Document the meaning of values for sys.float_info.rounds

[rhettinger]

BPO	41822
Nosy	@tim-one, @rhettinger, @mdickinson, @serhiy-storchaka, @eamanu, @rahul-kumi

^{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 2020-09-21.00:38:26.708>
labels = ['easy', '3.10', 'docs']
title = 'Document the meaning of values for sys.float_info.rounds'
updated_at = <Date 2020-10-18.09:13:18.609>
user = 'https://github.com/rhettinger'

bugs.python.org fields:

activity = <Date 2020-10-18.09:13:18.609>
actor = 'rhettinger'
assignee = 'docs@python'
closed = False
closed_date = None
closer = None
components = ['Documentation']
creation = <Date 2020-09-21.00:38:26.708>
creator = 'rhettinger'
dependencies = []
files = []
hgrepos = []
issue_num = 41822
keywords = ['newcomer friendly']
message_count = 5.0
messages = ['377236', '378852', '378853', '378854', '378856']
nosy_count = 7.0
nosy_names = ['tim.peters', 'rhettinger', 'mark.dickinson', 'docs@python', 'serhiy.storchaka', 'eamanu', 'rahul-kumi']
pr_nums = []
priority = 'normal'
resolution = None
stage = None
status = 'open'
superseder = None
type = None
url = 'https://bugs.python.org/issue41822'
versions = ['Python 3.10']

Linked PRs

• gh-85988: Change documentation for sys.float_info.rounds #99675

[rhettinger]

The current docs unnecessarily refer readers to the C99 standard. "integer constant representing the rounding mode used for arithmetic operations. This reflects the value of the system FLT_ROUNDS macro at interpreter startup time. See section 5.2.4.2.2 of the C99 standard for an explanation of the possible values and their meanings."

The docs should quote the standard, "The rounding mode for floating-point addition is characterized by the implementation defined value of FLT_ROUNDS: -1 indeterminable, 0 toward zero, 1 to nearest, 2 toward positive infinity, 3 toward negative infinity. All other values for FLT_ROUNDS characterize implementation-defined rounding."
behavior.

[rhettinger]

Also, we should document the meaning of the other fields. Several of them are not self explanatory:

min_10_exp = -307

does not mean that 2.3e-308 isn't normalized

does mean that 10**-307 is normalized and 10**-308 isn't

min_exp = -1021

does not mean that 2**-1021 is the lowest normalized value

does mean ldexp(0.5, -1021) == float_info.min

decimal_dig = 15

does not mean digits required to specify every unique float

does mean, "number of decimal digits, q, such that any
floating-point number with q decimal digits can be rounded
into a floating-point number with p radix b digits and back
again without change to the q decimal digits"

[serhiy-storchaka]

Python does not set the meaning and the value of these fields itself. It exposes C constants. Should we just copy definitions from corresponding C or floating-point standards?

[serhiy-storchaka]

Oh, seems it was what Raymond proposed initially.

[rhettinger]

Here's a link: http://c0x.coding-guidelines.com/5.2.4.2.2.html