Merge pull request #2851 from DimitriPapadopoulos/lgtm_errors

LGTM.com errors

nltk/internals.py

@@ -238,11 +238,11 @@ def read_str(s, start_position):
             break
 
     # Process it, using eval.  Strings with invalid escape sequences
-    # might raise ValueEerror.
+    # might raise ValueError.
     try:
         return eval(s[start_position : match.end()]), match.end()
     except ValueError as e:
-        raise ReadError("invalid string (%s)" % e) from e
+        raise ReadError("valid escape sequence", start_position) from e
 
 
 _READ_INT_RE = re.compile(r"-?\d+")

nltk/test/internals.doctest

@@ -138,3 +138,23 @@ It works for classic classes, too:
     False
     >>> overridden(D().f)
     True
+
+
+read_str()
+~~~~~~~~~~~~
+    >>> from nltk.internals import read_str
+
+Test valid scenarios
+
+    >>> read_str("'valid string'", 0)
+    ('valid string', 14)
+
+Now test invalid scenarios
+    >>> read_str("should error", 0)
+    Traceback (most recent call last):
+    ...
+    nltk.internals.ReadError: Expected open quote at 0
+    >>> read_str("'should error", 0)
+    Traceback (most recent call last):
+    ...
+    nltk.internals.ReadError: Expected close quote at 1

nltk/util.py

@@ -1013,7 +1013,7 @@ def skipgrams(sequence, n, k, **kwargs):
 ######################################################################
 
 # inherited from pywordnet, by Oliver Steele
-def binary_search_file(file, key, cache={}, cacheDepth=-1):
+def binary_search_file(file, key, cache=None, cacheDepth=-1):
     """
     Return the line from the file with first word key.
     Searches through a sorted file using the binary search algorithm.
@@ -1036,6 +1036,9 @@ def binary_search_file(file, key, cache={}, cacheDepth=-1):
         end = file.tell() - 1
         file.seek(0)
 
+    if cache is None:
+        cache = {}
+
     while start < end:
         lastState = start, end
         middle = (start + end) // 2