mirror/navidrome
1
0
Fork 0
mirror of https://github.com/navidrome/navidrome.git synced 2025-12-23 23:18:05 -05:00
Code Issues Packages Projects Releases Wiki Activity

fix(scanner): optimize refresh (#4059)

Browse Source 
* fix(artist): update RefreshStats to only process artists with recently updated media files

Signed-off-by: Deluan <deluan@navidrome.org>

* fix: paginate Artist's RefreshStats, also replace rawSQL with Expr

Signed-off-by: Deluan <deluan@navidrome.org>

---------

Signed-off-by: Deluan <deluan@navidrome.org>
This commit is contained in:
Deluan Quintão
2025-05-14 20:47:03 -04:00
committed by GitHub
parent ec9f9aa243
commit db92cf9e47
6 changed files with 118 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
Makefile
View File

@@ -36,8 +36,9 @@ watch: ##@Development Start Go tests in watch mode (re-run when code changes)
go tool ginkgo watch -tags=netgo -notify ./...
.PHONY: watch
PKG ?= ./...
test: ##@Development Run Go tests
go test -tags netgo ./...
go test -tags netgo $(PKG)
.PHONY: test
testrace: ##@Development Run Go tests with race detector

2
persistence/album_repository.go
View File

@@ -315,7 +315,7 @@ func (r *albumRepository) GetTouchedAlbums(libID int) (model.AlbumCursor, error)
// RefreshPlayCounts updates the play count and last play date annotations for all albums, based
// on the media files associated with them.
func (r *albumRepository) RefreshPlayCounts() (int64, error) {
query := rawSQL(`
query := Expr(`
with play_counts as (
select user_id, album_id, sum(play_count) as total_play_count, max(play_date) as last_play_date
from media_file

179
persistence/artist_repository.go
View File

@@ -239,7 +239,7 @@ func (r *artistRepository) purgeEmpty() error {
// RefreshPlayCounts updates the play count and last play date annotations for all artists, based
// on the media files associated with them.
func (r *artistRepository) RefreshPlayCounts() (int64, error) {
query := rawSQL(`
query := Expr(`
with play_counts as (
select user_id, atom as artist_id, sum(play_count) as total_play_count, max(play_date) as last_play_date
from media_file
@@ -259,76 +259,123 @@ on conflict (user_id, item_id, item_type) do update
return r.executeSQL(query)
}
// RefreshStats updates the stats field for all artists, based on the media files associated with them.
// BFR Maybe filter by "touched" artists?
// RefreshStats updates the stats field for artists whose associated media files were updated after the oldest recorded library scan time.
// It processes artists in batches to handle potentially large updates.
func (r *artistRepository) RefreshStats() (int64, error) {
// First get all counters, one query groups by artist/role, and another with totals per artist.
// Union both queries and group by artist to get a single row of counters per artist/role.
// Then format the counters in a JSON object, one key for each role.
// Finally update the artist table with the new counters
// In all queries, atom is the artist ID and path is the role (or "total" for the totals)
query := rawSQL(`
-- CTE to get counters for each artist, grouped by role
with artist_role_counters as (
-- Get counters for each artist, grouped by role
-- (remove the index from the role: composer[0] => composer
select atom as artist_id,
substr(
replace(jt.path, '$.', ''),
1,
case when instr(replace(jt.path, '$.', ''), '[') > 0
then instr(replace(jt.path, '$.', ''), '[') - 1
else length(replace(jt.path, '$.', ''))
end
) as role,
count(distinct album_id) as album_count,
count(mf.id) as count,
sum(size) as size
from media_file mf
left join json_tree(participants) jt
where atom is not null and key = 'id'
group by atom, role
),
touchedArtistsQuerySQL := `
SELECT DISTINCT mfa.artist_id
FROM media_file_artists mfa
JOIN media_file mf ON mfa.media_file_id = mf.id
WHERE mf.updated_at > (SELECT last_scan_at FROM library ORDER BY last_scan_at ASC LIMIT 1)
`
-- CTE to get the totals for each artist
artist_total_counters as (
select mfa.artist_id,
'total' as role,
count(distinct mf.album_id) as album_count,
count(distinct mf.id) as count,
sum(mf.size) as size
from (select artist_id, media_file_id
from main.media_file_artists) as mfa
join main.media_file mf on mfa.media_file_id = mf.id
group by mfa.artist_id
),
var allTouchedArtistIDs []string
if err := r.db.NewQuery(touchedArtistsQuerySQL).Column(&allTouchedArtistIDs); err != nil {
return 0, fmt.Errorf("fetching touched artist IDs: %w", err)
}
-- CTE to combine role and total counters
combined_counters as (
select artist_id, role, album_count, count, size
from artist_role_counters
union
select artist_id, role, album_count, count, size
from artist_total_counters
),
if len(allTouchedArtistIDs) == 0 {
log.Debug(r.ctx, "RefreshStats: No artists to update.")
return 0, nil
}
log.Debug(r.ctx, "RefreshStats: Found artists to update.", "count", len(allTouchedArtistIDs))
-- CTE to format the counters in a JSON object
artist_counters as (
select artist_id as id,
json_group_object(
replace(role, '"', ''),
json_object('a', album_count, 'm', count, 's', size)
) as counters
from combined_counters
group by artist_id
)
// Template for the batch update with placeholder markers that we'll replace
batchUpdateStatsSQL := `
WITH artist_role_counters AS (
SELECT jt.atom AS artist_id,
substr(
replace(jt.path, '$.', ''),
1,
CASE WHEN instr(replace(jt.path, '$.', ''), '[') > 0
THEN instr(replace(jt.path, '$.', ''), '[') - 1
ELSE length(replace(jt.path, '$.', ''))
END
) AS role,
count(DISTINCT mf.album_id) AS album_count,
count(mf.id) AS count,
sum(mf.size) AS size
FROM media_file mf
JOIN json_tree(mf.participants) jt ON jt.key = 'id' AND jt.atom IS NOT NULL
WHERE jt.atom IN (ROLE_IDS_PLACEHOLDER) -- Will replace with actual placeholders
GROUP BY jt.atom, role
),
artist_total_counters AS (
SELECT mfa.artist_id,
'total' AS role,
count(DISTINCT mf.album_id) AS album_count,
count(DISTINCT mf.id) AS count,
sum(mf.size) AS size
FROM media_file_artists mfa
JOIN media_file mf ON mfa.media_file_id = mf.id
WHERE mfa.artist_id IN (TOTAL_IDS_PLACEHOLDER) -- Will replace with actual placeholders
GROUP BY mfa.artist_id
),
combined_counters AS (
SELECT artist_id, role, album_count, count, size FROM artist_role_counters
UNION
SELECT artist_id, role, album_count, count, size FROM artist_total_counters
),
artist_counters AS (
SELECT artist_id AS id,
json_group_object(
replace(role, '"', ''),
json_object('a', album_count, 'm', count, 's', size)
) AS counters
FROM combined_counters
GROUP BY artist_id
)
UPDATE artist
SET stats = coalesce((SELECT counters FROM artist_counters ac WHERE ac.id = artist.id), '{}'),
updated_at = datetime(current_timestamp, 'localtime')
WHERE artist.id IN (UPDATE_IDS_PLACEHOLDER) AND artist.id <> '';` // Will replace with actual placeholders
-- Update the artist table with the new counters
update artist
set stats = coalesce((select counters from artist_counters where artist_counters.id = artist.id), '{}'),
updated_at = datetime(current_timestamp, 'localtime')
where id <> ''; -- always true, to avoid warnings`)
return r.executeSQL(query)
var totalRowsAffected int64 = 0
const batchSize = 1000
batchCounter := 0
for artistIDBatch := range slice.CollectChunks(slices.Values(allTouchedArtistIDs), batchSize) {
batchCounter++
log.Trace(r.ctx, "RefreshStats: Processing batch", "batchNum", batchCounter, "batchSize", len(artistIDBatch))
// Create placeholders for each ID in the IN clauses
placeholders := make([]string, len(artistIDBatch))
for i := range artistIDBatch {
placeholders[i] = "?"
}
// Don't add extra parentheses, the IN clause already expects them in SQL syntax
inClause := strings.Join(placeholders, ",")
// Replace the placeholder markers with actual SQL placeholders
batchSQL := strings.Replace(batchUpdateStatsSQL, "ROLE_IDS_PLACEHOLDER", inClause, 1)
batchSQL = strings.Replace(batchSQL, "TOTAL_IDS_PLACEHOLDER", inClause, 1)
batchSQL = strings.Replace(batchSQL, "UPDATE_IDS_PLACEHOLDER", inClause, 1)
// Create a single parameter array with all IDs (repeated 3 times for each IN clause)
// We need to repeat each ID 3 times (once for each IN clause)
var args []interface{}
for _, id := range artistIDBatch {
args = append(args, id) // For ROLE_IDS_PLACEHOLDER
}
for _, id := range artistIDBatch {
args = append(args, id) // For TOTAL_IDS_PLACEHOLDER
}
for _, id := range artistIDBatch {
args = append(args, id) // For UPDATE_IDS_PLACEHOLDER
}
// Now use Expr with the expanded SQL and all parameters
sqlizer := Expr(batchSQL, args...)
rowsAffected, err := r.executeSQL(sqlizer)
if err != nil {
return totalRowsAffected, fmt.Errorf("executing batch update for artist stats (batch %d): %w", batchCounter, err)
}
totalRowsAffected += rowsAffected
}
log.Debug(r.ctx, "RefreshStats: Successfully updated stats.", "totalArtistsProcessed", len(allTouchedArtistIDs), "totalDBRowsAffected", totalRowsAffected)
return totalRowsAffected, nil
}
func (r *artistRepository) Search(q string, offset int, size int, includeMissing bool) (model.Artists, error) {

8
persistence/helpers.go
View File

@@ -57,14 +57,6 @@ func toCamelCase(str string) string {
})
}
// rawSQL is a string that will be used as is in the SQL query executor
// It does not support arguments
type rawSQL string
func (r rawSQL) ToSql() (string, []interface{}, error) {
return string(r), nil, nil
}
func Exists(subTable string, cond squirrel.Sqlizer) existsCond {
return existsCond{subTable: subTable, cond: cond, not: false}
}

2
persistence/library_repository.go
View File

@@ -136,7 +136,7 @@ func (r *libraryRepository) ScanEnd(id int) error {
return err
}
// https://www.sqlite.org/pragma.html#pragma_optimize
_, err = r.executeSQL(rawSQL("PRAGMA optimize=0x10012;"))
_, err = r.executeSQL(Expr("PRAGMA optimize=0x10012;"))
return err
}

2
persistence/tag_repository.go
View File

@@ -60,7 +60,7 @@ where tag.id = updated_values.id;
`
for _, table := range []string{"album", "media_file"} {
start := time.Now()
query := rawSQL(fmt.Sprintf(template, table))
query := Expr(fmt.Sprintf(template, table))
c, err := r.executeSQL(query)
log.Debug(r.ctx, "Updated tag counts", "table", table, "elapsed", time.Since(start), "updated", c)
if err != nil {