Fix test imports and last batch cleanup

spacy/language.py

@@ -560,7 +560,11 @@ class Language(object):
             elif len(old_refs) == 0:
                 self.vocab.strings._cleanup_stale_strings()
                 nr_seen = 0
-        self.vocab.strings._reset_and_load(original_strings_data)
+        # Last batch can be not garbage collected and we cannot know it — last
+        # doc still here. Not erase that strings — just extend with original
+        # content
+        for string in original_strings_data:
+            self.vocab.strings.add(string)
 
     def to_disk(self, path, disable=tuple()):
         """Save the current state to a directory.  If a model is loaded, this

spacy/tests/regression/test_issue1506.py

@@ -1,12 +1,6 @@
 # coding: utf8
 from __future__ import unicode_literals
 
-import random
-import string
-
-import itertools
-from compat import izip
-
 from ...lang.en import English
 
 
@@ -14,14 +8,14 @@ def test_issue1506():
     nlp = English()
 
     def string_generator():
-        for (_, t) in izip(range(10001), itertools.repeat("It's sentence produced by that bug.")):
-            yield t
+        for _ in range(10001):
+            yield "It's sentence produced by that bug."
 
-        for (_, t) in izip(range(10001), itertools.repeat("I erase lemmas.")):
-            yield t
+        for _ in range(10001):
+            yield "I erase lemmas."
 
-        for (_, t) in izip(range(10001), itertools.repeat("It's sentence produced by that bug.")):
-            yield t
+        for _ in range(10001):
+            yield "It's sentence produced by that bug."
 
     for d in nlp.pipe(string_generator()):
         for t in d: