gensim/corpora/textcorpus.py

Survived

Mutant 4

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -47,7 +47,7 @@
 from gensim.parsing.preprocessing import STOPWORDS, RE_WHITESPACE
 from gensim.utils import deaccent, simple_tokenize
 
-logger = logging.getLogger(__name__)
+logger = None
 
 
 def remove_stopwords(tokens, stopwords=STOPWORDS):

Mutant 5

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -69,7 +69,7 @@
     return [token for token in tokens if token not in stopwords]
 
 
-def remove_short(tokens, minsize=3):
+def remove_short(tokens, minsize=4):
     """Remove tokens shorter than `minsize` chars.
 
     Parameters

Mutant 6

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -88,7 +88,7 @@
     return [token for token in tokens if len(token) >= minsize]
 
 
-def lower_to_unicode(text, encoding='utf8', errors='strict'):
+def lower_to_unicode(text, encoding='XXutf8XX', errors='strict'):
     """Lowercase `text` and convert to unicode, using :func:`gensim.utils.any2unicode`.
 
     Parameters

Mutant 7

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -88,7 +88,7 @@
     return [token for token in tokens if len(token) >= minsize]
 
 
-def lower_to_unicode(text, encoding='utf8', errors='strict'):
+def lower_to_unicode(text, encoding='utf8', errors='XXstrictXX'):
     """Lowercase `text` and convert to unicode, using :func:`gensim.utils.any2unicode`.
 
     Parameters

Mutant 8

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -477,7 +477,7 @@
     """
 
     def __init__(self, input, dictionary=None, metadata=False, min_depth=0, max_depth=None,
-                 pattern=None, exclude_pattern=None, lines_are_documents=False, **kwargs):
+                 pattern=None, exclude_pattern=None, lines_are_documents=True, **kwargs):
         """
 
         Parameters

Mutant 10

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -516,7 +516,6 @@
     def lines_are_documents(self):
         return self._lines_are_documents
 
-    @lines_are_documents.setter
     def lines_are_documents(self, lines_are_documents):
         self._lines_are_documents = lines_are_documents
         self.length = None

Mutant 12

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -525,7 +525,6 @@
     def pattern(self):
         return self._pattern
 
-    @pattern.setter
     def pattern(self, pattern):
         self._pattern = None if pattern is None else re.compile(pattern)
         self.length = None

Mutant 14

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -534,7 +534,6 @@
     def exclude_pattern(self):
         return self._exclude_pattern
 
-    @exclude_pattern.setter
     def exclude_pattern(self, pattern):
         self._exclude_pattern = None if pattern is None else re.compile(pattern)
         self.length = None

Mutant 16

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -543,7 +543,6 @@
     def min_depth(self):
         return self._min_depth
 
-    @min_depth.setter
     def min_depth(self, min_depth):
         self._min_depth = min_depth
         self.length = None

Mutant 18

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -552,7 +552,6 @@
     def max_depth(self):
         return self._max_depth
 
-    @max_depth.setter
     def max_depth(self, max_depth):
         self._max_depth = max_depth
         self.length = None

Mutant 19

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -620,7 +620,7 @@
             self.length = sum(1 for _ in self.getstream())
 
 
-def walk(top, topdown=True, onerror=None, followlinks=False, depth=0):
+def walk(top, topdown=False, onerror=None, followlinks=False, depth=0):
     """Generate the file names in a directory tree by walking the tree either top-down or bottom-up.
     For each directory in the tree rooted at directory top (including top itself), it yields a 4-tuple
     (depth, dirpath, dirnames, filenames).

Mutant 20

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -620,7 +620,7 @@
             self.length = sum(1 for _ in self.getstream())
 
 
-def walk(top, topdown=True, onerror=None, followlinks=False, depth=0):
+def walk(top, topdown=True, onerror=None, followlinks=True, depth=0):
     """Generate the file names in a directory tree by walking the tree either top-down or bottom-up.
     For each directory in the tree rooted at directory top (including top itself), it yields a 4-tuple
     (depth, dirpath, dirnames, filenames).

Mutant 21

--- gensim/corpora/textcorpus.py
+++ gensim/corpora/textcorpus.py
@@ -620,7 +620,7 @@
             self.length = sum(1 for _ in self.getstream())
 
 
-def walk(top, topdown=True, onerror=None, followlinks=False, depth=0):
+def walk(top, topdown=True, onerror=None, followlinks=False, depth=1):
     """Generate the file names in a directory tree by walking the tree either top-down or bottom-up.
     For each directory in the tree rooted at directory top (including top itself), it yields a 4-tuple
     (depth, dirpath, dirnames, filenames).