From dd405054fe4cc9eb87c5750fb94d3ba7215be836 Mon Sep 17 00:00:00 2001
From: sneak <sneak@sneak.berlin>
Date: Mon, 21 Sep 2026 22:49:12 +0000
Subject: [PATCH] Refuse an unversioned database that already has a files table
 (closes #11)

An unversioned database (user_version 0) that already contains a files
table was not created by this build; it is a foreign or partially
initialized file. Adopting it silently could corrupt unrelated data, so
createSchema now checks for a files table first and, when one exists,
returns the schema-version error telling the operator to remove the file
and rescan. A genuinely empty database is still created and stamped as
before.

Model: opus-4-8
---
 TODO.md    |  4 ++++
 db.go      | 25 +++++++++++++++++++++++--
 db_test.go | 32 ++++++++++++++++++++++++++++++++
 3 files changed, 59 insertions(+), 2 deletions(-)

diff --git a/TODO.md b/TODO.md
index 2044a7d..fb35c9c 100644
--- a/TODO.md
+++ b/TODO.md
@@ -29,6 +29,10 @@
 
 # Completed Steps
 
+- refuse an unversioned database that already has a `files` table with
+  a clear schema-version error (2026-09-21, closes
+  https://git.eeqj.de/sneak/sfdupes/issues/11)
+
 - remove the dead `files.dat` references from `Makefile`, `.gitignore`
   and `.dockerignore` (2026-09-21, branch `next`, closes
   https://git.eeqj.de/sneak/sfdupes/issues/22)
diff --git a/db.go b/db.go
index 573d944..a79c660 100644
--- a/db.go
+++ b/db.go
@@ -174,9 +174,30 @@ func initSchema(ctx context.Context, db *sql.DB) error {
 }
 
 // createSchema applies the schema to a fresh database and stamps the
-// schema version.
+// schema version. A database with user_version 0 that already has a
+// files table was not created by this build — a foreign or partially
+// initialized file. Adopting it silently could corrupt unrelated data,
+// so that is a fatal schema-version error telling the operator to
+// remove the file and rescan.
 func createSchema(ctx context.Context, db *sql.DB) error {
-	_, err := db.ExecContext(ctx, createTableSQL)
+	var name string
+
+	err := db.QueryRowContext(ctx,
+		"SELECT name FROM sqlite_master "+
+			"WHERE type = 'table' AND name = 'files'").Scan(&name)
+
+	switch {
+	case err == nil:
+		return fmt.Errorf(
+			"has a files table but no schema version; "+
+				"remove the file and rescan: %w", errSchemaVersion)
+	case errors.Is(err, sql.ErrNoRows):
+		// Genuinely empty: create the schema below.
+	default:
+		return fmt.Errorf("check for files table: %w", err)
+	}
+
+	_, err = db.ExecContext(ctx, createTableSQL)
 	if err != nil {
 		return fmt.Errorf("create schema: %w", err)
 	}
diff --git a/db_test.go b/db_test.go
index 41ab6f2..29f7e26 100644
--- a/db_test.go
+++ b/db_test.go
@@ -78,6 +78,38 @@ func TestOpenScanDatabaseCreates(t *testing.T) {
 	}
 }
 
+func TestOpenScanDatabaseUnversionedForeign(t *testing.T) {
+	t.Parallel()
+
+	path := testDBPath(t)
+
+	// A database that has a files table but user_version 0 — a foreign
+	// or partially initialized file. scan must refuse it with a clear
+	// schema-version error, not adopt it and not emit a raw SQLite
+	// "table files already exists".
+	db, err := openDB(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	_, err = db.ExecContext(t.Context(), "CREATE TABLE files (x INTEGER)")
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	_ = db.Close()
+
+	_, err = openScanDatabase(t.Context(), path)
+	if !errors.Is(err, errSchemaVersion) {
+		t.Fatalf("err = %v, want errSchemaVersion", err)
+	}
+
+	if !strings.Contains(err.Error(), "remove the file and rescan") {
+		t.Fatalf("err = %v, want it to tell the operator to remove and rescan",
+			err)
+	}
+}
+
 func TestOpenReportDatabaseMissing(t *testing.T) {
 	t.Parallel()