Issue1336
Created on 2009-05-03.21:28:03 by pjenvey, last changed 2009-07-28.05:58:21 by pjenvey.
|
| File name |
Uploaded |
Description |
Edit |
Remove |
|
deque_repeated_code.patch
|
thatmattbone,
2009-07-24.01:31:38
|
deque extends pysequence, reduces duplication |
|
|
|
| msg4629 (view) |
Author: Philip Jenvey (pjenvey) |
Date: 2009-05-03.21:28:02 |
|
PyDeque obviously cut and pasted much of PySequence, it looks very likely
that it should be subclassing it instead.
Its tests also run through some of tests.seq_tests (but not all of them).
|
| msg4938 (view) |
Author: Matt Bone (thatmattbone) |
Date: 2009-07-24.01:31:38 |
|
Here's my stab at a patch. The tests pass, but it's probably all that
useful (the cpython deque does not extend sequence).
|
| msg4954 (view) |
Author: Philip Jenvey (pjenvey) |
Date: 2009-07-28.05:58:20 |
|
Thanks for this, a couple things:
no tabs, just 4 space indents
pyget looks like cut and paste of getNode, I think they need
refactoring..
Actually I think getNode is probably unnecessary if we have a working
pyget. Then the PySequence __get/set/delitem__ methods can do all the
work for us and we wouldn't have to override them to use getNode.
Those PySequence methods should also do the work of the first few lines
of getNode for us (the adjusting of the index and throwing an IndexError
if it's invalid) via PySequence.delegator
|
|
| Date |
User |
Action |
Args |
| 2009-07-28 05:58:22 | pjenvey | set | messages:
+ msg4954 |
| 2009-07-26 14:01:39 | fwierzbicki | set | nosy:
+ fwierzbicki |
| 2009-07-24 01:31:40 | thatmattbone | set | files:
+ deque_repeated_code.patch
keywords:
+ patch
messages:
+ msg4938
nosy:
+ thatmattbone |
| 2009-05-03 21:28:03 | pjenvey | create | |
|
