msg107556 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2010-06-11 15:32 |
Module level pack, unpack etc. methods have similar functionality with Struct instance methods, but docs are different. The immediate issue is the lack of signature in the module level methods' docstrings.
$ ./python.exe -m pydoc struct.Struct.pack
Help on method_descriptor in struct.Struct:
struct.Struct.pack = pack(...)
S.pack(v1, v2, ...) -> bytes
Return a bytes containing values v1, v2, ... packed according to this
Struct's format. See struct.__doc__ for more on format strings.
and
$ ./python.exe -m pydoc struct.pack
Help on built-in function pack in struct:
struct.pack = pack(...)
Return bytes containing values v1, v2, ... packed according to fmt.
|
msg107557 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2010-06-11 15:41 |
Two more bits:
1. "See struct.__doc__", while technically correct, is not user friendly. If you copy struct.__doc__ to >>> prompt, you get an ugly repr of a multiline string. I suggest s/struct.__doc__/help(struct)/.
2. For some reason struct.Struct does not show up in help(struct).
|
msg107558 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-11 15:42 |
Thanks for the reports; I'll look at this a little later.
|
msg107655 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 14:34 |
Here's a patch. Alexander, do you want to check it for sanity?
|
msg107656 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 14:34 |
N.B. The patch doesn't fix the missing struct.Struct documentation issue; I'll look at that separately.
|
msg107657 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 14:45 |
The docstrings need to be rewrapped to 72 characters; I'll do this once the wording is settled.
|
msg107658 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2010-06-12 15:18 |
calcsize() still does not have signature in docstring.
A nit: maybe follow calcsize() lead and say "format string fmt" rather than just "fmt".
One more __doc__ mention: in Struct docstring,
| __init__(...)
| x.__init__(...) initializes x; see x.__class__.__doc__ for signature
I think this is inherited, but quite confusing here because help(Struct) does not have signature. Maybe just add signature to help(Struct).
|
msg107659 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 15:22 |
r81947 fixes the missing struct.Struct entry in struct.help. (pydoc was relying on the inspect module to correctly guess the module for struct.Struct, and inspect was getting it wrong. Adding __all__ to the struct module saves inspect from having to guess.)
|
msg107660 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 15:22 |
I also accidentally included the docstring changes in r81947; those were reverted in r81948.
|
msg107663 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 15:44 |
Committed patch (with additional changes suggested by Alexander) in r81949. Only the __init__ doc issue remains.
|
msg107664 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2010-06-12 15:51 |
See issue #8983 for the the __init__ doc discussion since it is not specific to the struct module.
|
msg107666 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 15:53 |
Apart from the pydoc __init__ issue, the Struct class documentation should probably be expanded a bit. At the moment, it's just:
PyDoc_STRVAR(s__doc__, "Compiled struct object");
Alexander, interested in producing a patch?
|
msg107686 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2010-06-12 19:25 |
How does issue8973-Struct.diff look?
|
msg107687 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2010-06-12 19:30 |
Looks fine. I'd probably call the argument 'fmt' rather than 'format', for consistency. Please commit, with or without the 'format' -> 'fmt' change, as you choose.
|
msg107688 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2010-06-12 19:38 |
Committed in r81961. Yes, I used "format" for consistency with the manual, but on the second thought consistency within help() is more important.
|
msg127593 - (view) |
Author: SilentGhost (SilentGhost) * |
Date: 2011-01-31 11:50 |
r81947 introduced this issue:
>>> from struct import *
>>> pack_into
Traceback (most recent call last):
File "<pyshell#1>", line 1, in <module>
pack_into
NameError: name 'pack_into' is not defined
struct.__all__ has a duplicate entry and misses pack_into.
Patch is attached, test passes. Seems quite innocent to me, could it make into 3.2, George?
|
msg127606 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2011-01-31 15:48 |
The proposed patch looks fine to me, but it is attached to the wrong issue. It belongs to #8973 or better yet to follow RC2 rules pedantically, it should be posted in a separate issue.
This is important, because this issue is limited to docstrings and may be considered documentation only, but the proposed parch affects behavior. Still I would be +1 on applying it.
|
msg127609 - (view) |
Author: Alexander Belopolsky (belopolsky) * |
Date: 2011-01-31 15:56 |
Wait, #8973 *is* this issue. But r81947 is clearly not only about docstrings. We definitely need a separate issue. This is too confusing.
|
msg127626 - (view) |
Author: SilentGhost (SilentGhost) * |
Date: 2011-01-31 17:23 |
new issue #11081 was created for struct.__all__ fix
|
|
Date |
User |
Action |
Args |
2022-04-11 14:57:02 | admin | set | github: 53219 |
2011-01-31 17:23:54 | SilentGhost | set | status: open -> closed nosy:
georg.brandl, mark.dickinson, belopolsky, SilentGhost messages:
+ msg127626
|
2011-01-31 15:56:04 | belopolsky | set | nosy:
georg.brandl, mark.dickinson, belopolsky, SilentGhost messages:
+ msg127609 |
2011-01-31 15:55:42 | belopolsky | set | nosy:
georg.brandl, mark.dickinson, belopolsky, SilentGhost messages:
- msg127607 |
2011-01-31 15:51:14 | belopolsky | set | nosy:
georg.brandl, mark.dickinson, belopolsky, SilentGhost messages:
+ msg127607 |
2011-01-31 15:48:46 | belopolsky | set | nosy:
georg.brandl, mark.dickinson, belopolsky, SilentGhost messages:
+ msg127606 |
2011-01-31 11:50:15 | SilentGhost | set | status: closed -> open files:
+ struct.py.diff
nosy:
+ georg.brandl, SilentGhost messages:
+ msg127593
|
2010-06-12 19:42:03 | belopolsky | set | status: open -> closed resolution: accepted stage: resolved |
2010-06-12 19:38:10 | belopolsky | set | messages:
+ msg107688 |
2010-06-12 19:30:25 | mark.dickinson | set | messages:
+ msg107687 |
2010-06-12 19:25:01 | belopolsky | set | files:
+ issue8973-Struct.diff
messages:
+ msg107686 |
2010-06-12 15:53:50 | mark.dickinson | set | assignee: mark.dickinson -> belopolsky messages:
+ msg107666 |
2010-06-12 15:51:21 | belopolsky | set | messages:
+ msg107664 |
2010-06-12 15:44:21 | mark.dickinson | set | messages:
+ msg107663 |
2010-06-12 15:22:48 | mark.dickinson | set | messages:
+ msg107660 |
2010-06-12 15:22:16 | mark.dickinson | set | messages:
+ msg107659 |
2010-06-12 15:18:03 | belopolsky | set | messages:
+ msg107658 |
2010-06-12 14:45:10 | mark.dickinson | set | messages:
+ msg107657 |
2010-06-12 14:39:20 | mark.dickinson | set | files:
- issue8973.patch |
2010-06-12 14:39:15 | mark.dickinson | set | files:
+ issue8973.patch |
2010-06-12 14:34:30 | mark.dickinson | set | messages:
+ msg107656 |
2010-06-12 14:34:01 | mark.dickinson | set | files:
+ issue8973.patch keywords:
+ patch messages:
+ msg107655
|
2010-06-11 15:42:54 | mark.dickinson | set | assignee: mark.dickinson messages:
+ msg107558 |
2010-06-11 15:41:38 | belopolsky | set | messages:
+ msg107557 |
2010-06-11 15:32:36 | belopolsky | create | |