Reserve capacity for regular file artifacts

internal/artifact/reader.go

@@ -21,6 +21,7 @@ var (
 )
 
 const fileReadChunkSize = 64 * 1024
+const fileReadInitialCapacityLimit = 1 << 20
 
 // Reader resolves artifact references into actual artifacts.
 type Reader interface {
@@ -133,7 +134,11 @@ func readFileArtifact(ctx context.Context, path string, open func(string) (artif
 		return nil, fmt.Errorf("%w: %s", ErrUnsupportedFile, path)
 	}
 
-	data := make([]byte, 0)
+	capacity := 0
+	if size := openedInfo.Size(); size > 0 {
+		capacity = int(min(size, int64(fileReadInitialCapacityLimit)))
+	}
+	data := make([]byte, 0, capacity)
 	chunk := make([]byte, fileReadChunkSize)
 	for {
 		if err := ctx.Err(); err != nil {
@@ -153,6 +158,13 @@ func readFileArtifact(ctx context.Context, path string, open func(string) (artif
 			return nil, fmt.Errorf("failed to read file %s: %w", path, readErr)
 		}
 	}
+	if len(data) > 0 && cap(data)-len(data) > len(data) {
+		// A stale size hint should not make a small completed artifact retain a
+		// disproportionately large backing array.
+		compact := make([]byte, len(data))
+		copy(compact, data)
+		data = compact
+	}
 
 	contentType := mime.TypeByExtension(filepath.Ext(path))
 	if contentType == "" {

internal/artifact/reader_test.go

@@ -3,7 +3,9 @@ package artifact
 import (
 	"bytes"
 	"context"
+	"crypto/sha256"
 	"errors"
+	"fmt"
 	"os"
 	"path/filepath"
 	"strings"
@@ -232,6 +234,60 @@ func TestFileReaderFailuresAndMetadata(t *testing.T) {
 	})
 }
 
+func TestFileReaderUsesOpenedSizeOnlyAsACapacityHint(t *testing.T) {
+	tests := []struct {
+		name         string
+		body         []byte
+		reportedSize int64
+	}{
+		{name: "overreported size", body: []byte("actual file content"), reportedSize: 128},
+		{name: "underreported size", body: bytes.Repeat([]byte("x"), fileReadChunkSize+17), reportedSize: 1},
+		{name: "very large hint with small content", body: []byte("small actual content"), reportedSize: int64(^uint64(0) >> 1)},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			filePath := filepath.Join(t.TempDir(), "artifact.txt")
+			if err := os.WriteFile(filePath, tc.body, 0o600); err != nil {
+				t.Fatal(err)
+			}
+			reader := &fileReader{open: func(path string) (artifactFile, error) {
+				file, err := os.Open(path)
+				if err != nil {
+					return nil, err
+				}
+				info, err := file.Stat()
+				if err != nil {
+					file.Close()
+					return nil, err
+				}
+				return &reportedInfoFile{
+					artifactFile: file,
+					info:         reportedFileSizeInfo{FileInfo: info, size: tc.reportedSize},
+				}, nil
+			}}
+
+			artifact, err := reader.Read(context.Background(), domain.ArtifactRef{
+				Type: domain.ArtifactRefFile,
+				URI:  filePath,
+			})
+			if err != nil {
+				t.Fatalf("read artifact: %v", err)
+			}
+			if !bytes.Equal(artifact.Body, tc.body) {
+				t.Fatalf("artifact body length=%d, want %d", len(artifact.Body), len(tc.body))
+			}
+			if artifact.Size != int64(len(tc.body)) {
+				t.Fatalf("artifact size=%d, want %d", artifact.Size, len(tc.body))
+			}
+			wantHash := fmt.Sprintf("%x", sha256.Sum256(tc.body))
+			if artifact.Hash != wantHash {
+				t.Fatalf("artifact hash=%q, want %q", artifact.Hash, wantHash)
+			}
+		})
+	}
+}
+
 func TestFileReaderCancelsAfterReadProgress(t *testing.T) {
 	filePath := filepath.Join(t.TempDir(), "artifact.bin")
 	content := bytes.Repeat([]byte("x"), fileReadChunkSize*2)
@@ -268,6 +324,13 @@ func (f *reportedInfoFile) Stat() (os.FileInfo, error) {
 	return f.info, nil
 }
 
+type reportedFileSizeInfo struct {
+	os.FileInfo
+	size int64
+}
+
+func (info reportedFileSizeInfo) Size() int64 { return info.size }
+
 type cancelAfterProgressFile struct {
 	artifactFile
 	cancel context.CancelFunc