From d6261de1e721d6f6e1a5dd9a17e4160d6bd5804b Mon Sep 17 00:00:00 2001
From: Claude <noreply@anthropic.com>
Date: Mon, 20 Jul 2026 02:39:07 +0000
Subject: [PATCH] Re-clone bootstrap.sh's target directory when the existing
 copy is broken
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If an existing ~/ubuntu-post-install checkout has a broken/SSH-only
origin remote, `git pull --ff-only` fails and the script fell through
to "continuing with existing version" — even when that existing copy
is missing setup.sh entirely, guaranteeing a crash right after. Now
checks for setup.sh post-pull and wipes + re-clones over HTTPS (no SSH
key needed) if it's still missing.
---
 bootstrap.sh | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/bootstrap.sh b/bootstrap.sh
index 0d0f7b3..d12e6e0 100755
--- a/bootstrap.sh
+++ b/bootstrap.sh
@@ -87,6 +87,21 @@ else
     fi
 fi
 
+# The pull-failed fallback above assumes the existing checkout is at least
+# usable — it isn't always (e.g. a stale/broken .git dir, an origin remote
+# pointed at SSH with no key available under sudo, an interrupted first
+# clone). Rather than hand the user a guaranteed "setup.sh: No such file or
+# directory" crash, treat a missing setup.sh as "this copy is unusable" and
+# re-clone fresh over HTTPS (which doesn't need any SSH key at all).
+if [ ! -f "$DEST/setup.sh" ]; then
+    echo "  Existing copy at $DEST looks incomplete or broken — re-cloning fresh..."
+    rm -rf "$DEST"
+    git clone "$CLONE_URL" "$DEST"
+    if [ -n "$PAT" ]; then
+        git -C "$DEST" remote set-url origin "$REPO_URL"
+    fi
+fi
+
 echo ""
 echo "  Launching setup..."
 echo ""