add helper function to get changed files

Author: joshrwolf

go.mod

@@ -15,6 +15,7 @@ require (
 	github.com/chainguard-dev/clog v1.5.1-0.20240811185937-4c523ae4593f
 	github.com/cloudevents/sdk-go/v2 v2.15.2
 	github.com/coreos/go-oidc/v3 v3.11.0
+	github.com/go-git/go-billy/v5 v5.5.0
 	github.com/go-git/go-git/v5 v5.12.0
 	github.com/google/go-cmp v0.6.0
 	github.com/google/go-github/v61 v61.0.0
@@ -57,7 +58,6 @@ require (
 	github.com/emirpasic/gods v1.18.1 // indirect
 	github.com/felixge/httpsnoop v1.0.4 // indirect
 	github.com/go-git/gcfg v1.5.1-0.20230307220236-3a3c6141e376 // indirect
-	github.com/go-git/go-billy/v5 v5.5.0 // indirect
 	github.com/go-jose/go-jose/v4 v4.0.4 // indirect
 	github.com/go-logr/logr v1.4.2 // indirect
 	github.com/go-logr/stdr v1.2.2 // indirect

modules/github-bots/sdk/github.go

@@ -18,6 +18,7 @@ import (
 	"github.com/go-git/go-git/v5/config"
 	"github.com/go-git/go-git/v5/plumbing"
 	gitHttp "github.com/go-git/go-git/v5/plumbing/transport/http"
+	"github.com/go-git/go-git/v5/utils/merkletrie"
 	"github.com/snabb/httpreaderat"
 
 	"chainguard.dev/sdk/octosts"
@@ -433,6 +434,83 @@ func (c GitHubClient) ListArtifactsFunc(ctx context.Context, wr *github.Workflow
 	return nil
 }
 
+// GetChangedFiles uses the git package to get a map of the files changed between two branches.
+func (c GitHubClient) GetChangedFiles(ctx context.Context, repo *git.Repository, from, to string) (map[string]struct{}, error) {
+	if repo == nil {
+		return nil, fmt.Errorf("repository is nil")
+	}
+
+	parseRef := func(ref string) (*plumbing.Reference, error) {
+		if plumbing.IsHash(ref) {
+			return plumbing.NewHashReference(plumbing.ReferenceName(ref), plumbing.NewHash(ref)), nil
+		}
+		return repo.Reference(plumbing.ReferenceName(ref), true)
+	}
+
+	fromRef, err := parseRef(from)
+	if err != nil {
+		return nil, fmt.Errorf("failed to get reference for %s: %w", from, err)
+	}
+
+	toRef, err := parseRef(to)
+	if err != nil {
+		return nil, fmt.Errorf("failed to get reference for %s: %w", to, err)
+	}
+
+	if fromRef == nil || toRef == nil {
+		return nil, fmt.Errorf("one or both references are nil")
+	}
+
+	fromCommit, err := repo.CommitObject(fromRef.Hash())
+	if err != nil {
+		return nil, fmt.Errorf("failed to get commit for %s: %w", fromRef.Hash().String(), err)
+	}
+
+	toCommit, err := repo.CommitObject(toRef.Hash())
+	if err != nil {
+		return nil, fmt.Errorf("failed to get commit for %s: %w", toRef.Hash().String(), err)
+	}
+
+	if fromCommit == nil || toCommit == nil {
+		return nil, fmt.Errorf("one or both commits are nil")
+	}
+
+	changed := make(map[string]struct{})
+
+	fromTree, err := fromCommit.Tree()
+	if err != nil {
+		return nil, fmt.Errorf("failed to get tree for %s: %w", fromRef.Hash().String(), err)
+	}
+
+	toTree, err := toCommit.Tree()
+	if err != nil {
+		return nil, fmt.Errorf("failed to get tree for %s: %w", toRef.String(), err)
+	}
+
+	changes, err := fromTree.Diff(toTree)
+	if err != nil {
+		return nil, fmt.Errorf("failed to diff trees: %w", err)
+	}
+
+	for _, change := range changes {
+		action, err := change.Action()
+		if err != nil {
+			return nil, fmt.Errorf("failed to get change action: %w", err)
+		}
+
+		switch action {
+		case merkletrie.Insert:
+			changed[change.To.Name] = struct{}{}
+		case merkletrie.Delete:
+			changed[change.From.Name] = struct{}{}
+		case merkletrie.Modify:
+			changed[change.From.Name] = struct{}{}
+		}
+	}
+
+	return changed, nil
+}
+
 func validateResponse(ctx context.Context, err error, resp *github.Response, action string) error {
 	// resp may be nil if err is nonempty. However, err may contain a rate limit
 	// error so we have to inspect for rate limiting if resp is non-nil

modules/github-bots/sdk/github_test.go

@@ -0,0 +1,205 @@
+package sdk_test
+
+import (
+	"context"
+	"fmt"
+	"testing"
+
+	"github.com/chainguard-dev/terraform-infra-common/modules/github-bots/sdk"
+	"github.com/go-git/go-billy/v5/osfs"
+	"github.com/go-git/go-git/v5"
+	"github.com/go-git/go-git/v5/plumbing"
+	"github.com/go-git/go-git/v5/plumbing/object"
+	"github.com/go-git/go-git/v5/storage/memory"
+	"github.com/google/go-cmp/cmp"
+)
+
+func TestGithubClient_GetChangedFiles(t *testing.T) {
+	ctx := context.Background()
+	client := sdk.NewGitHubClient(ctx, "org", "repo", "bot")
+
+	tests := []struct {
+		name     string
+		patches  []gpatch
+		expected map[string]struct{}
+		wantErr  bool
+	}{
+		{
+			name: "Added file",
+			patches: []gpatch{
+				{action: "add", path: "new.txt"},
+			},
+			expected: map[string]struct{}{
+				"new.txt": {},
+			},
+		},
+		{
+			name: "Modified file",
+			patches: []gpatch{
+				{action: "add", path: "existing.txt"},
+				{action: "modify", path: "existing.txt"},
+			},
+			expected: map[string]struct{}{
+				"existing.txt": {},
+			},
+		},
+		{
+			name: "Renamed file",
+			patches: []gpatch{
+				{action: "add", path: "old.txt"},
+				{action: "rename", path: "old.txt", newPath: "new.txt"},
+			},
+			expected: map[string]struct{}{
+				"new.txt": {},
+			},
+		},
+		{
+			name: "Deleted file",
+			patches: []gpatch{
+				{action: "add", path: "to_delete.txt"},
+				{action: "delete", path: "initial.txt"},
+			},
+			expected: map[string]struct{}{
+				"to_delete.txt": {},
+				"initial.txt":   {},
+			},
+		},
+		{
+			name: "Multiple changes",
+			patches: []gpatch{
+				{action: "add", path: "new.txt"},
+				{action: "add", path: "to_modify.txt"},
+				{action: "modify", path: "to_modify.txt"},
+				{action: "add", path: "to_rename.txt"},
+				{action: "rename", path: "to_rename.txt", newPath: "renamed.txt"},
+				{action: "add", path: "to_delete.txt"},
+				{action: "delete", path: "to_delete.txt"},
+			},
+			expected: map[string]struct{}{
+				"new.txt":       {},
+				"to_modify.txt": {},
+				"renamed.txt":   {},
+			},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			repo := setupTestRepo(t, tt.patches)
+
+			got, err := client.GetChangedFiles(ctx, repo, "feature", "main")
+			if (err != nil) != tt.wantErr {
+				t.Errorf("GetChangedFiles() error = %v, wantErr %v", err, tt.wantErr)
+				return
+			}
+			if err != nil {
+				t.Logf("Error details: %+v", err)
+			}
+			if diff := cmp.Diff(got, tt.expected); diff != "" {
+				t.Errorf("GetChangedFiles() mismatch (-want +got):\n%s", diff)
+			}
+		})
+	}
+}
+
+type gpatch struct {
+	action  string
+	path    string
+	newPath string
+}
+
+func setupTestRepo(t *testing.T, patches []gpatch) *git.Repository {
+	t.Helper()
+
+	dir := t.TempDir()
+	fs := osfs.New(dir)
+	repo, err := git.Init(memory.NewStorage(), fs)
+	if err != nil {
+		t.Fatalf("failed to init git repo: %v", err)
+	}
+
+	w, err := repo.Worktree()
+	if err != nil {
+		t.Fatalf("failed to get worktree: %v", err)
+	}
+
+	// Create and commit an initial file on main
+	if err := writeFile(w, "initial.txt", "initial content"); err != nil {
+		t.Fatalf("failed to create initial file: %v", err)
+	}
+	if _, err := w.Add("initial.txt"); err != nil {
+		t.Fatalf("failed to add initial file: %v", err)
+	}
+	if _, err := w.Commit("Initial commit", &git.CommitOptions{
+		Author: &object.Signature{Name: "Test Author", Email: "[email protected]"},
+	}); err != nil {
+		t.Fatalf("failed to commit initial file: %v", err)
+	}
+
+	// Create main branch
+	mainRef := plumbing.NewBranchReferenceName("main")
+	headRef, err := repo.Head()
+	if err != nil {
+		t.Fatalf("failed to get HEAD reference: %v", err)
+	}
+	if err := repo.Storer.SetReference(plumbing.NewHashReference(mainRef, headRef.Hash())); err != nil {
+		t.Fatalf("failed to set main branch reference: %v", err)
+	}
+
+	// Create and checkout feature branch
+	if err := w.Checkout(&git.CheckoutOptions{
+		Create: true,
+		Branch: plumbing.NewBranchReferenceName("feature"),
+	}); err != nil {
+		t.Fatalf("failed to checkout feature branch: %v", err)
+	}
+
+	// Apply patches
+	for _, p := range patches {
+		if err := applyPatch(w, p); err != nil {
+			t.Fatalf("failed to apply patch: %v", err)
+		}
+	}
+
+	// Commit changes
+	if _, err := w.Add("."); err != nil {
+		t.Fatalf("failed to add changes: %v", err)
+	}
+	if _, err := w.Commit("Test changes", &git.CommitOptions{
+		Author: &object.Signature{Name: "Test Author", Email: "[email protected]"},
+	}); err != nil {
+		t.Fatalf("failed to commit changes: %v", err)
+	}
+
+	return repo
+}
+
+func applyPatch(w *git.Worktree, p gpatch) error {
+	switch p.action {
+	case "add":
+		return writeFile(w, p.path, "content")
+	case "modify":
+		return writeFile(w, p.path, "modified content")
+	case "rename":
+		if err := w.Filesystem.Rename(p.path, p.newPath); err != nil {
+			return fmt.Errorf("failed to rename file: %w", err)
+		}
+	case "delete":
+		if err := w.Filesystem.Remove(p.path); err != nil {
+			return fmt.Errorf("failed to delete file: %w", err)
+		}
+	}
+	return nil
+}
+
+func writeFile(w *git.Worktree, path, content string) error {
+	f, err := w.Filesystem.Create(path)
+	if err != nil {
+		return fmt.Errorf("failed to create file %s: %w", path, err)
+	}
+	defer f.Close()
+	if _, err := f.Write([]byte(content)); err != nil {
+		return fmt.Errorf("failed to write to file %s: %w", path, err)
+	}
+	return nil
+}