From c396581a0b908ea2ff5244c01ce5d87153b7d729 Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <matthew@honnibal.com>
Date: Mon, 15 Sep 2014 06:34:45 +0200
Subject: [PATCH] * Fiddle with the way strings are interned in lexeme

---
 spacy/lexeme.pyx | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/spacy/lexeme.pyx b/spacy/lexeme.pyx
index 3c0a28c7f..223fc5e06 100644
--- a/spacy/lexeme.pyx
+++ b/spacy/lexeme.pyx
@@ -24,8 +24,8 @@ cdef int lexeme_free(LexemeC* lexeme) except -1:
     
 
 cdef char* intern_and_encode(unicode string, size_t* length):
-    global _strings
-    cdef bytes utf8_string = intern(string.encode('utf8'))
+    cdef bytes byte_string = string.encode('utf8')
+    cdef bytes utf8_string = intern(byte_string)
     Py_INCREF(utf8_string)
     length[0] = len(utf8_string)
     return <char*>utf8_string