diff --git a/.gitignore b/.gitignore
index dabdbca..8f5ae89 100644
--- a/.gitignore
+++ b/.gitignore
@@ -4,4 +4,4 @@ __pycache__/
 *.py[o|c]
 **_build/
 geckodriver.log
-sql_app.db
+*.sqlite3
diff --git a/Makefile b/Makefile
index 403e1e2..80ea42f 100644
--- a/Makefile
+++ b/Makefile
@@ -1,5 +1,7 @@
 coverage_opt=--cov  --cov-report=term-missing:skip-covered --durations=10
 
+target=dev
+
 clean:
 	pipenv --rm
 
@@ -48,12 +50,25 @@ venv_install:
 venv_test:
 	make -f MakefileVenv test
 
-db_clean:
-	rm  sql_app.db && touch sql_app.db
+db_clean: db_delete
+	touch db_$(target).sqlite3
 
 db_fill:
 	pipenv run python database.py
 
 
+db_delete:
+	([ -f db_$(target).sqlite3 ] && rm db_$(target).sqlite3) || echo "No file db_$(target).sqlite3 to drop"
+
+
+db_save:
+	cp -f db_$(target).sqlite3 backup_$(USER)_$(target).sqlite3
+
+db_restore:
+	cp  backup_$(USER)_$(target).sqlite3 db_$(target).sqlite3
+
+
+
+
 db_reset: db_clean db_fill
 
diff --git a/database.py b/database.py
index d93f256..a1fc403 100644
--- a/database.py
+++ b/database.py
@@ -2,7 +2,7 @@ from sqlalchemy import create_engine
 from sqlalchemy.ext.declarative import declarative_base
 from sqlalchemy.orm import sessionmaker
 
-SQLALCHEMY_DATABASE_URL = "sqlite:///./sql_app.db"
+SQLALCHEMY_DATABASE_URL = "sqlite:///./db_dev.sqlite3"  # to be configure
 # SQLALCHEMY_DATABASE_URL = "postgresql://user:password@postgresserver/db"
 
 engine = create_engine(
@@ -46,11 +46,17 @@ def fill_db(
     page_size = 1_00
     with open(movie_input_file) as csvfile:
         for count, movie_data in enumerate(csv.DictReader(csvfile), start=1):
-            adapted_data = adapt_movie_data(movie_data)
-            crud.create_movie(db, batch_mode=True, **adapted_data)
             if count % page_size == 0:
                 db.commit()
 
+            adapted_data = adapt_movie_data(movie_data)
+
+            if not adapted_data["title"]:
+                print(count, "should be fixed")
+                print(adapted_data)
+                continue
+
+            crud.create_movie(db, batch_mode=True, **adapted_data)
         db.commit()
 
 
diff --git a/utests/test_sql_database.py b/utests/test_sql_database.py
index 6a1f474..46c5c42 100644
--- a/utests/test_sql_database.py
+++ b/utests/test_sql_database.py
@@ -143,4 +143,25 @@ def test_sample_import_toy_story():
 
     assert "Andy" in toy_story["description"]
 
-    # non regression
+
+def test_title_is_taken_form_original_title_is_missing():
+    """
+    t0113002,en,Midnight Man
+    19763 Midnight Man (among others)  has an unescaped \n that makes import fail
+
+    in the csv the movie 'Avalanche Sharks' @ line 35587
+    has no tiltle, we fix this here to get  quicker but we need a better solution
+    """
+    movie_title = "Midnight Man"
+
+    file_path = "utests/movie_error_missing_title.csv"
+    file_path = "input_data/movies_metadata.csv"
+    movies = client.get("movies")
+    movies_by_title = {m["title"]: m for m in movies.json()}
+
+    assert movie_title not in movies_by_title, "The movie should not be pre existing"
+    with db_context() as db:
+        fill_db(db, file_path)
+
+    movies = client.get("movies")
+    movies_by_title = {m["title"]: m for m in movies.json()}