From b5098079d84a2472d4d8552d8c7867c39b46da88 Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal+gh@gmail.com>
Date: Thu, 29 Mar 2018 00:08:16 +0200
Subject: [PATCH] Fix error on urllib

---
 spacy/cli/download.py |  8 ++++----
 spacy/cli/validate.py |  8 ++++----
 spacy/compat.py       | 18 +++++++++++++-----
 3 files changed, 21 insertions(+), 13 deletions(-)

diff --git a/spacy/cli/download.py b/spacy/cli/download.py
index 977545dfe..71c1a65dd 100644
--- a/spacy/cli/download.py
+++ b/spacy/cli/download.py
@@ -9,7 +9,7 @@ import ujson
 
 from .link import link
 from ..util import prints, get_package_path
-from ..compat import url_open, url_error
+from ..compat import url_read, HTTPError
 from .. import about
 
 
@@ -58,13 +58,13 @@ def download(model, direct=False):
 
 def get_json(url, desc):
     try:
-        r = url_open(url)
-    except url_error as e:
+        data = url_read(url)
+    except HTTPError as e:
         msg = ("Couldn't fetch %s. Please find a model for your spaCy "
                "installation (v%s), and download it manually.")
         prints(msg % (desc, about.__version__), about.__docs_models__,
                title="Server error (%d: %s)" % (e.code, e.reason), exits=1)
-    return ujson.load(r)
+    return ujson.loads(data)
 
 
 def get_compatibility():
diff --git a/spacy/cli/validate.py b/spacy/cli/validate.py
index 2b7388988..c140f6bff 100644
--- a/spacy/cli/validate.py
+++ b/spacy/cli/validate.py
@@ -6,7 +6,7 @@ from pathlib import Path
 import sys
 import ujson
 
-from ..compat import path2str, locale_escape, url_open, url_error
+from ..compat import path2str, locale_escape, url_read, HTTPError
 from ..util import prints, get_data_path, read_json
 from .. import about
 
@@ -16,11 +16,11 @@ def validate():
     with the installed models. Should be run after `pip install -U spacy`.
     """
     try:
-        r = url_open(about.__compatibility__)
-    except url_error as e:
+        data = url_read(about.__compatibility__)
+    except HTTPError as e:
         prints("Couldn't fetch compatibility table.",
                title="Server error (%d: %s)" % (e.code, e.reason), exits=1)
-    compat = ujson.load(r)['spacy']
+    compat = ujson.loads(data)['spacy']
     current_compat = compat.get(about.__version__)
     if not current_compat:
         prints(about.__compatibility__, exits=1,
diff --git a/spacy/compat.py b/spacy/compat.py
index 5359edefc..d6260a850 100644
--- a/spacy/compat.py
+++ b/spacy/compat.py
@@ -39,9 +39,9 @@ except ImportError:
     import urllib2 as urllib
 
 try:
-    from urllib.error import HTTPError as url_error
+    from urllib.error import HTTPError
 except ImportError:
-    from urllib2 import HTTPError as url_error
+    from urllib2 import HTTPError
 
 pickle = pickle
 copy_reg = copy_reg
@@ -49,7 +49,6 @@ CudaStream = CudaStream
 cupy = cupy
 copy_array = copy_array
 urllib = urllib
-url_error = url_error
 izip = getattr(itertools, 'izip', zip)
 
 is_windows = sys.platform.startswith('win')
@@ -68,7 +67,7 @@ if is_python2:
     input_ = raw_input  # noqa: F821
     json_dumps = lambda data: ujson.dumps(data, indent=2, escape_forward_slashes=False).decode('utf8')
     path2str = lambda path: str(path).decode('utf8')
-    url_open = lambda url: urllib.urlopen(url)
+    url_open = urllib.urlopen
 
 elif is_python3:
     bytes_ = bytes
@@ -77,7 +76,16 @@ elif is_python3:
     input_ = input
     json_dumps = lambda data: ujson.dumps(data, indent=2, escape_forward_slashes=False)
     path2str = lambda path: str(path)
-    url_open = lambda url: urllib.request.urlopen(url)
+    url_open = urllib.request.urlopen
+
+
+def url_read(url):
+    file_ = url_open(url)
+    code = file_.getcode()
+    if code != 200:
+        raise HTTPError(url, code, "Cannot GET url", [], file_)  
+    data = file_.read()
+    return data
 
 
 def b_to_str(b_str):