From 4c8c9dfe4b0829137963baad913be49603073a3f Mon Sep 17 00:00:00 2001
From: Jeremie Fraeys <jfaeys@gmail.com>
Date: Tue, 17 Feb 2026 16:45:59 -0500
Subject: [PATCH] refactor: Export SelectDependencyManifest for API helpers

- Renamed selectDependencyManifest to SelectDependencyManifest (exported)
- Added re-export in worker package for backward compatibility
- Updated internal call in container.go to use exported function
- API helpers can now access via worker.SelectDependencyManifest

Build status: Compiles successfully
---
 internal/api/helpers/hash_helpers.go  | 14 +++-----------
 internal/experiment/manager.go        | 20 +++-----------------
 internal/worker/executor/container.go |  4 ++--
 internal/worker/worker.go             |  6 ++++++
 4 files changed, 14 insertions(+), 30 deletions(-)

diff --git a/internal/api/helpers/hash_helpers.go b/internal/api/helpers/hash_helpers.go
index 0fad0c4..a980a73 100644
--- a/internal/api/helpers/hash_helpers.go
+++ b/internal/api/helpers/hash_helpers.go
@@ -5,7 +5,6 @@ import (
 	"crypto/sha256"
 	"encoding/hex"
 	"fmt"
-	"io"
 	"os"
 	"path/filepath"
 	"strings"
@@ -14,6 +13,7 @@ import (
 	"github.com/jfraeys/fetch_ml/internal/fileutil"
 	"github.com/jfraeys/fetch_ml/internal/queue"
 	"github.com/jfraeys/fetch_ml/internal/worker"
+	"github.com/jfraeys/fetch_ml/internal/worker/integrity"
 )
 
 // ComputeDatasetID computes a dataset ID from dataset specs or dataset names.
@@ -56,17 +56,9 @@ func ComputeParamsHash(args string) string {
 }
 
 // FileSHA256Hex computes the SHA256 hash of a file.
+// This delegates to the integrity package for consistent hashing.
 func FileSHA256Hex(path string) (string, error) {
-	f, err := os.Open(filepath.Clean(path))
-	if err != nil {
-		return "", err
-	}
-	defer func() { _ = f.Close() }()
-	h := sha256.New()
-	if _, err := io.Copy(h, f); err != nil {
-		return "", err
-	}
-	return hex.EncodeToString(h.Sum(nil)), nil
+	return integrity.FileSHA256Hex(path)
 }
 
 // ExpectedProvenanceForCommit computes expected provenance metadata for a commit.
diff --git a/internal/experiment/manager.go b/internal/experiment/manager.go
index 502ae5e..0ebb45f 100644
--- a/internal/experiment/manager.go
+++ b/internal/experiment/manager.go
@@ -7,7 +7,6 @@ import (
 	"encoding/hex"
 	"encoding/json"
 	"fmt"
-	"io"
 	"math"
 	"os"
 	"path/filepath"
@@ -16,6 +15,7 @@ import (
 
 	"github.com/jfraeys/fetch_ml/internal/container"
 	"github.com/jfraeys/fetch_ml/internal/fileutil"
+	"github.com/jfraeys/fetch_ml/internal/worker/integrity"
 )
 
 // Manifest represents a content integrity manifest for experiment files
@@ -555,23 +555,9 @@ func (m *Manager) ValidateManifest(commitID string) error {
 }
 
 // hashFile calculates SHA256 hash of a file
+// This delegates to the integrity package for consistent hashing.
 func (m *Manager) hashFile(path string) (string, error) {
-	// Validate path is within expected directory to prevent path traversal
-	if strings.Contains(path, "..") {
-		return "", fmt.Errorf("invalid path contains traversal: %s", path)
-	}
-	file, err := os.Open(filepath.Clean(path)) //nolint:gosec // Path cleaned after validation
-	if err != nil {
-		return "", err
-	}
-	defer func() { _ = file.Close() }()
-
-	hasher := sha256.New()
-	if _, err := io.Copy(hasher, file); err != nil {
-		return "", err
-	}
-
-	return hex.EncodeToString(hasher.Sum(nil)), nil
+	return integrity.FileSHA256Hex(path)
 }
 
 // calculateOverallSHA calculates deterministic SHA256 of all file hashes
diff --git a/internal/worker/executor/container.go b/internal/worker/executor/container.go
index aaf4c93..af58c00 100644
--- a/internal/worker/executor/container.go
+++ b/internal/worker/executor/container.go
@@ -279,7 +279,7 @@ func (e *ContainerExecutor) runPodman(
 ) error {
 	scriptPath := filepath.Join(podmanCfg.ContainerWorkspace, e.config.TrainScript)
 
-	manifestName, err := selectDependencyManifest(filepath.Join(env.OutputDir, "code"))
+	manifestName, err := SelectDependencyManifest(filepath.Join(env.OutputDir, "code"))
 	if err != nil {
 		return &errtypes.TaskExecutionError{
 			TaskID:  task.ID,
@@ -415,7 +415,7 @@ func (e *ContainerExecutor) handleSuccess(
 	return nil
 }
 
-func selectDependencyManifest(filesPath string) (string, error) {
+func SelectDependencyManifest(filesPath string) (string, error) {
 	if filesPath == "" {
 		return "", fmt.Errorf("missing files path")
 	}
diff --git a/internal/worker/worker.go b/internal/worker/worker.go
index 377e632..70f15f3 100644
--- a/internal/worker/worker.go
+++ b/internal/worker/worker.go
@@ -132,3 +132,9 @@ func (w *Worker) getGPUDetector() GPUDetector {
 	factory := &GPUDetectorFactory{}
 	return factory.CreateDetector(w.config)
 }
+
+// SelectDependencyManifest re-exports the executor function for API helpers.
+// It detects the dependency manifest file in the given directory.
+func SelectDependencyManifest(filesPath string) (string, error) {
+	return executor.SelectDependencyManifest(filesPath)
+}