Merge branch 'master' of ssh://github.com/honnibal/spaCy

setup.py

@@ -72,7 +72,7 @@ MOD_NAMES = [
 # which is really known only after finalize_options
 # http://stackoverflow.com/questions/724664/python-distutils-how-to-get-a-compiler-that-is-going-to-be-used
 compile_options =  {
-    'msvc': ['/Ox', '/EHsc', '/openmp'],
+    'msvc': ['/Ox', '/EHsc'],
     'mingw32' : ['-O3', '-Wno-strict-prototypes', '-Wno-unused-function'],
     'other' : ['-O3', '-Wno-strict-prototypes', '-Wno-unused-function']
 }
@@ -85,6 +85,10 @@ link_options = {
 }
 
 
+if os.environ.get('USE_OPENMP') == '1':
+    compile_options['msvc'].append('/openmp')
+
+
 if not sys.platform.startswith('darwin'):
     compile_options['other'].append('-fopenmp')
     link_options['other'].append('-fopenmp')
@@ -189,7 +193,6 @@ def setup_package():
                 'thinc>=5.0.0,<5.1.0',
                 'plac',
                 'six',
-                'ujson',
                 'cloudpickle',
                 'sputnik>=0.9.2,<0.10.0'],
             classifiers=[

spacy/__init__.py

@@ -1,11 +1,11 @@
 from .util import set_lang_class, get_lang_class, get_package, get_package_by_name
 
-from .en import English
-from .de import German
+from . import en
+from . import de
 
 
-set_lang_class(English.lang, English)
-set_lang_class(German.lang, German)
+set_lang_class(en.English.lang, en.English)
+set_lang_class(de.German.lang, de.German)
 
 
 def load(name, vectors=None, via=None):

spacy/about.py

@@ -14,4 +14,3 @@ __models__ = {
     'en': 'en>=1.1.0,<1.2.0',
     'de': 'de>=1.0.0,<1.1.0',
 }
-__default_lang__ = 'en'

spacy/gold.pyx

@@ -1,7 +1,6 @@
 import numpy
 import io
 import json
-import ujson
 import random
 import re
 import os

spacy/language.py

@@ -20,6 +20,7 @@ from .syntax.ner import BiluoPushDown
 from .syntax.arc_eager import ArcEager
 
 from . import util
+from . import about
 from .attrs import TAG, DEP, ENT_IOB, ENT_TYPE, HEAD
 
 
@@ -166,7 +167,7 @@ class Language(object):
         """
         if package is None:
             if data_dir is None:
-                package = util.get_package_by_name()
+                package = util.get_package_by_name(about.__models__[self.lang])
             else:
                 package = util.get_package(data_dir)
 

spacy/strings.pyx

@@ -10,7 +10,10 @@ from preshed.maps cimport map_iter, key_t
 
 from .typedefs cimport hash_t
 
-import ujson as json
+try:
+    import ujson as json
+except ImportError:
+    import json
 
 
 cpdef hash_t hash_string(unicode string) except 0:

spacy/tests/serialize/test_packer.py

@@ -24,7 +24,7 @@ from spacy.serialize.bits import BitArray
 def vocab():
     data_dir = os.environ.get('SPACY_DATA')
     if data_dir is None:
-        package = util.get_package_by_name()
+        package = util.get_package_by_name('en')
     else:
         package = util.get_package(data_dir)
 

spacy/tests/tagger/test_lemmatizer.py

@@ -14,7 +14,7 @@ import pytest
 def package():
     data_dir = os.environ.get('SPACY_DATA')
     if data_dir is None:
-        return util.get_package_by_name()
+        return util.get_package_by_name('en')
     else:
         return util.get_package(data_dir)
 

spacy/util.py

@@ -36,11 +36,12 @@ def get_package(data_dir):
 
 
 def get_package_by_name(name=None, via=None):
-    package_name = name or about.__models__[about.__default_lang__]
-    lang = get_lang_class(package_name)
+    if name is None:
+        return
+    lang = get_lang_class(name)
     try:
         return sputnik.package(about.__title__, about.__version__,
-            package_name, data_path=via)
+            name, data_path=via)
     except PackageNotFoundException as e:
         raise RuntimeError("Model '%s' not installed. Please run 'python -m "
                            "%s.download' to install latest compatible "