From 57385506c1eb0c40e48756f4dbe0a357fe8d1291 Mon Sep 17 00:00:00 2001
From: Yiorgis Gozadinos <ggozadinos@gmail.com>
Date: Thu, 23 Apr 2026 15:29:10 +0300
Subject: [PATCH] Fix is_new_db detection when create=True is passed on an
 existing DB

---
 haiku_rag_slim/haiku/rag/store/engine.py | 16 +++++++---------
 tests/store/test_migrations.py           | 11 +++++++++++
 2 files changed, 18 insertions(+), 9 deletions(-)

diff --git a/haiku_rag_slim/haiku/rag/store/engine.py b/haiku_rag_slim/haiku/rag/store/engine.py
index 0c3c3907..08f3279a 100644
--- a/haiku_rag_slim/haiku/rag/store/engine.py
+++ b/haiku_rag_slim/haiku/rag/store/engine.py
@@ -188,6 +188,7 @@ class Store:
         self._skip_validation = skip_validation
         self._skip_migration_check = skip_migration_check
         self._vacuum_lock = asyncio.Lock()
+        self._is_new_db = False
 
         # Check if database exists (for local filesystem only)
         if self._connection_mode == ConnectionMode.LOCAL:
@@ -197,6 +198,7 @@ class Store:
                         f"Database does not exist at {self.db_path.absolute()}. "
                         "Use 'haiku-rag init' to create a new database."
                     )
+                self._is_new_db = True
                 # Ensure parent directories exist for new databases
                 if not db_path.parent.exists():
                     Path.mkdir(db_path.parent, parents=True)
@@ -211,15 +213,11 @@ class Store:
             self._config, self.db_path
         )
 
-        # Detect new vs existing database
-        is_new_db = False
-        if self._connection_mode == ConnectionMode.LOCAL:
-            is_new_db = not self.db_path.exists() or self._create
-            # Re-check after connect: if path didn't exist before, it's new
-            existing_tables = (await self.db.list_tables()).tables
-            if not existing_tables:
-                is_new_db = True
-        else:
+        # For remote stores (and as a safety net for local paths that exist but
+        # have no tables — e.g. a previously failed init), detect new DB by
+        # checking whether any tables exist.
+        is_new_db = self._is_new_db
+        if not is_new_db:
             existing_tables = (await self.db.list_tables()).tables
             if not existing_tables:
                 is_new_db = True
diff --git a/tests/store/test_migrations.py b/tests/store/test_migrations.py
index 85c7104b..ff8f0f94 100644
--- a/tests/store/test_migrations.py
+++ b/tests/store/test_migrations.py
@@ -67,6 +67,17 @@ class TestMigrationCheck:
                 pass
         assert "migrate" in str(exc_info.value).lower()
 
+    @pytest.mark.asyncio
+    async def test_pending_migrations_raises_error_with_create_flag(self, temp_db_path):
+        """Opening an existing DB with create=True must still check migrations."""
+        async with Store(temp_db_path, create=True) as store:
+            await store.set_haiku_version("0.19.0")
+
+        # create=True is idempotent — must not mark an existing populated DB as new
+        with pytest.raises(MigrationRequiredError):
+            async with Store(temp_db_path, create=True) as store:
+                pass
+
     @pytest.mark.asyncio
     async def test_pending_migrations_read_only_raises_error(self, temp_db_path):
         """Read-only mode with pending migrations should still raise."""