Repository navigation
Python Decimal supports '#' format, C Decimal does not. #91060
Description
| BPO | 46904 |
|---|---|
| Nosy | @mdickinson, @ericvsmith, @cgevans |
Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.
Show more details
GitHub fields:
assignee = None
closed_at = None
created_at = <Date 2022-03-02.23:39:51.993>
labels = ['type-bug', 'library', '3.10', '3.11']
title = "Python Decimal supports '#' format, C Decimal does not."
updated_at = <Date 2022-03-03.00:03:06.014>
user = 'https://github.com/cgevans'bugs.python.org fields:
activity = <Date 2022-03-03.00:03:06.014>
actor = 'eric.smith'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['Library (Lib)']
creation = <Date 2022-03-02.23:39:51.993>
creator = 'const'
dependencies = []
files = []
hgrepos = []
issue_num = 46904
keywords = []
message_count = 1.0
messages = ['414390']
nosy_count = 3.0
nosy_names = ['mark.dickinson', 'eric.smith', 'const']
pr_nums = []
priority = 'normal'
resolution = None
stage = None
status = 'open'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue46904'
versions = ['Python 3.10', 'Python 3.11']
Activity
cgevans commentedon Mar 2, 2022
Decimal as implemented in _pydecimal supports the '#' alternate form in format strings, though this does not appear to be documented: as with floats, it causes a decimal point to be included regardless of value. Decimal as implemented in cdecimal, as it uses libmpdec, does not.
This is a rather minor discrepancy; I hadn't actually intended to have # in my format string at all, but it had ended up there and was causing errors for half of the users of my library. # also could be useful for decimals, as it can be for floats.
Like bpo-45739, it's unclear to me whether this is a discrepancy that should be corrected by implementing # for cdecimal, removing # support for _pydecimal, or just documenting the inconsistency.
mdickinson commentedon Apr 12, 2022
I agree that the Python and C versions of Decimal should be consistent. The support in the Python version definitely was intentional, and I'd be reluctant to remove it. So options are: (a) do nothing, (b) document the discrepancy, or (c) fix cdecimal.
#58046 is somewhat related.
skrah commentedon Jan 28, 2024
This is fixed by applying the general format patch in #114563.
gh-114563: C decimal falls back to pydecimal for unsupported format s…
belm0 commentedon Feb 12, 2024
please close-- fIxed in PR #114879
[3.12] gh-114563: C decimal falls back to pydecimal for unsupported f…
[3.11] [3.12] pythongh-114563: C decimal falls back to pydecimal for …
[3.11] gh-114563: C decimal falls back to pydecimal for unsupported f…
pythongh-114563: C decimal falls back to pydecimal for unsupported fo…