From b098ef69a4a45efa8011a2f4b93ee55eca4bcc26 Mon Sep 17 00:00:00 2001
From: dzhwinter <dzhwinter@gmail.com>
Date: Wed, 24 May 2017 19:26:39 +0800
Subject: [PATCH] "remove the rar extractfile, prevent small files"

---
 python/paddle/v2/dataset/__init__.py | 3 ++-
 python/paddle/v2/dataset/mq2007.py   | 6 ++----
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/python/paddle/v2/dataset/__init__.py b/python/paddle/v2/dataset/__init__.py
index 80ff6295c34..26252d5bbd7 100644
--- a/python/paddle/v2/dataset/__init__.py
+++ b/python/paddle/v2/dataset/__init__.py
@@ -24,8 +24,9 @@ import conll05
 import uci_housing
 import sentiment
 import wmt14
+import mq2007
 
 __all__ = [
     'mnist', 'imikolov', 'imdb', 'cifar', 'movielens', 'conll05', 'sentiment'
-    'uci_housing', 'wmt14'
+    'uci_housing', 'wmt14', 'mq2007'
 ]
diff --git a/python/paddle/v2/dataset/mq2007.py b/python/paddle/v2/dataset/mq2007.py
index fd71b341662..d8c9918d140 100644
--- a/python/paddle/v2/dataset/mq2007.py
+++ b/python/paddle/v2/dataset/mq2007.py
@@ -41,9 +41,7 @@ def __initialize_meta_info__():
   """
     fn = fetch()
     rar = rarfile.RarFile(fn)
-    dirpath = os.path.dirname(fn)
-    rar.extractall(path=dirpath)
-    return dirpath
+    return rar
 
 
 class Query(object):
@@ -273,7 +271,7 @@ def load_from_text(filepath, shuffle=True, fill_missing=-1):
     querylists = []
     querylist = None
     fn = __initialize_meta_info__()
-    with open(os.path.join(fn, filepath)) as f:
+    with fn.open(os.path.join(fn, filepath)) as f:
         for line in f:
             query = Query()
             query = query._parse_(line)
-- 
GitLab