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

@@ -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:

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):

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)

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)