navidrome/persistence/sql_annotations.go
ts c6732e1fdf
feat(cli): add missing file list and remap subcommands (#5928)
* feat(cli): add missing file list and remap subcommands

Signed-off-by: zerovox <933064+zerovox@users.noreply.github.com>

* fix: prevent remapping from dropping participants on target track

* fix: after remapping, refresh stats synchronously

* fix: only move album annotations if moving a track would empty the old album

* fix(persistence): keep the new item's annotation when reassigning onto an item the user already annotated

ReassignAnnotation was a plain UPDATE; the annotation table is unique on
(user_id, item_id, item_type), so when a user had annotated both items the
statement aborted and none of the rows moved. In the scanner that surfaced as
a warning; in the missing-file remap it rolled back the whole operation.
UPDATE OR IGNORE moves what it can and leaves the conflicting rows for GC.

* fix(core): keep the target track's history when remapping a missing file onto it

The remap discards the target's row, and GC then dropped its play counts,
stars, ratings, bookmarks and every playlist entry pointing at it. That is
harmless in the scanner, whose target was imported seconds earlier, but the
CLI lets the user pick any existing track. Move those references onto the
surviving id first; where a user already has a row for both, theirs on the
missing file wins.

* fix(persistence): stop FindByPaths dropping plain paths that contain a colon

Any colon was taken as the libraryID separator, and a non-numeric prefix
made the whole path vanish from the lookup. 'missing fix' then rejected the
very paths 'missing list' printed, and M3U imports silently skipped such
tracks. Only a numeric prefix qualifies a path now.

* perf(cli): stream 'missing list' instead of loading every missing file into memory

GetAll materialised the whole result set before a single row was written;
on a library with 97k missing files that peaked at 1.28 GB of RSS. Iterate
the repository cursor and write rows as they arrive.

* refactor(core): tidy the missing-file remap

Drop the log lines copied from deleteMissing that still said 'after deleting
missing files', the debug-on-success branches, and the what-comments; build
the affected album list without slice helpers.

* fix(cli): move path to the last column of 'missing list'

Path is the only variable-width field, so leading with it misaligns every
row that follows. Applies to both csv and json.

* fix(persistence): also try a numeric colon prefix as a plain path

'1999: A Different Life/01.mp3' parsed as library 1999 plus a truncated path
and matched nothing. The prefix is ambiguous, so search both ways.

Also buffer the json branch of 'missing list', which wrote a syscall per row.

* fix(persistence): move scrobbles and buffered scrobbles off a discarded media file

Both tables carry ON DELETE CASCADE on media_file_id, so 'missing fix'
deleting the target erased its play history and dropped scrobbles still
waiting on an external service. scrobble_buffer needs OR IGNORE for its
unique (user_id, service, media_file_id, play_time).

* fix(persistence): recompute the cached average rating after merging annotations

Merging the discarded row's annotations grows the rating population of the
surviving track, so media_file.average_rating no longer matched what the
annotation rows say. Only reachable since the remap started merging those
rows instead of deleting them.

* fix(persistence): recompute the cached average rating inside ReassignAnnotation

Moving annotation rows always changes the new item's rating population, so
the recompute belongs with the move rather than at each call site. Covers
the album reassign in the remap and the two scanner sites, and replaces the
explicit call ReassignReferences was making.

Album was the worse case: rate an album, move its files, and 'missing fix'
handed the rating to an album still caching an average of 0.

* fix(cli): let libraryID:path win over a file literally named like one

FindByPaths searches a numeric-prefixed reference both ways, so a top-level
file named '1:foo.mp3' can tie with library 1's 'foo.mp3'. The CLI then
rejected the reference as ambiguous while advising the exact syntax the
caller had used. Also disambiguates the same path in two libraries, which
is what the qualified form is for.

---------

Signed-off-by: zerovox <933064+zerovox@users.noreply.github.com>
Co-authored-by: Deluan Quintão <deluan@navidrome.org>
2026-09-12 12:08:25 -04:00

208 lines
6.4 KiB
Go

package persistence
import (
"database/sql"
"errors"
"fmt"
"regexp"
"sort"
"strings"
"sync"
"time"
. "github.com/Masterminds/squirrel"
"github.com/fatih/structs"
"github.com/navidrome/navidrome/conf"
"github.com/navidrome/navidrome/consts"
"github.com/navidrome/navidrome/log"
"github.com/navidrome/navidrome/model"
)
const annotationTable = "annotation"
// annotationColumns are the columns withAnnotation's LEFT JOIN contributes, derived from
// model.Annotations so the set tracks schema changes. average_rating is excluded: it lives on the
// base table, not the annotation join.
var annotationColumns = sync.OnceValue(func() map[string]struct{} {
cols := map[string]struct{}{}
for name := range structs.Map(model.Annotations{}) {
if name == "average_rating" {
continue
}
cols[name] = struct{}{}
}
return cols
})
// annotationColumnRE matches any annotation column as a whole word. The word boundaries keep the
// base-table column average_rating from matching the annotation column rating (Go's \b treats '_'
// as a word char). It is case-insensitive because SQLite column names are, so a raw filter using
// e.g. "RATING" must still be detected.
var annotationColumnRE = sync.OnceValue(func() *regexp.Regexp {
cols := make([]string, 0, len(annotationColumns()))
for col := range annotationColumns() {
cols = append(cols, regexp.QuoteMeta(col))
}
sort.Strings(cols) // map iteration is random; sort for a stable pattern
return regexp.MustCompile(`(?i)\b(?:` + strings.Join(cols, "|") + `)\b`)
})
// filtersNeedAnnotation reports whether the rendered query references an annotation column, i.e.
// whether the annotation LEFT JOIN must be kept. Scanning the rendered SQL catches every filter
// path. The placeholder column is needed because squirrel won't render a column-less SELECT; on a
// render error, keep the join to be safe.
func filtersNeedAnnotation(query SelectBuilder) bool {
sql, _, err := query.Columns("1").ToSql()
if err != nil {
return true
}
return annotationColumnRE().MatchString(sql)
}
func (r sqlRepository) withAnnotation(query SelectBuilder, idField string) SelectBuilder {
userID := loggedUser(r.ctx).ID
if userID == invalidUserId {
return query.Columns(fmt.Sprintf("%s.average_rating", r.tableName))
}
query = query.
LeftJoin("annotation on ("+
"annotation.item_id = "+idField+
" AND annotation.item_type = ?"+
" AND annotation.user_id = ?)", r.tableName, userID).
Columns(
"coalesce(starred, 0) as starred",
"coalesce(rating, 0) as rating",
"starred_at",
"play_date",
"rated_at",
)
if conf.Server.AlbumPlayCountMode == consts.AlbumPlayCountModeNormalized && r.tableName == "album" {
query = query.Columns(
fmt.Sprintf("round(coalesce(round(cast(play_count as float) / coalesce(%[1]s.song_count, 1), 1), 0)) as play_count", r.tableName),
)
} else {
query = query.Columns("coalesce(play_count, 0) as play_count")
}
query = query.Columns(fmt.Sprintf("%s.average_rating", r.tableName))
return query
}
func annotationBoolFilter(field string) func(string, any) Sqlizer {
return func(_ string, value any) Sqlizer {
v, ok := value.(string)
if !ok {
return nil
}
if strings.ToLower(v) == "true" {
return Expr(fmt.Sprintf("COALESCE(%s, 0) > 0", field))
}
return Expr(fmt.Sprintf("COALESCE(%s, 0) = 0", field))
}
}
func (r sqlRepository) annId(itemID ...string) And {
userID := loggedUser(r.ctx).ID
return And{
Eq{annotationTable + ".user_id": userID},
Eq{annotationTable + ".item_type": r.tableName},
Eq{annotationTable + ".item_id": itemID},
}
}
func (r sqlRepository) annUpsert(values map[string]any, itemIDs ...string) error {
upd := Update(annotationTable).Where(r.annId(itemIDs...))
for f, v := range values {
upd = upd.Set(f, v)
}
c, err := r.executeSQL(upd)
if c == 0 || errors.Is(err, sql.ErrNoRows) {
userID := loggedUser(r.ctx).ID
for _, itemID := range itemIDs {
values["user_id"] = userID
values["item_type"] = r.tableName
values["item_id"] = itemID
ins := Insert(annotationTable).SetMap(values)
_, err = r.executeSQL(ins)
if err != nil {
return err
}
}
}
return err
}
func (r sqlRepository) SetStar(starred bool, ids ...string) error {
starredAt := time.Now()
return r.annUpsert(map[string]any{"starred": starred, "starred_at": starredAt}, ids...)
}
func (r sqlRepository) SetRating(rating int, itemID string) error {
ratedAt := time.Now()
err := r.annUpsert(map[string]any{"rating": rating, "rated_at": ratedAt}, itemID)
if err != nil {
return err
}
return r.updateAvgRating(itemID)
}
func (r sqlRepository) updateAvgRating(itemID string) error {
upd := Update(r.tableName).
Where(Eq{"id": itemID}).
Set("average_rating", Expr(
"coalesce((select round(avg(rating), 2) from annotation where item_id = ? and item_type = ? and rating > 0), 0)",
itemID, r.tableName,
))
_, err := r.executeSQL(upd)
return err
}
func (r sqlRepository) IncPlayCount(itemID string, ts time.Time) error {
upd := Update(annotationTable).Where(r.annId(itemID)).
Set("play_count", Expr("play_count+1")).
Set("play_date", Expr("max(ifnull(play_date,''),?)", ts))
c, err := r.executeSQL(upd)
if c == 0 || errors.Is(err, sql.ErrNoRows) {
userID := loggedUser(r.ctx).ID
values := map[string]any{}
values["user_id"] = userID
values["item_type"] = r.tableName
values["item_id"] = itemID
values["play_count"] = 1
values["play_date"] = ts
ins := Insert(annotationTable).SetMap(values)
_, err = r.executeSQL(ins)
if err != nil {
return err
}
}
return err
}
func (r sqlRepository) ReassignAnnotation(prevID string, newID string) error {
if prevID == newID || prevID == "" || newID == "" {
return nil
}
// OR IGNORE keeps newID's own row where a user annotated both, instead of aborting the whole statement
upd := Expr("update or ignore "+annotationTable+" set item_id = ? where item_type = ? and item_id = ?",
newID, r.tableName, prevID)
if _, err := r.executeSQL(upd); err != nil {
return err
}
// The moved rows change newID's rating population, so its cached average no longer matches
return r.updateAvgRating(newID)
}
func (r sqlRepository) cleanAnnotations() error {
del := Delete(annotationTable).Where(Eq{"item_type": r.tableName}).Where("item_id not in (select id from " + r.tableName + ")")
c, err := r.executeSQL(del)
if err != nil {
return fmt.Errorf("error cleaning up %s annotations: %w", r.tableName, err)
}
if c > 0 {
log.Debug(r.ctx, "Clean-up annotations", "table", r.tableName, "totalDeleted", c)
}
return nil
}