From 9cca5cc41babbd4d6491f5253e09edb8ae8da256 Mon Sep 17 00:00:00 2001
From: jordan <jordan.washburn@gmail.com>
Date: Sat, 7 Feb 2026 07:59:53 -0700
Subject: [PATCH] fix: add proper instrumentation to git clone for debugging

- Log clone request with work_dir, URL, and token presence
- Log workspace state (is_git_repo, existing remote)
- Log all decision points (pull vs clone, clear workspace)
- Detect and clear non-empty non-git directories before clone
- Capture both stdout and stderr for clone failures
- Include exit code in error messages
---
 internal/claudebox/git.go | 55 ++++++++++++++++++++++++++++++++++++---
 1 file changed, 51 insertions(+), 4 deletions(-)

diff --git a/internal/claudebox/git.go b/internal/claudebox/git.go
index 772625b..f101561 100644
--- a/internal/claudebox/git.go
+++ b/internal/claudebox/git.go
@@ -68,50 +68,97 @@ type CloneResult struct {
 func (g *GitOperations) CloneRepo(ctx context.Context, workDir, cloneURL string) *CloneResult {
 	result := &CloneResult{}
 
+	g.logger.Info("git clone request",
+		"work_dir", workDir,
+		"clone_url", g.redactToken(cloneURL),
+		"has_token", g.giteaToken != "")
+
 	if cloneURL == "" {
 		result.Error = fmt.Errorf("git clone URL is required")
+		g.logger.Error("git clone failed: empty URL")
 		return result
 	}
 
 	// Check if already a git repo with the correct remote
-	if g.isGitRepo(ctx, workDir) {
+	isRepo := g.isGitRepo(ctx, workDir)
+	g.logger.Info("workspace check", "is_git_repo", isRepo, "work_dir", workDir)
+
+	if isRepo {
 		currentRemote, err := g.runGitOutput(ctx, workDir, "config", "--get", "remote.origin.url")
 		currentRemote = strings.TrimSpace(currentRemote)
+		g.logger.Info("existing repo detected",
+			"current_remote", g.redactToken(currentRemote),
+			"expected_remote", g.redactToken(cloneURL),
+			"remote_match", currentRemote == cloneURL,
+			"remote_err", err)
 
 		if err == nil && currentRemote == cloneURL {
 			// Pull latest changes
+			g.logger.Info("pulling latest changes", "work_dir", workDir)
 			if err := g.runGit(ctx, workDir, "pull", "--ff-only"); err != nil {
 				// Pull failed but repo exists - continue with existing state
-				g.logger.Debug("git pull failed, continuing with existing state", "error", err, "work_dir", workDir)
+				g.logger.Warn("git pull failed, continuing with existing state",
+					"error", err,
+					"work_dir", workDir)
 			}
+			g.logger.Info("git clone complete (existing repo)", "cloned", false)
 			return result
 		}
 
 		// Different remote - clear and re-clone
+		g.logger.Info("clearing workspace for re-clone", "work_dir", workDir)
 		if err := g.clearDir(ctx, workDir); err != nil {
 			result.Error = fmt.Errorf("clear workspace: %w", err)
+			g.logger.Error("failed to clear workspace", "error", result.Error)
 			return result
 		}
 	}
 
+	// Check if directory exists and is non-empty (would cause clone to fail)
+	if !isRepo {
+		checkCmd := exec.CommandContext(ctx, "sh", "-c", fmt.Sprintf("ls -A %s 2>/dev/null | head -1", workDir))
+		output, _ := checkCmd.Output()
+		if len(strings.TrimSpace(string(output))) > 0 {
+			g.logger.Warn("workspace not empty but not a git repo, clearing",
+				"work_dir", workDir,
+				"first_file", strings.TrimSpace(string(output)))
+			if err := g.clearDir(ctx, workDir); err != nil {
+				result.Error = fmt.Errorf("clear non-empty workspace: %w", err)
+				g.logger.Error("failed to clear non-empty workspace", "error", result.Error)
+				return result
+			}
+		}
+	}
+
 	// Inject token for authentication
 	authCloneURL := cloneURL
 	if g.giteaToken != "" {
 		authCloneURL = strings.Replace(cloneURL, "https://", "https://token:"+g.giteaToken+"@", 1)
+	} else {
+		g.logger.Warn("no gitea token configured, clone may fail for private repos")
 	}
 
 	// Clone the repository
+	g.logger.Info("executing git clone", "work_dir", workDir)
 	cmd := exec.CommandContext(ctx, "git", "clone", authCloneURL, workDir)
-	var stderr bytes.Buffer
+	var stdout, stderr bytes.Buffer
+	cmd.Stdout = &stdout
 	cmd.Stderr = &stderr
 
 	if err := cmd.Run(); err != nil {
 		errMsg := g.redactToken(stderr.String())
-		result.Error = fmt.Errorf("git clone failed: %s: %s", err, errMsg)
+		stdoutMsg := g.redactToken(stdout.String())
+		result.Error = fmt.Errorf("git clone exit %v: %s", err, errMsg)
+		g.logger.Error("git clone failed",
+			"error", err,
+			"stderr", errMsg,
+			"stdout", stdoutMsg,
+			"work_dir", workDir)
 		return result
 	}
 
 	result.Cloned = true
+	g.logger.Info("git clone complete", "cloned", true, "work_dir", workDir)
 	return result
 }