Record each snapshot's uncompressed size when its backup runs (closes #360)
check / check (push) Waiting to run

snapshot list summed the chunks a snapshot reaches through its file rows
in the local index. A file's row is shared by every snapshot that lists
it and rewritten when the file changes, so backing up a changed file
changed every older snapshot's UNCOMPRESSED SIZE.

The backup now takes that total when it records the snapshot's stats,
into a new snapshots column, uncompressed_chunk_size, and snapshot list
reads the column, as it already does for new_chunk_size.

Model: opus-5-5
This commit was merged in pull request #366.
This commit is contained in:
2026-10-13 20:43:19 +02:00
parent fb8abcd049
commit 3361e9ebd9
6 changed files with 120 additions and 25 deletions
+23 -17
View File
@@ -120,7 +120,11 @@ func (r *SnapshotRepository) UpdateCounts(
return nil
}
// UpdateExtendedStats updates extended statistics for a snapshot
// UpdateExtendedStats updates extended statistics for a snapshot. It also
// records the total plaintext size of the distinct chunks the snapshot's
// files reference. Only the snapshot's own backup can take that total: a
// file's row is shared by every snapshot that lists it, and a later
// backup of a changed file rewrites the row with the new chunks.
func (r *SnapshotRepository) UpdateExtendedStats(
ctx context.Context,
tx *sql.Tx,
@@ -145,7 +149,17 @@ func (r *SnapshotRepository) UpdateExtendedStats(
compression_level = ?,
upload_bytes = ?,
upload_duration_ms = ?,
new_chunk_size = ?
new_chunk_size = ?,
uncompressed_chunk_size = (
SELECT COALESCE(SUM(c.size), 0)
FROM (
SELECT DISTINCT fc.chunk_hash
FROM snapshot_files sf
JOIN file_chunks fc ON sf.file_id = fc.file_id
WHERE sf.snapshot_id = snapshots.id
) sc
JOIN chunks c ON sc.chunk_hash = c.chunk_hash
)
WHERE id = ?
`
@@ -555,30 +569,22 @@ func (r *SnapshotRepository) GetSnapshotBlobSizes(
return compressed, uncompressed, nil
}
// GetSnapshotUncompressedChunkSize returns the sum of plaintext sizes of all unique
// chunks referenced by a snapshot (via snapshot_files → file_chunks → chunks).
// GetSnapshotUncompressedChunkSize returns the total plaintext size of the
// distinct chunks the snapshot's files reference, as the snapshot's backup
// recorded it (see UpdateExtendedStats).
func (r *SnapshotRepository) GetSnapshotUncompressedChunkSize(
ctx context.Context, snapshotID string,
) (int64, error) {
query := `
SELECT COALESCE(SUM(c.size), 0)
FROM (
SELECT DISTINCT fc.chunk_hash
FROM snapshot_files sf
JOIN file_chunks fc ON sf.file_id = fc.file_id
WHERE sf.snapshot_id = ?
) sc
JOIN chunks c ON sc.chunk_hash = c.chunk_hash
`
query := `SELECT uncompressed_chunk_size FROM snapshots WHERE id = ?`
var totalSize int64
var uncompressedChunkSize int64
err := r.db.conn.QueryRowContext(ctx, query, snapshotID).Scan(&totalSize)
err := r.db.conn.QueryRowContext(ctx, query, snapshotID).Scan(&uncompressedChunkSize)
if err != nil {
return 0, fmt.Errorf("querying uncompressed chunk size: %w", err)
}
return totalSize, nil
return uncompressedChunkSize, nil
}
// GetSnapshotNewChunkSize returns the total plaintext size of the chunks the