Add compat functions and remove old workarounds

Add ensure_path util function to handle checking instance of path

spacy/cli/link.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import pip
 from pathlib import Path
 import importlib
+from ..compat import unicode_, symlink_to
 from .. import util
 
 
@@ -43,23 +44,17 @@ def symlink(model_path, link_name, force):
     elif link_path.exists():
         link_path.unlink()
 
-    # Add workaround for Python 2 on Windows (see issue #909)
-    if util.is_python2() and util.is_windows():
-        import subprocess
-        command = ['mklink', '/d', unicode(link_path), unicode(model_path)]
     try:
-            subprocess.call(command, shell=True)
+        symlink_to(link_path, model_path)
     except:
-            # This is quite dirty, but just making sure other Windows-specific
+        # This is quite dirty, but just making sure other errors are caught so
-            # errors are caught so users at least see a proper error message.
+        # users at least see a proper message.
         util.sys_exit(
             "Creating a symlink in spacy/data failed. You can still import "
             "the model as a Python package and call its load() method, or "
             "create the symlink manually:",
-                "{a} --> {b}".format(a=unicode(model_path), b=unicode(link_path)),
+            "{a} --> {b}".format(a=unicode_(model_path), b=unicode_(link_path)),
             title="Error: Couldn't link model to '{l}'".format(l=link_name))
-    else:
-        link_path.symlink_to(model_path)
 
     util.print_msg(
         "{a} --> {b}".format(a=model_path.as_posix(), b=link_path.as_posix()),

spacy/cli/package.py

@@ -1,20 +1,13 @@
 # coding: utf8
 from __future__ import unicode_literals
 
-import json
 import shutil
 import requests
 from pathlib import Path
 
-import six
+from ..compat import unicode_, json_dumps
-
-from .. import about
 from .. import util
 
-if six.PY2:
-    json_dumps = lambda data: json.dumps(data, indent=2).decode("utf8")
-elif six.PY3:
-    json_dumps = lambda data: json.dumps(data, indent=2)
 
 def package(input_dir, output_dir, force):
     input_path = Path(input_dir)
@@ -32,31 +25,31 @@ def package(input_dir, output_dir, force):
     package_path = main_path / model_name
 
     create_dirs(package_path, force)
-    shutil.copytree(input_path.as_posix(), (package_path / model_name_v).as_posix())
+    shutil.copytree(unicode_(input_path), unicode_(package_path / model_name_v))
     create_file(main_path / 'meta.json', json_dumps(meta))
     create_file(main_path / 'setup.py', template_setup)
     create_file(main_path / 'MANIFEST.in', template_manifest)
     create_file(package_path / '__init__.py', template_init)
 
     util.print_msg(
-        main_path.as_posix(),
+        unicode_(main_path),
         "To build the package, run `python setup.py sdist` in that directory.",
         title="Successfully created package {p}".format(p=model_name_v))
 
 
 def check_dirs(input_path, output_path):
     if not input_path.exists():
-        util.sys_exit(input_path.as_poisx(), title="Model directory not found")
+        util.sys_exit(unicode_(input_path.as_poisx), title="Model directory not found")
     if not output_path.exists():
-        util.sys_exit(output_path.as_posix(), title="Output directory not found")
+        util.sys_exit(unicode_(output_path), title="Output directory not found")
 
 
 def create_dirs(package_path, force):
     if package_path.exists():
         if force:
-            shutil.rmtree(package_path.as_posix())
+            shutil.rmtree(unicode_(package_path.as_posix))
         else:
-            util.sys_exit(package_path.as_posix(),
+            util.sys_exit(unicode_(package_path.as_posix),
                 "Please delete the directory and try again.",
                 title="Package directory already exists")
     Path.mkdir(package_path, parents=True)

spacy/deprecated.py

@@ -6,12 +6,6 @@ from .cli import download
 from .cli import link
 
 
-try:
-    basestring
-except NameError:
-    basestring = str
-
-
 def read_lang_data(package):
     tokenization = package.load_json(('tokenizer', 'specials.json'))
     with package.open(('tokenizer', 'prefix.txt'), default=None) as file_:
@@ -73,9 +67,7 @@ def fix_glove_vectors_loading(overrides):
     if overrides.get('path') in (None, True):
         data_path = util.get_data_path()
     else:
-        path = overrides['path']
+        path = util.ensure_path(overrides['path'])
-        if isinstance(path, basestring):
-            path = Path(path)
         data_path = path.parent
     vec_path = None
     if 'add_vectors' not in overrides:

spacy/language.py

@@ -4,17 +4,6 @@ from contextlib import contextmanager
 import shutil
 import ujson
 
-
-try:
-    basestring
-except NameError:
-    basestring = str
-
-try:
-    unicode
-except NameError:
-    unicode = str
-
 from .tokenizer import Tokenizer
 from .vocab import Vocab
 from .tagger import Tagger
@@ -26,6 +15,7 @@ from .syntax.nonproj import PseudoProjectivity
 from .pipeline import DependencyParser, EntityRecognizer
 from .syntax.arc_eager import ArcEager
 from .syntax.ner import BiluoPushDown
+from .compat import unicode_
 from .attrs import IS_STOP
 from . import attrs
 from . import orth
@@ -205,7 +195,7 @@ class Language(object):
             directory.mkdir()
             with (directory / 'config.json').open('wb') as file_:
                 data = ujson.dumps(config, indent=2)
-                if isinstance(data, unicode):
+                if isinstance(data, unicode_):
                     data = data.encode('utf8')
                 file_.write(data)
         if not (path / 'vocab').exists():
@@ -252,9 +242,7 @@ class Language(object):
     def __init__(self, **overrides):
         if 'data_dir' in overrides and 'path' not in overrides:
             raise ValueError("The argument 'data_dir' has been renamed to 'path'")
-        path = overrides.get('path', True)
+        path = util.ensure_path(overrides.get('path', True))
-        if isinstance(path, basestring):
-            path = pathlib.Path(path)
         if path is True:
             path = util.get_data_path() / self.lang
             if not path.exists() and 'path' not in overrides:

spacy/tagger.pyx

@@ -15,6 +15,7 @@ from .tokens.doc cimport Doc
 from .attrs cimport TAG
 from .gold cimport GoldParse
 from .attrs cimport *
+from . import util
 
 
 cpdef enum:
@@ -127,7 +128,7 @@ cdef class Tagger:
         """
         # TODO: Change this to expect config.json when we don't have to
         # support old data.
-        path = path if not isinstance(path, basestring) else pathlib.Path(path)
+        path = util.ensure_path(path)
         if (path / 'templates.json').exists():
             with (path / 'templates.json').open('r', encoding='utf8') as file_:
                 templates = json.load(file_)

spacy/tokenizer.pyx

@@ -48,10 +48,8 @@ cdef class Tokenizer:
             infix_finditer:
                 Signature of re.compile(string).finditer
         Returns Tokenizer
-        if isinstance(path, basestring):
-            path = pathlib.Path(path)
-
         """
+        path = util.ensure_path(path)
         if rules is None:
             with (path / 'tokenizer' / 'specials.json').open('r', encoding='utf8') as file_:
                 rules = json.load(file_)

spacy/util.py

@@ -8,17 +8,7 @@ from pathlib import Path
 import sys
 import textwrap
 
-
+from .compat import basestring_, unicode_, input_
-try:
-    basestring
-except NameError:
-    basestring = str
-
-
-try:
-    raw_input
-except NameError: # Python 3
-    raw_input = input
 
 
 LANGUAGES = {}
@@ -46,9 +36,14 @@ def get_data_path(require_exists=True):
 
 def set_data_path(path):
     global _data_path
-    if isinstance(path, basestring):
+    _data_path = ensure_path(path)
-        path = pathlib.Path(path)
+
-    _data_path = path
+
+def ensure_path(path):
+    if isinstance(path, basestring_):
+        return Path(path)
+    else:
+        return path
 
 
 def or_(val1, val2):
@@ -94,7 +89,7 @@ def constraint_match(constraint_string, version):
 
 
 def read_regex(path):
-    path = path if not isinstance(path, basestring) else pathlib.Path(path)
+    path = ensure_path(path)
     with path.open() as file_:
         entries = file_.read().split('\n')
     expression = '|'.join(['^' + re.escape(piece) for piece in entries if piece.strip()])
@@ -151,16 +146,6 @@ def check_renamed_kwargs(renamed, kwargs):
             raise TypeError("Keyword argument %s now renamed to %s" % (old, new))
 
 
-def is_windows():
-    """Check if user is on Windows."""
-    return sys.platform.startswith('win')
-
-
-def is_python2():
-    """Check if Python 2 is used."""
-    return sys.version.startswith('2.')
-
-
 def parse_package_meta(package_path, package, require=True):
     location = package_path / package / 'meta.json'
     if location.is_file():
@@ -180,7 +165,7 @@ def get_raw_input(description, default=False):
 
     additional = ' (default: {d})'.format(d=default) if default else ''
     prompt = '    {d}{a}: '.format(d=description, a=additional)
-    user_input = raw_input(prompt)
+    user_input = input_(prompt)
     return user_input
 
 

spacy/vocab.pyx

@@ -5,11 +5,6 @@ import bz2
 import ujson as json
 import re
 
-try:
-    import cPickle as pickle
-except ImportError:
-    import pickle
-
 from libc.string cimport memset
 from libc.stdint cimport int32_t
 from libc.math cimport sqrt
@@ -23,10 +18,7 @@ from .tokens.token cimport Token
 from .serialize.packer cimport Packer
 from .attrs cimport PROB, LANG
 
-try:
+from .compat import copy_reg, pickle
-    import copy_reg
-except ImportError:
-    import copyreg as copy_reg
 from .lemmatizer import Lemmatizer
 from .attrs import intify_attrs
 from . import util
@@ -69,8 +61,7 @@ cdef class Vocab:
         Returns:
             Vocab: The newly constructed vocab object.
         """
-        if isinstance(path, basestring):
+        path = util.ensure_path(path)
-            path = Path(path)
         util.check_renamed_kwargs({'get_lex_attr': 'lex_attr_getters'}, deprecated_kwargs)
         if 'vectors' in deprecated_kwargs:
             raise AttributeError(