch.
// abort the entire batch. thumbChan dispatches thumbnail generation
// to the async worker pool.
func (l *Library) commitBatch(
batch []importResult,
cache *entityCache,
metrics *ScanMetrics,
added, updated *atomic.Int64,
thumbChan chan<- thumbnailWork,
) error {
tx, err := l.db.BeginTx()
if err != nil {
@@ -475,12 +639,18 @@ func (l *Library) commitBatch(
var saveErr error
if result.needsUpdate {
saveErr = l.updateAudioFileMetadata(txq, cache, *result)
saveErr = l.updateAudioFileMetadata(
txq, cache, metrics, *result,
thumbChan,
)
if saveErr == nil {
updated.Add(1)
}
} else {
saveErr = l.saveAudioFile(txq, cache, *result)
saveErr = l.saveAudioFile(
txq, cache, metrics, *result,
thumbChan,
)
if saveErr == nil {
added.Add(1)
}
@@ -511,7 +681,9 @@ func (l *Library) commitBatch(
func (l *Library) saveAudioFile(
q *sqlcgen.Queries,
cache *entityCache,
metrics *ScanMetrics,
result importResult,
thumbChan chan<- thumbnailWork,
) error {
l.logger.Debug(
"saving audio file to db",
@@ -526,7 +698,9 @@ func (l *Library) saveAudioFile(
)
// Process metadata and create related records.
recordingID, err := l.processMetadata(q, cache, result)
recordingID, err := l.processMetadata(
q, cache, metrics, result, thumbChan,
)
if err != nil {
return fmt.Errorf("could not process metadata: %w", err)
}
@@ -560,7 +734,9 @@ func (l *Library) saveAudioFile(
func (l *Library) updateAudioFileMetadata(
q *sqlcgen.Queries,
cache *entityCache,
metrics *ScanMetrics,
result importResult,
thumbChan chan<- thumbnailWork,
) error {
l.logger.Debug(
"updating audio file metadata",
@@ -569,7 +745,9 @@ func (l *Library) updateAudioFileMetadata(
)
// Process metadata and create related records.
recordingID, err := l.processMetadata(q, cache, result)
recordingID, err := l.processMetadata(
q, cache, metrics, result, thumbChan,
)
if err != nil {
return fmt.Errorf("could not process metadata: %w", err)
}
@@ -596,10 +774,14 @@ func (l *Library) updateAudioFileMetadata(
// and returns the recording ID. It uses the provided queries object
// (which may be transaction-scoped) and the entity cache to avoid
// redundant upserts for repeated artist/album/cover-art values.
// When thumbChan is non-nil, thumbnail generation is dispatched
// asynchronously.
func (l *Library) processMetadata(
q *sqlcgen.Queries,
cache *entityCache,
metrics *ScanMetrics,
result importResult,
thumbChan chan<- thumbnailWork,
) (int64, error) {
tags := result.tags
if tags == nil {
@@ -607,7 +789,9 @@ func (l *Library) processMetadata(
}
// 1. Handle cover art (if present).
coverArtID := l.processCoverArt(q, cache, tags)
coverArtID := l.processCoverArt(
q, cache, metrics, tags, thumbChan,
)
// 2. Get or create artist credit for track artist.
artistName := tags.Artist
@@ -681,17 +865,23 @@ func (l *Library) processMetadata(
}
// processCoverArt saves cover art to disk and upserts the DB record,
// using the cache to skip work for previously seen images.
// using the cache to skip work for previously seen images. When
// thumbChan is non-nil, thumbnail generation is dispatched to the
// async worker pool.
func (l *Library) processCoverArt(
q *sqlcgen.Queries,
cache *entityCache,
metrics *ScanMetrics,
tags *metadata.TrackMetadata,
thumbChan chan<- thumbnailWork,
) sql.NullInt64 {
if tags.Picture == nil {
return sql.NullInt64{}
}
coverPath, err := l.saveCoverArt(tags.Picture)
coverPath, err := l.saveCoverArt(
tags.Picture, metrics, thumbChan,
)
if err != nil {
l.logger.Warn("could not save cover art", "err", err)
@@ -937,10 +1127,14 @@ func (l *Library) handleConfigUpdate(updatedConfigValues Config) error {
l.logger.Info("new library, scanning")
l.conf.DirectoryPath = updatedConfigValues.DirectoryPath
if err := l.Scan(); err != nil {
if _, err := l.Scan(); err != nil {
updateErr = errors.Join(
updateErr,
fmt.Errorf("problem scanning library on config update: %w", err),
fmt.Errorf(
"problem scanning library on config update: %w",
err,
),
)
}
}