Move fix_text to spacy.compat (see #1002)

spacy/cli/model.py

@@ -6,9 +6,9 @@ import math
 from ast import literal_eval
 from pathlib import Path
 from preshed.counter import PreshCounter
-import ftfy
 
 from ..vocab import write_binary_vectors
+from ..compat import fix_text
 from .. import util
 
 
@@ -77,7 +77,7 @@ def read_clusters(clusters_path):
         for line in f:
             try:
                 cluster, word, freq = line.split()
-                word = ftfy.fix_text(word)
+                word = fix_text(word)
             except ValueError:
                 continue
             # If the clusterer has only seen the word a few times, its

spacy/compat.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 import six
+import ftfy
 import sys
 import ujson
 
@@ -38,6 +39,9 @@ elif is_python3:
     json_dumps = lambda data: ujson.dumps(data, indent=2)
 
 
+fix_text = lambda text: ftfy.fix_text(text)
+
+
 def symlink_to(orig, dest):
     if is_python2 and is_windows:
         import subprocess