homepage

multiprocessing.sharedctypes.Array and
multiprocessing.sharedctypes.Value if used according to documentation
fail with AssertionError.
Python 3.0b1+ (py3k:64518, Jun 25 2008, 12:52:38)
[GCC 4.2.3 (Ubuntu 4.2.3-2ubuntu7)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from multiprocessing import sharedctypes
>>> sharedctypes.Array('i', 1, lock=True)
Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 File "/usr/local/lib/python3.0/multiprocessing/sharedctypes.py", line
88, in Array
assert hasattr(lock, 'acquire')
AssertionError
>>> sharedctypes.Array('i', 1, lock=False)
Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 File "/usr/local/lib/python3.0/multiprocessing/sharedctypes.py", line
88, in Array
assert hasattr(lock, 'acquire')
AssertionError
>>> sharedctypes.Array('i', 1, lock=None)
<SynchronizedArray wrapper for
<multiprocessing.sharedctypes.c_long_Array_1 object at 0x83214f4>>
>>> Value('i', lock=True)
Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 File "/usr/local/lib/python3.0/multiprocessing/__init__.py", line 246,
in Value
return Value(typecode_or_type, *args, **kwds)
 File "/usr/local/lib/python3.0/multiprocessing/sharedctypes.py", line
75, in Value
assert hasattr(lock, 'acquire')
AssertionError
The same goes for multiprocessing.Array and multiprocessing.Value.
Comparing code to documentation it's obvious that lock argument can be
one of Lock, RLock or None objects (or any object with "acquire"
attribute), but not True or False. Also, looking at the code it seems
strange to me that 'lock' is a keyword-only argument. Why not use simple
default argument "lock=None" for Array() function?
Proposed patch tries to address these issues.

And here is the patch itself. :)

Note that if the error is in the documentation semantics, and not in the
implementation, then the benchmark code in the documentation is also
broken, and should be change to not use lock=True/False respectively...
I'm not sure if the patch here is good or rather lock=True/False should
be the right API and the implementation should be changed along this lines:
diff --git a/Lib/multiprocessing/sharedctypes.py
b/Lib/multiprocessing/sharedctypes.py
index b94cd52..2f68e74 100644
--- a/Lib/multiprocessing/sharedctypes.py
+++ b/Lib/multiprocessing/sharedctypes.py
@@ -79,10 +79,11 @@ def Array(typecode_or_type, size_or_initializer,
**kwds):
 if kwds:
 raise ValueError('unrecognized keyword argument(s): %s' %
list(kwds.keys()))
 obj = RawArray(typecode_or_type, size_or_initializer)
- if lock is None:
+ if lock is True:
 lock = RLock()
- assert hasattr(lock, 'acquire')
- return synchronized(obj, lock)
+ return synchronized(obj, lock)
+ return obj
+ 
 
 def copy(obj):
 new_obj = _new_value(type(obj))

oops, there is a proper code patch in issue4449 , and the documentation
is right.

Dupe, issue4449