batch upload impl

This commit is contained in:
jay brown
2025-08-06 15:06:18 -07:00
parent fac5615c15
commit 7014fa790b
17 changed files with 1272 additions and 97 deletions
@@ -0,0 +1,84 @@
package storage
import (
"bytes"
"context"
"fmt"
"io"
"path/filepath"
"time"
"github.com/aws/aws-sdk-go-v2/service/s3"
"github.com/google/uuid"
"queryorchestration/internal/serviceconfig/objectstore"
)
type Service struct {
cfg objectstore.ConfigProvider
}
type StorageResult struct {
Bucket string
Key string
SizeBytes int64
}
func New(cfg objectstore.ConfigProvider) *Service {
return &Service{cfg: cfg}
}
// StoreZIPFile uploads ZIP file to S3 and returns storage metadata
func (s *Service) StoreZIPFile(ctx context.Context, clientID string, filename string, content io.Reader) (*StorageResult, error) {
// Create unique S3 key with timestamp and UUID
timestamp := time.Now().Format("2006/01/02/15")
batchID := uuid.New()
key := fmt.Sprintf("batches/%s/%s/%s-%s", clientID, timestamp, batchID.String(), filename)
// Get bucket from existing configuration
bucketName := s.cfg.GetBucket()
// Read content to get size (required for S3 PutObject)
buf := &bytes.Buffer{}
size, err := io.Copy(buf, content)
if err != nil {
return nil, fmt.Errorf("failed to read file content: %w", err)
}
// Upload to S3 using the store client
_, err = s.cfg.GetStoreClient().PutObject(ctx, &s3.PutObjectInput{
Bucket: &bucketName,
Key: &key,
Body: bytes.NewReader(buf.Bytes()),
})
if err != nil {
return nil, fmt.Errorf("failed to upload ZIP to S3: %w", err)
}
return &StorageResult{
Bucket: bucketName,
Key: key,
SizeBytes: size,
}, nil
}
// ValidateZIPFile performs basic ZIP file validation
func (s *Service) ValidateZIPFile(filename string, sizeBytes int64) error {
// Check file extension
ext := filepath.Ext(filename)
if ext != ".zip" {
return fmt.Errorf("invalid file type: expected .zip, got %s", ext)
}
// Check file size (100MB limit for Part 1)
const maxSizeBytes = 100 * 1024 * 1024 // 100MB
if sizeBytes > maxSizeBytes {
return fmt.Errorf("file too large: %d bytes exceeds limit of %d bytes", sizeBytes, maxSizeBytes)
}
if sizeBytes == 0 {
return fmt.Errorf("empty file not allowed")
}
return nil
}
@@ -0,0 +1,165 @@
package storage
import (
"archive/zip"
"bytes"
"fmt"
"testing"
"github.com/aws/aws-sdk-go-v2/service/s3"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"queryorchestration/internal/serviceconfig"
"queryorchestration/internal/serviceconfig/objectstore"
"queryorchestration/internal/test"
)
// TestConfig provides a test configuration that implements both service config and AWS interfaces
type TestConfig struct {
serviceconfig.BaseConfig
objectstore.ObjectStoreConfig
}
func TestStoreZIPFile_Success(t *testing.T) {
// Setup: Create testcontainer with localstack
cfg := &TestConfig{}
test.CreateDB(t, cfg)
test.CreateAWSResources(t, cfg)
// Create storage service
service := New(cfg)
// Create test ZIP content with 3 PDFs
zipContent := createTestZIPContent(t)
// Test: Store ZIP file
result, err := service.StoreZIPFile(t.Context(), "test-client", "test.zip", zipContent)
// Verify: Storage result
require.NoError(t, err)
assert.NotEmpty(t, result.Bucket)
assert.NotEmpty(t, result.Key)
assert.Greater(t, result.SizeBytes, int64(0))
assert.Contains(t, result.Key, "test-client")
assert.Contains(t, result.Key, "test.zip")
// Verify: File exists in S3 (using actual S3 client)
s3Client := cfg.GetStoreClient()
_, err = s3Client.HeadObject(t.Context(), &s3.HeadObjectInput{
Bucket: &result.Bucket,
Key: &result.Key,
})
require.NoError(t, err, "ZIP file should exist in S3")
}
func TestStoreZIPFile_LargeFile(t *testing.T) {
// Setup: Create testcontainer with localstack
cfg := &TestConfig{}
test.CreateDB(t, cfg)
test.CreateAWSResources(t, cfg)
// Create storage service
service := New(cfg)
// Create a large test ZIP content (but still under 100MB limit)
var buf bytes.Buffer
zipWriter := zip.NewWriter(&buf)
// Create multiple PDFs to make a larger ZIP
for i := 0; i < 10; i++ {
filename := fmt.Sprintf("document_%d.pdf", i)
fileWriter, err := zipWriter.Create(filename)
require.NoError(t, err)
// Write a larger PDF content
pdfContent := fmt.Sprintf("%%PDF-1.4\n")
// Add some bulk content
for j := 0; j < 1000; j++ {
pdfContent += fmt.Sprintf("%%Page %d Line %d content\n", i, j)
}
pdfContent += "%%EOF"
_, err = fileWriter.Write([]byte(pdfContent))
require.NoError(t, err)
}
err := zipWriter.Close()
require.NoError(t, err)
// Test: Store larger ZIP file
result, err := service.StoreZIPFile(t.Context(), "test-client", "large-batch.zip", bytes.NewReader(buf.Bytes()))
// Verify: Storage result
require.NoError(t, err)
assert.NotEmpty(t, result.Bucket)
assert.NotEmpty(t, result.Key)
assert.Greater(t, result.SizeBytes, int64(0)) // Should have some size
}
func TestValidateZIPFile_Success(t *testing.T) {
service := New(nil) // No config needed for validation
// Test: Valid ZIP file
err := service.ValidateZIPFile("test.zip", 1024*1024) // 1MB
// Verify: No error
assert.NoError(t, err)
}
func TestValidateZIPFile_InvalidExtension(t *testing.T) {
service := New(nil)
// Test: Invalid file extension
err := service.ValidateZIPFile("test.pdf", 1024*1024)
// Verify: Error returned
require.Error(t, err)
assert.Contains(t, err.Error(), "invalid file type")
}
func TestValidateZIPFile_TooLarge(t *testing.T) {
service := New(nil)
// Test: File too large (101MB)
err := service.ValidateZIPFile("test.zip", 101*1024*1024)
// Verify: Error returned
require.Error(t, err)
assert.Contains(t, err.Error(), "file too large")
}
func TestValidateZIPFile_Empty(t *testing.T) {
service := New(nil)
// Test: Empty file
err := service.ValidateZIPFile("test.zip", 0)
// Verify: Error returned
require.Error(t, err)
assert.Contains(t, err.Error(), "empty file not allowed")
}
// Helper function to create test ZIP content with 3 PDF files
func createTestZIPContent(t *testing.T) *bytes.Reader {
var buf bytes.Buffer
zipWriter := zip.NewWriter(&buf)
// Create 3 PDF files in the ZIP
pdfFiles := []string{"test1.pdf", "test2.pdf", "test3.pdf"}
for _, filename := range pdfFiles {
fileWriter, err := zipWriter.Create(filename)
require.NoError(t, err)
// Write minimal PDF content (PDF magic header)
// Each file has slightly different content for identification
pdfContent := fmt.Sprintf("%%PDF-1.4\n%%Test content for %s\n%%%%EOF", filename)
_, err = fileWriter.Write([]byte(pdfContent))
require.NoError(t, err)
}
err := zipWriter.Close()
require.NoError(t, err)
return bytes.NewReader(buf.Bytes())
}