From 6076213c167640c5f2c4cf24ec2cd354b7a8bda0 Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal@gmail.com>
Date: Sat, 25 Jul 2015 22:35:52 +0200
Subject: [PATCH 1/4] * Fix init_model script

---
 bin/init_model.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/bin/init_model.py b/bin/init_model.py
index d40e7813d..2d3c62368 100644
--- a/bin/init_model.py
+++ b/bin/init_model.py
@@ -91,7 +91,7 @@ def _read_probs(loc):
 def _read_freqs(loc):
     counts = PreshCounter()
     total = 0
-    for line in open(loc):
+    for line in loc.open():
         freq, doc_freq, key = line.split('\t', 2)
         freq = int(freq)
         counts[hash_string(key)] = freq
@@ -99,7 +99,7 @@ def _read_freqs(loc):
     counts.smooth()
     log_total = math.log(total)
     probs = {}
-    for line in open(loc):
+    for line in loc.open():
         freq, doc_freq, key = line.split('\t', 2)
         if int(doc_freq) >= 2 and int(freq) >= 5 and len(key) < 200:
             word = literal_eval(key)

From 65f3ce6c52c688a26497cbb10c3a82b15f684f7c Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal@gmail.com>
Date: Sat, 25 Jul 2015 22:36:43 +0200
Subject: [PATCH 2/4] * Require preshed 0.41

---
 requirements.txt | 2 +-
 setup.py         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/requirements.txt b/requirements.txt
index 7a5292e96..bf10a1165 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,7 +1,7 @@
 cython
 cymem == 1.11
 pathlib
-preshed == 0.37
+preshed == 0.41
 thinc == 3.3
 murmurhash == 0.24
 unidecode
diff --git a/setup.py b/setup.py
index 36f43a296..15c850627 100644
--- a/setup.py
+++ b/setup.py
@@ -119,7 +119,7 @@ def run_setup(exts):
                       "spacy.syntax": ["*.pxd"]},
         ext_modules=exts,
         license="Dual: Commercial or AGPL",
-        install_requires=['numpy', 'murmurhash', 'cymem >= 1.11', 'preshed == 0.37',
+        install_requires=['numpy', 'murmurhash', 'cymem >= 1.11', 'preshed == 0.41',
                           'thinc == 3.3', "unidecode", 'wget', 'plac', 'six',
                           'ujson'],
         setup_requires=["headers_workaround"],

From 8e800adfbcfde54958b27b5ef41f8240a8384d27 Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal@gmail.com>
Date: Sat, 25 Jul 2015 22:54:08 +0200
Subject: [PATCH 3/4] * Fix init_model

---
 bin/init_model.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/bin/init_model.py b/bin/init_model.py
index 2d3c62368..414e2b05c 100644
--- a/bin/init_model.py
+++ b/bin/init_model.py
@@ -89,6 +89,9 @@ def _read_probs(loc):
 
 
 def _read_freqs(loc):
+    if not loc.exists():
+        print("Warning: Frequencies file not found")
+        return None
     counts = PreshCounter()
     total = 0
     for line in loc.open():

From 3fe14b8ed6673b05199a64e2b397202baf75eba9 Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal@gmail.com>
Date: Sat, 25 Jul 2015 22:55:53 +0200
Subject: [PATCH 4/4] * Fix CFile for Python2

---
 spacy/cfile.pyx | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/spacy/cfile.pyx b/spacy/cfile.pyx
index 6346302b9..781759466 100644
--- a/spacy/cfile.pyx
+++ b/spacy/cfile.pyx
@@ -5,6 +5,8 @@ cdef class CFile:
     def __init__(self, loc, mode):
         if isinstance(mode, unicode):
             mode_str = mode.encode('ascii')
+        else:
+            mode_str = mode
         cdef bytes bytes_loc = loc.encode('utf8') if type(loc) == unicode else loc
         self.fp = fopen(<char*>bytes_loc, mode_str)
         if self.fp == NULL: