Store IVFSparse deletes as int (#1241)

src/python/txtai/ann/sparse/ivfsparse.py

@@ -116,10 +116,11 @@ class IVFSparse(ANN):
 
     def delete(self, ids):
         # Set index ids as deleted, ignoring ids that were never indexed and ids already marked deleted
+        # Ids are stored as int, the deletes list is serialized with msgpack
         size = self.size()
         for x in ids:
             if x < size and x not in self.deletes:
-                self.deletes.append(x)
+                self.deletes.append(int(x))
 
     def search(self, queries, limit):
         results = []

test/python/testann/testsparse.py

@@ -8,6 +8,8 @@ import unittest
 
 from unittest.mock import patch
 
+import numpy as np
+
 from scipy.sparse import random
 from sklearn.preprocessing import normalize
 
@@ -100,6 +102,31 @@ class TestSparse(unittest.TestCase):
         self.assertLessEqual(len(ann.blocks), 15)
         ann.close()
 
+    def testIVFSparseDeleteArray(self):
+        """
+        Test IVFSparse with ids deleted using a NumPy array
+        """
+
+        # Generate test record
+        data = self.generate(50, 30522)
+
+        # Create ANN
+        path = os.path.join(tempfile.gettempdir(), "ivfsparse.deletes")
+        ann = SparseANNFactory.create({"backend": "ivfsparse"})
+        ann.index(data)
+
+        # Test delete with ids passed in as a NumPy array
+        ann.delete(np.array([0, 1]))
+        self.assertEqual(ann.count(), 48)
+
+        # Validate save/load
+        ann.save(path)
+        ann.load(path)
+        self.assertEqual(ann.count(), 48)
+
+        # Close ANN
+        ann.close()
+
     def testIVFSparseSortOrder(self):
         """
         Test IVFSparse returns results sorted by score descending