Merge branch 'main' into fix/issue-29

Reviewed-on: #36

internal/blobgen/compress_test.go

@@ -0,0 +1,64 @@
+package blobgen
+
+import (
+	"bytes"
+	"crypto/rand"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// testRecipient is a static age recipient for tests.
+const testRecipient = "age1cplgrwj77ta54dnmydvvmzn64ltk83ankxl5sww04mrtmu62kv3s89gmvv"
+
+// TestCompressStreamNoDoubleClose is a regression test for issue #28.
+// It verifies that CompressStream does not panic or return an error due to
+// double-closing the underlying blobgen.Writer. Before the fix in PR #33,
+// the explicit Close() on the happy path combined with defer Close() would
+// cause a double close.
+func TestCompressStreamNoDoubleClose(t *testing.T) {
+	input := []byte("regression test data for issue #28 double-close fix")
+	var buf bytes.Buffer
+
+	written, hash, err := CompressStream(&buf, bytes.NewReader(input), 3, []string{testRecipient})
+	require.NoError(t, err, "CompressStream should not return an error")
+	assert.True(t, written > 0, "expected bytes written > 0")
+	assert.NotEmpty(t, hash, "expected non-empty hash")
+	assert.True(t, buf.Len() > 0, "expected non-empty output")
+}
+
+// TestCompressStreamLargeInput exercises CompressStream with a larger payload
+// to ensure no double-close issues surface under heavier I/O.
+func TestCompressStreamLargeInput(t *testing.T) {
+	data := make([]byte, 512*1024) // 512 KB
+	_, err := rand.Read(data)
+	require.NoError(t, err)
+
+	var buf bytes.Buffer
+	written, hash, err := CompressStream(&buf, bytes.NewReader(data), 3, []string{testRecipient})
+	require.NoError(t, err)
+	assert.True(t, written > 0)
+	assert.NotEmpty(t, hash)
+}
+
+// TestCompressStreamEmptyInput verifies CompressStream handles empty input
+// without double-close issues.
+func TestCompressStreamEmptyInput(t *testing.T) {
+	var buf bytes.Buffer
+	_, hash, err := CompressStream(&buf, strings.NewReader(""), 3, []string{testRecipient})
+	require.NoError(t, err)
+	assert.NotEmpty(t, hash)
+}
+
+// TestCompressDataNoDoubleClose mirrors the stream test for CompressData,
+// ensuring the explicit Close + error-path Close pattern is also safe.
+func TestCompressDataNoDoubleClose(t *testing.T) {
+	input := []byte("CompressData regression test for double-close")
+	result, err := CompressData(input, 3, []string{testRecipient})
+	require.NoError(t, err)
+	assert.True(t, result.CompressedSize > 0)
+	assert.True(t, result.UncompressedSize == int64(len(input)))
+	assert.NotEmpty(t, result.SHA256)
+}

internal/vaultik/blob_fetch_stub.go

@@ -0,0 +1,55 @@
+package vaultik
+
+import (
+	"context"
+	"fmt"
+	"io"
+
+	"filippo.io/age"
+	"git.eeqj.de/sneak/vaultik/internal/blobgen"
+)
+
+// FetchAndDecryptBlobResult holds the result of fetching and decrypting a blob.
+type FetchAndDecryptBlobResult struct {
+	Data []byte
+}
+
+// FetchAndDecryptBlob downloads a blob, decrypts it, and returns the plaintext data.
+func (v *Vaultik) FetchAndDecryptBlob(ctx context.Context, blobHash string, expectedSize int64, identity age.Identity) (*FetchAndDecryptBlobResult, error) {
+	rc, _, err := v.FetchBlob(ctx, blobHash, expectedSize)
+	if err != nil {
+		return nil, err
+	}
+	defer func() { _ = rc.Close() }()
+
+	reader, err := blobgen.NewReader(rc, identity)
+	if err != nil {
+		return nil, fmt.Errorf("creating blob reader: %w", err)
+	}
+	defer func() { _ = reader.Close() }()
+
+	data, err := io.ReadAll(reader)
+	if err != nil {
+		return nil, fmt.Errorf("reading blob data: %w", err)
+	}
+
+	return &FetchAndDecryptBlobResult{Data: data}, nil
+}
+
+// FetchBlob downloads a blob and returns a reader for the encrypted data.
+func (v *Vaultik) FetchBlob(ctx context.Context, blobHash string, expectedSize int64) (io.ReadCloser, int64, error) {
+	blobPath := fmt.Sprintf("blobs/%s/%s/%s", blobHash[:2], blobHash[2:4], blobHash)
+
+	rc, err := v.Storage.Get(ctx, blobPath)
+	if err != nil {
+		return nil, 0, fmt.Errorf("downloading blob %s: %w", blobHash[:16], err)
+	}
+
+	info, err := v.Storage.Stat(ctx, blobPath)
+	if err != nil {
+		_ = rc.Close()
+		return nil, 0, fmt.Errorf("stat blob %s: %w", blobHash[:16], err)
+	}
+
+	return rc, info.Size, nil
+}

internal/vaultik/blobcache.go

@@ -0,0 +1,207 @@
+package vaultik
+
+import (
+	"fmt"
+	"os"
+	"path/filepath"
+	"sync"
+)
+
+// blobDiskCacheEntry tracks a cached blob on disk.
+type blobDiskCacheEntry struct {
+	key  string
+	size int64
+	prev *blobDiskCacheEntry
+	next *blobDiskCacheEntry
+}
+
+// blobDiskCache is an LRU cache that stores blobs on disk instead of in memory.
+// Blobs are written to a temp directory keyed by their hash. When total size
+// exceeds maxBytes, the least-recently-used entries are evicted (deleted from disk).
+type blobDiskCache struct {
+	mu       sync.Mutex
+	dir      string
+	maxBytes int64
+	curBytes int64
+	items    map[string]*blobDiskCacheEntry
+	head     *blobDiskCacheEntry // most recent
+	tail     *blobDiskCacheEntry // least recent
+}
+
+// newBlobDiskCache creates a new disk-based blob cache with the given max size.
+func newBlobDiskCache(maxBytes int64) (*blobDiskCache, error) {
+	dir, err := os.MkdirTemp("", "vaultik-blobcache-*")
+	if err != nil {
+		return nil, fmt.Errorf("creating blob cache dir: %w", err)
+	}
+	return &blobDiskCache{
+		dir:      dir,
+		maxBytes: maxBytes,
+		items:    make(map[string]*blobDiskCacheEntry),
+	}, nil
+}
+
+func (c *blobDiskCache) path(key string) string {
+	return filepath.Join(c.dir, key)
+}
+
+func (c *blobDiskCache) unlink(e *blobDiskCacheEntry) {
+	if e.prev != nil {
+		e.prev.next = e.next
+	} else {
+		c.head = e.next
+	}
+	if e.next != nil {
+		e.next.prev = e.prev
+	} else {
+		c.tail = e.prev
+	}
+	e.prev = nil
+	e.next = nil
+}
+
+func (c *blobDiskCache) pushFront(e *blobDiskCacheEntry) {
+	e.prev = nil
+	e.next = c.head
+	if c.head != nil {
+		c.head.prev = e
+	}
+	c.head = e
+	if c.tail == nil {
+		c.tail = e
+	}
+}
+
+func (c *blobDiskCache) evictLRU() {
+	if c.tail == nil {
+		return
+	}
+	victim := c.tail
+	c.unlink(victim)
+	delete(c.items, victim.key)
+	c.curBytes -= victim.size
+	_ = os.Remove(c.path(victim.key))
+}
+
+// Put writes blob data to disk cache. Entries larger than maxBytes are silently skipped.
+func (c *blobDiskCache) Put(key string, data []byte) error {
+	entrySize := int64(len(data))
+
+	c.mu.Lock()
+	defer c.mu.Unlock()
+
+	if entrySize > c.maxBytes {
+		return nil
+	}
+
+	// Remove old entry if updating
+	if e, ok := c.items[key]; ok {
+		c.unlink(e)
+		c.curBytes -= e.size
+		_ = os.Remove(c.path(key))
+		delete(c.items, key)
+	}
+
+	if err := os.WriteFile(c.path(key), data, 0600); err != nil {
+		return fmt.Errorf("writing blob to cache: %w", err)
+	}
+
+	e := &blobDiskCacheEntry{key: key, size: entrySize}
+	c.pushFront(e)
+	c.items[key] = e
+	c.curBytes += entrySize
+
+	for c.curBytes > c.maxBytes && c.tail != nil {
+		c.evictLRU()
+	}
+
+	return nil
+}
+
+// Get reads a cached blob from disk. Returns data and true on hit.
+func (c *blobDiskCache) Get(key string) ([]byte, bool) {
+	c.mu.Lock()
+	e, ok := c.items[key]
+	if !ok {
+		c.mu.Unlock()
+		return nil, false
+	}
+	c.unlink(e)
+	c.pushFront(e)
+	c.mu.Unlock()
+
+	data, err := os.ReadFile(c.path(key))
+	if err != nil {
+		c.mu.Lock()
+		if e2, ok2 := c.items[key]; ok2 && e2 == e {
+			c.unlink(e)
+			delete(c.items, key)
+			c.curBytes -= e.size
+		}
+		c.mu.Unlock()
+		return nil, false
+	}
+	return data, true
+}
+
+// ReadAt reads a slice of a cached blob without loading the entire blob into memory.
+func (c *blobDiskCache) ReadAt(key string, offset, length int64) ([]byte, error) {
+	c.mu.Lock()
+	e, ok := c.items[key]
+	if !ok {
+		c.mu.Unlock()
+		return nil, fmt.Errorf("key %q not in cache", key)
+	}
+	if offset+length > e.size {
+		c.mu.Unlock()
+		return nil, fmt.Errorf("read beyond blob size: offset=%d length=%d size=%d", offset, length, e.size)
+	}
+	c.unlink(e)
+	c.pushFront(e)
+	c.mu.Unlock()
+
+	f, err := os.Open(c.path(key))
+	if err != nil {
+		return nil, err
+	}
+	defer func() { _ = f.Close() }()
+
+	buf := make([]byte, length)
+	if _, err := f.ReadAt(buf, offset); err != nil {
+		return nil, err
+	}
+	return buf, nil
+}
+
+// Has returns whether a key exists in the cache.
+func (c *blobDiskCache) Has(key string) bool {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+	_, ok := c.items[key]
+	return ok
+}
+
+// Size returns current total cached bytes.
+func (c *blobDiskCache) Size() int64 {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+	return c.curBytes
+}
+
+// Len returns number of cached entries.
+func (c *blobDiskCache) Len() int {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+	return len(c.items)
+}
+
+// Close removes the cache directory and all cached blobs.
+func (c *blobDiskCache) Close() error {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+	c.items = nil
+	c.head = nil
+	c.tail = nil
+	c.curBytes = 0
+	return os.RemoveAll(c.dir)
+}

internal/vaultik/blobcache_test.go

@@ -0,0 +1,189 @@
+package vaultik
+
+import (
+	"bytes"
+	"crypto/rand"
+	"fmt"
+	"testing"
+)
+
+func TestBlobDiskCache_BasicGetPut(t *testing.T) {
+	cache, err := newBlobDiskCache(1 << 20)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = cache.Close() }()
+
+	data := []byte("hello world")
+	if err := cache.Put("key1", data); err != nil {
+		t.Fatal(err)
+	}
+
+	got, ok := cache.Get("key1")
+	if !ok {
+		t.Fatal("expected cache hit")
+	}
+	if !bytes.Equal(got, data) {
+		t.Fatalf("got %q, want %q", got, data)
+	}
+
+	_, ok = cache.Get("nonexistent")
+	if ok {
+		t.Fatal("expected cache miss")
+	}
+}
+
+func TestBlobDiskCache_EvictionUnderPressure(t *testing.T) {
+	maxBytes := int64(1000)
+	cache, err := newBlobDiskCache(maxBytes)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = cache.Close() }()
+
+	for i := 0; i < 5; i++ {
+		data := make([]byte, 300)
+		if err := cache.Put(fmt.Sprintf("key%d", i), data); err != nil {
+			t.Fatal(err)
+		}
+	}
+
+	if cache.Size() > maxBytes {
+		t.Fatalf("cache size %d exceeds max %d", cache.Size(), maxBytes)
+	}
+
+	if !cache.Has("key4") {
+		t.Fatal("expected key4 to be cached")
+	}
+	if cache.Has("key0") {
+		t.Fatal("expected key0 to be evicted")
+	}
+}
+
+func TestBlobDiskCache_OversizedEntryRejected(t *testing.T) {
+	cache, err := newBlobDiskCache(100)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = cache.Close() }()
+
+	data := make([]byte, 200)
+	if err := cache.Put("big", data); err != nil {
+		t.Fatal(err)
+	}
+
+	if cache.Has("big") {
+		t.Fatal("oversized entry should not be cached")
+	}
+}
+
+func TestBlobDiskCache_UpdateInPlace(t *testing.T) {
+	cache, err := newBlobDiskCache(1 << 20)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = cache.Close() }()
+
+	if err := cache.Put("key1", []byte("v1")); err != nil {
+		t.Fatal(err)
+	}
+	if err := cache.Put("key1", []byte("version2")); err != nil {
+		t.Fatal(err)
+	}
+
+	got, ok := cache.Get("key1")
+	if !ok {
+		t.Fatal("expected hit")
+	}
+	if string(got) != "version2" {
+		t.Fatalf("got %q, want %q", got, "version2")
+	}
+	if cache.Len() != 1 {
+		t.Fatalf("expected 1 entry, got %d", cache.Len())
+	}
+	if cache.Size() != int64(len("version2")) {
+		t.Fatalf("expected size %d, got %d", len("version2"), cache.Size())
+	}
+}
+
+func TestBlobDiskCache_ReadAt(t *testing.T) {
+	cache, err := newBlobDiskCache(1 << 20)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = cache.Close() }()
+
+	data := make([]byte, 1024)
+	if _, err := rand.Read(data); err != nil {
+		t.Fatal(err)
+	}
+	if err := cache.Put("blob1", data); err != nil {
+		t.Fatal(err)
+	}
+
+	chunk, err := cache.ReadAt("blob1", 100, 200)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if !bytes.Equal(chunk, data[100:300]) {
+		t.Fatal("ReadAt returned wrong data")
+	}
+
+	_, err = cache.ReadAt("blob1", 900, 200)
+	if err == nil {
+		t.Fatal("expected error for out-of-bounds read")
+	}
+
+	_, err = cache.ReadAt("missing", 0, 10)
+	if err == nil {
+		t.Fatal("expected error for missing key")
+	}
+}
+
+func TestBlobDiskCache_Close(t *testing.T) {
+	cache, err := newBlobDiskCache(1 << 20)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if err := cache.Put("key1", []byte("data")); err != nil {
+		t.Fatal(err)
+	}
+	if err := cache.Close(); err != nil {
+		t.Fatal(err)
+	}
+}
+
+func TestBlobDiskCache_LRUOrder(t *testing.T) {
+	cache, err := newBlobDiskCache(200)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = cache.Close() }()
+
+	d := make([]byte, 100)
+	if err := cache.Put("a", d); err != nil {
+		t.Fatal(err)
+	}
+	if err := cache.Put("b", d); err != nil {
+		t.Fatal(err)
+	}
+
+	// Access "a" to make it most recently used
+	cache.Get("a")
+
+	// Adding "c" should evict "b" (LRU), not "a"
+	if err := cache.Put("c", d); err != nil {
+		t.Fatal(err)
+	}
+
+	if !cache.Has("a") {
+		t.Fatal("expected 'a' to survive")
+	}
+	if !cache.Has("c") {
+		t.Fatal("expected 'c' to be present")
+	}
+	if cache.Has("b") {
+		t.Fatal("expected 'b' to be evicted")
+	}
+}

internal/vaultik/restore.go

@@ -109,7 +109,11 @@ func (v *Vaultik) Restore(opts *RestoreOptions) error {
 
 	// Step 5: Restore files
 	result := &RestoreResult{}
-	blobCache := make(map[string][]byte) // Cache downloaded and decrypted blobs
+	blobCache, err := newBlobDiskCache(4 * v.Config.BlobSizeLimit.Int64())
+	if err != nil {
+		return fmt.Errorf("creating blob cache: %w", err)
+	}
+	defer func() { _ = blobCache.Close() }()
 
 	for i, file := range files {
 		if v.ctx.Err() != nil {
@@ -299,7 +303,7 @@ func (v *Vaultik) restoreFile(
 	targetDir string,
 	identity age.Identity,
 	chunkToBlobMap map[string]*database.BlobChunk,
-	blobCache map[string][]byte,
+	blobCache *blobDiskCache,
 	result *RestoreResult,
 ) error {
 	// Calculate target path - use full original path under target directory
@@ -383,7 +387,7 @@ func (v *Vaultik) restoreRegularFile(
 	targetPath string,
 	identity age.Identity,
 	chunkToBlobMap map[string]*database.BlobChunk,
-	blobCache map[string][]byte,
+	blobCache *blobDiskCache,
 	result *RestoreResult,
 ) error {
 	// Get file chunks in order
@@ -417,13 +421,15 @@ func (v *Vaultik) restoreRegularFile(
 
 		// Download and decrypt blob if not cached
 		blobHashStr := blob.Hash.String()
-		blobData, ok := blobCache[blobHashStr]
+		blobData, ok := blobCache.Get(blobHashStr)
 		if !ok {
 			blobData, err = v.downloadBlob(ctx, blobHashStr, blob.CompressedSize, identity)
 			if err != nil {
 				return fmt.Errorf("downloading blob %s: %w", blobHashStr[:16], err)
 			}
-			blobCache[blobHashStr] = blobData
+			if putErr := blobCache.Put(blobHashStr, blobData); putErr != nil {
+				log.Debug("Failed to cache blob on disk", "hash", blobHashStr[:16], "error", putErr)
+			}
 			result.BlobsDownloaded++
 			result.BytesDownloaded += blob.CompressedSize
 		}

internal/vaultik/snapshot.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"os"
 	"path/filepath"
+	"regexp"
 	"sort"
 	"strings"
 	"text/tabwriter"
@@ -1126,12 +1127,20 @@ func (v *Vaultik) PruneDatabase() (*PruneResult, error) {
 	return result, nil
 }
 
-// getTableCount returns the count of rows in a table
+// validTableNameRe matches table names containing only lowercase alphanumeric characters and underscores.
+var validTableNameRe = regexp.MustCompile(`^[a-z0-9_]+$`)
+
+// getTableCount returns the count of rows in a table.
+// The tableName is sanitized to only allow [a-z0-9_] characters to prevent SQL injection.
 func (v *Vaultik) getTableCount(tableName string) (int64, error) {
 	if v.DB == nil {
 		return 0, nil
 	}
 
+	if !validTableNameRe.MatchString(tableName) {
+		return 0, fmt.Errorf("invalid table name: %q", tableName)
+	}
+
 	var count int64
 	query := fmt.Sprintf("SELECT COUNT(*) FROM %s", tableName)
 	err := v.DB.Conn().QueryRowContext(v.ctx, query).Scan(&count)

internal/vaultik/vaultik.go

@@ -129,7 +129,7 @@ func (v *Vaultik) GetFilesystem() afero.Fs {
 	return v.Fs
 }
 
-// printfStdout writes formatted output to stdout for user-facing messages.
+// printfStdout writes formatted output to stdout.
 func (v *Vaultik) printfStdout(format string, args ...any) {
 	_, _ = fmt.Fprintf(v.Stdout, format, args...)
 }
@@ -148,6 +148,7 @@ func (v *Vaultik) printfStderr(format string, args ...any) {
 func (v *Vaultik) scanStdin(a ...any) (int, error) {
 	return fmt.Fscanln(v.Stdin, a...)
 }
+
 // TestVaultik wraps a Vaultik with captured stdout/stderr for testing
 type TestVaultik struct {
 	*Vaultik