From d703bce30236f278818d7f346b2d746256871380 Mon Sep 17 00:00:00 2001
From: JesseyXujin <xu.jinxujinxujin@163.com>
Date: Fri, 10 Jan 2020 16:30:46 +0800
Subject: [PATCH] paddle coredump exception fix (#807)

* paddle_null_point_fix

* add core expception note

* delete yield

* modify test paddle for supporting enable_paddle()
---
 jieba/__init__.py          | 3 ++-
 jieba/posseg/__init__.py   | 3 ++-
 test/test_paddle.py        | 2 +-
 test/test_paddle_postag.py | 3 ++-
 4 files changed, 7 insertions(+), 4 deletions(-)
 mode change 100644 => 100755 jieba/__init__.py
 mode change 100644 => 100755 test/test_paddle.py
 mode change 100644 => 100755 test/test_paddle_postag.py

diff --git a/jieba/__init__.py b/jieba/__init__.py
old mode 100644
new mode 100755
index 992039e..04f53b1
--- a/jieba/__init__.py
+++ b/jieba/__init__.py
@@ -298,8 +298,9 @@ class Tokenizer(object):
         is_paddle_installed = check_paddle_install['is_paddle_installed']
         sentence = strdecode(sentence)
         if use_paddle and is_paddle_installed:
+            # if sentence is null, it will raise core exception in paddle.
             if sentence is None or sentence == "" or sentence == u"":
-                yield sentence
+                return
             import jieba.lac_small.predict as predict
             results = predict.get_sent(sentence)
             for sent in results:
diff --git a/jieba/posseg/__init__.py b/jieba/posseg/__init__.py
index 05d7c01..df47c99 100755
--- a/jieba/posseg/__init__.py
+++ b/jieba/posseg/__init__.py
@@ -279,8 +279,9 @@ def cut(sentence, HMM=True, use_paddle=False):
     """
     is_paddle_installed = check_paddle_install['is_paddle_installed']
     if use_paddle and is_paddle_installed:
+        # if sentence is null, it will raise core exception in paddle.
         if sentence is None or sentence == "" or sentence == u"":
-            yield pair(None, None)
+            return
         import jieba.lac_small.predict as predict
         sents, tags = predict.get_result(strdecode(sentence))
         for i, sent in enumerate(sents):
diff --git a/test/test_paddle.py b/test/test_paddle.py
old mode 100644
new mode 100755
index a964533..6003dea
--- a/test/test_paddle.py
+++ b/test/test_paddle.py
@@ -2,7 +2,7 @@
 import sys
 sys.path.append("../")
 import jieba
-
+jieba.enable_paddle()
 
 def cuttest(test_sent):
     result = jieba.cut(test_sent, use_paddle=True)
diff --git a/test/test_paddle_postag.py b/test/test_paddle_postag.py
old mode 100644
new mode 100755
index b8c29c1..c7d083c
--- a/test/test_paddle_postag.py
+++ b/test/test_paddle_postag.py
@@ -2,7 +2,8 @@
 import sys
 sys.path.append("../")
 import jieba.posseg as pseg
-
+import jieba
+jieba.enable_paddle()
 
 def cuttest(test_sent):
     result = pseg.cut(test_sent, use_paddle=True)