c3b39245a7
Issue #4580 : slicing of memoryviews when itemsize != 1 is wrong.
...
Also fix len() to return number of items rather than length in bytes.
I'm sorry it was not possible for me to work on this without reindenting
a bit some stuff around. The indentation in memoryobject.c is a mess,
I'll open a separate bug for it.
2009-01-03 16:59:18 +00:00
bc420400eb
Issue #4569 : Interpreter crash when mutating a memoryview with an item size larger than 1.
...
(together with a bit of reindenting)
2008-12-07 20:14:49 +00:00
5e19e44413
kill memoryview.size in favor of len(view)
...
Reviewer: Antoine Pitrou #3827
2008-09-10 21:47:03 +00:00
c6b09ebe58
#3712 : The memoryview object had a reference leak and didn't support cyclic garbage collection.
...
Reviewed by Benjamin Peterson.
2008-09-01 15:10:14 +00:00
616d28566b
Issue #2394 : implement more of the memoryview API.
2008-08-19 22:09:34 +00:00
ee8712cda4
#2621 rename test.test_support to test.support
2008-05-20 21:35:26 +00:00
7b6fc8e19d
Fixed memoryview constructor. It allowed arbitrary keyword arguments. The bug was found by mykhal from #python. I've also added a small test case in the new test_memoryview.py
2007-11-08 02:28:11 +00:00
Antoine Pitrou
Benjamin Peterson
Christian Heimes