Identify binary operators used as unary operators

Previously we handled the case where binary operators were being used
as unary operators except in the case where they followed another
binary operator, e.g.,

    foo = (1 +
           -10 *
           2 /
           -5)

This change updates the check for W503 to track the previous
non-whitespace token type and token text to check if it is in fact also
a binary operator (because you should never have two binary operators in
a row).

This does not handle invalid syntax, e.g.,

    foo = (bar /
           /baz)

But the false-positive generated for other cases was more harmful than
not catching what will instead be caught by the interpreter.

Closes gh-484

1 files changed, 10 insertions, 1 deletions

diff --git a/pep8.py b/pep8.py
index bfbea7f..39eb426 100755
--- a/pep8.py
+++ b/pep8.py
@@ -1018,6 +1018,8 @@ def break_around_binary_operator(logical_line, tokens):
     Okay: foo(x,\n    -y)
     Okay: foo(x,  # comment\n    -y)
     Okay: var = (1 &\n       ~2)
+    Okay: var = (1 /\n       -2)
+    Okay: var = (1 +\n       -1 +\n       -2)
     """
     def is_binary_operator(token_type, text):
         # The % character is strictly speaking a binary operator, but the
@@ -1028,6 +1030,9 @@ def break_around_binary_operator(logical_line, tokens):
 
     line_break = False
     unary_context = True
+    # Previous non-newline token types and text
+    previous_token_type = None
+    previous_text = None
     for token_type, text, start, end, line in tokens:
         if token_type == tokenize.COMMENT:
             continue
@@ -1035,10 +1040,14 @@ def break_around_binary_operator(logical_line, tokens):
             line_break = True
         else:
             if (is_binary_operator(token_type, text) and line_break and
-                    not unary_context):
+                    not unary_context and
+                    not is_binary_operator(previous_token_type,
+                                           previous_text)):
                 yield start, "W503 line break before binary operator"
             unary_context = text in '([{,;'
             line_break = False
+            previous_token_type = token_type
+            previous_text = text
 
 
 def comparison_to_singleton(logical_line, noqa):