List a file:// destination through a symlinked directory (closes #296)

List and ListStream walked the destination with afero.Walk, which uses
Lstat and so reported a symlinked metadata/ or blobs/ directory as one
file, while Put, Get and Stat follow the link. The store then listed as
empty, and prune removed the local snapshot records and deleted every
blob. The listings, and DeletePartialUploads, which remote nuke uses,
now walk a view of the filesystem without its Lstat method, so afero
falls back to Stat and enters the link. A prefix lists as empty only
when it is truly missing: when it or a directory above it is a symlink
whose target is missing, such as a link to an unmounted volume, the
listing fails.

Judgement call: a symlink loop is not detected; it is followed until the
system refuses the path, and the listing fails.

Model: opus-5-5

README.md

@@ -294,8 +294,7 @@ needed.
 of the local index and the backup destination store — with timestamps
 and three sizes per snapshot (compressed remote size; total
 uncompressed chunk size; size of the chunks that snapshot stored that
-were not stored before). Both chunk sizes are recorded when the
-snapshot's backup runs.
+were not stored before, recorded when its backup ran).
 
 Listing the destination store does **not** require the age secret key,
 so it works in vaultik's intended configuration, where the backed-up

docs/DATAMODEL.md

@@ -122,7 +122,6 @@ Tracks backup snapshots.
 - `new_chunk_size` (INTEGER) - Total size of those chunks before compression; `snapshot list` shows it as NEW CHUNK SIZE
 - `blob_count` (INTEGER) - Number of blobs this snapshot created
 - `total_size` (INTEGER) - Total size of all files
-- `uncompressed_chunk_size` (INTEGER) - Total size before compression of the distinct chunks the snapshot's files reference, recorded when its backup ran; `snapshot list` shows it as UNCOMPRESSED SIZE
 - `blob_size` (INTEGER) - Total compressed size of all referenced blobs
 - `blob_uncompressed_size` (INTEGER) - Total uncompressed size of all referenced blobs
 - `compression_ratio` (REAL) - Compression ratio achieved

internal/database/schema/001.sql

@@ -89,7 +89,6 @@ CREATE TABLE IF NOT EXISTS snapshots (
     new_chunk_size INTEGER NOT NULL DEFAULT 0,
     blob_count INTEGER NOT NULL DEFAULT 0,
     total_size INTEGER NOT NULL DEFAULT 0,
-    uncompressed_chunk_size INTEGER NOT NULL DEFAULT 0,
     blob_size INTEGER NOT NULL DEFAULT 0,
     blob_uncompressed_size INTEGER NOT NULL DEFAULT 0,
     compression_ratio REAL NOT NULL DEFAULT 1.0,

internal/database/snapshots.go

@@ -120,11 +120,7 @@ func (r *SnapshotRepository) UpdateCounts(
 	return nil
 }
 
-// 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.
+// UpdateExtendedStats updates extended statistics for a snapshot
 func (r *SnapshotRepository) UpdateExtendedStats(
 	ctx context.Context,
 	tx *sql.Tx,
@@ -149,17 +145,7 @@ func (r *SnapshotRepository) UpdateExtendedStats(
 		    compression_level = ?,
 		    upload_bytes = ?,
 		    upload_duration_ms = ?,
-		    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
-		    )
+		    new_chunk_size = ?
 		WHERE id = ?
 	`
 
@@ -569,22 +555,30 @@ func (r *SnapshotRepository) GetSnapshotBlobSizes(
 	return compressed, uncompressed, nil
 }
 
-// GetSnapshotUncompressedChunkSize returns the total plaintext size of the
-// distinct chunks the snapshot's files reference, as the snapshot's backup
-// recorded it (see UpdateExtendedStats).
+// GetSnapshotUncompressedChunkSize returns the sum of plaintext sizes of all unique
+// chunks referenced by a snapshot (via snapshot_files → file_chunks → chunks).
 func (r *SnapshotRepository) GetSnapshotUncompressedChunkSize(
 	ctx context.Context, snapshotID string,
 ) (int64, error) {
-	query := `SELECT uncompressed_chunk_size FROM snapshots WHERE id = ?`
+	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
+	`
 
-	var uncompressedChunkSize int64
+	var totalSize int64
 
-	err := r.db.conn.QueryRowContext(ctx, query, snapshotID).Scan(&uncompressedChunkSize)
+	err := r.db.conn.QueryRowContext(ctx, query, snapshotID).Scan(&totalSize)
 	if err != nil {
 		return 0, fmt.Errorf("querying uncompressed chunk size: %w", err)
 	}
 
-	return uncompressedChunkSize, nil
+	return totalSize, nil
 }
 
 // GetSnapshotNewChunkSize returns the total plaintext size of the chunks the

internal/vaultik/snapshot_list_test.go

@@ -483,14 +483,13 @@ func TestListSnapshots_UnreadableManifestDoesNotHideOthers(t *testing.T) {
 //
 //nolint:tagliatelle // snake_case is the established output format
 type listJSONRow struct {
-	ID               string `json:"id"`
-	RemoteKey        string `json:"remote_key"`
-	Timestamp        string `json:"timestamp"`
-	CompressedSize   int64  `json:"compressed_size"`
-	UncompressedSize int64  `json:"uncompressed_size"`
-	NewChunkSize     int64  `json:"new_chunk_size"`
-	LocallyTracked   bool   `json:"locally_tracked"`
-	RemotePresent    *bool  `json:"remote_present"`
+	ID             string `json:"id"`
+	RemoteKey      string `json:"remote_key"`
+	Timestamp      string `json:"timestamp"`
+	CompressedSize int64  `json:"compressed_size"`
+	NewChunkSize   int64  `json:"new_chunk_size"`
+	LocallyTracked bool   `json:"locally_tracked"`
+	RemotePresent  *bool  `json:"remote_present"`
 }
 
 // decodeListJSON parses the command's stdout, which must contain

internal/vaultik/uncompressed_size_test.go

@@ -1,85 +0,0 @@
-package vaultik_test
-
-import (
-	"bytes"
-	"context"
-	"path/filepath"
-	"testing"
-
-	"github.com/spf13/afero"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-	"sneak.berlin/go/vaultik/internal/database"
-	"sneak.berlin/go/vaultik/internal/log"
-	"sneak.berlin/go/vaultik/internal/storage"
-	"sneak.berlin/go/vaultik/internal/vaultik"
-)
-
-// TestUncompressedSizeOfChangedFileStaysFixed covers
-// https://git.eeqj.de/sneak/vaultik/issues/360. A backup that changes a
-// file leaves the earlier snapshot's UNCOMPRESSED SIZE as it was when that
-// snapshot's backup completed. The file's row in the local index is shared
-// by both snapshots and rewritten by the second backup.
-func TestUncompressedSizeOfChangedFileStaysFixed(t *testing.T) {
-	log.Initialize(log.Config{})
-	t.Parallel()
-
-	fs := afero.NewOsFs()
-	tempDir := t.TempDir()
-	dataDir := filepath.Join(tempDir, "src")
-	dbPath := filepath.Join(tempDir, "index.sqlite")
-	changedPath := filepath.Join(dataDir, "changed.bin")
-
-	ctx := context.Background()
-	files := writeFaultSourceTree(t, fs, dataDir)
-	cfg := changedFileConfig(dataDir, dbPath)
-
-	appendRandomBytes(t, fs, files, changedPath, appendedFileSize)
-
-	store, err := storage.NewFileStorer(filepath.Join(tempDir, "remote"))
-	require.NoError(t, err)
-
-	db, err := database.New(ctx, dbPath)
-	require.NoError(t, err)
-	t.Cleanup(func() { _ = db.Close() })
-
-	repos := database.NewRepositories(db)
-	v := newBackupVaultik(ctx, cfg, store, repos, db, fs)
-
-	require.NoError(t, backUp(v, "first"))
-
-	// With nothing to compare against, sizeOfNewChunks totals every
-	// distinct chunk of the files.
-	firstID := localSnapshotID(ctx, t, repos, "first")
-	firstSize := sizeOfNewChunks(t, files, nil)
-	require.Equal(t, firstSize, listedUncompressedSizes(t, v)[firstID],
-		"the first snapshot must count the chunks of every file it holds")
-
-	appendRandomBytes(t, fs, files, changedPath, faultChunkSize)
-	require.NoError(t, backUp(v, "second"))
-
-	secondID := localSnapshotID(ctx, t, repos, "second")
-	sizes := listedUncompressedSizes(t, v)
-	assert.Equal(t, firstSize, sizes[firstID],
-		"a later backup must not change the first snapshot's figure")
-	assert.Equal(t, sizeOfNewChunks(t, files, nil), sizes[secondID],
-		"the second snapshot must count the changed file's chunks")
-}
-
-// listedUncompressedSizes runs `snapshot list --json` and returns each
-// listed snapshot's uncompressed_size by snapshot ID.
-func listedUncompressedSizes(t *testing.T, v *vaultik.Vaultik) map[string]int64 {
-	t.Helper()
-
-	var out bytes.Buffer
-
-	v.Stdout = &out
-	require.NoError(t, v.ListSnapshots(true))
-
-	sizes := make(map[string]int64)
-	for _, row := range decodeListJSON(t, out.String()) {
-		sizes[row.ID] = row.UncompressedSize
-	}
-
-	return sizes
-}