Implement the formatting part of PEP 515, '_' in numeric literals.

[ericvsmith]

BPO	27080
Nosy	@brettcannon, @birkenfeld, @ericvsmith, @ned-deily, @Rosuav
Files	underscores_decimal_only.patch underscores_all_bases.patch underscores_with_tests.patch moar-tests.patch

^{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 = 'https://github.com/ericvsmith'
closed_at = <Date 2016-09-10.03:08:14.367>
created_at = <Date 2016-05-21.20:38:09.840>
labels = ['interpreter-core', 'type-feature', 'release-blocker']
title = "Implement the formatting part of PEP 515, '_' in numeric literals."
updated_at = <Date 2016-09-10.03:08:14.366>
user = 'https://github.com/ericvsmith'

bugs.python.org fields:

activity = <Date 2016-09-10.03:08:14.366>
actor = 'eric.smith'
assignee = 'eric.smith'
closed = True
closed_date = <Date 2016-09-10.03:08:14.367>
closer = 'eric.smith'
components = ['Interpreter Core']
creation = <Date 2016-05-21.20:38:09.840>
creator = 'eric.smith'
dependencies = []
files = ['44152', '44153', '44154', '44425']
hgrepos = []
issue_num = 27080
keywords = ['patch']
message_count = 13.0
messages = ['266023', '273093', '273117', '273120', '273123', '274769', '275150', '275151', '275165', '275367', '275545', '275549', '275550']
nosy_count = 6.0
nosy_names = ['brett.cannon', 'georg.brandl', 'eric.smith', 'ned.deily', 'python-dev', 'Rosuav']
pr_nums = []
priority = 'release blocker'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue27080'
versions = ['Python 3.6']

[ericvsmith]

I've separated this out from bpo-26331, so as to not interfere with discussions and code reviews for the parsing portion of the PEP.

[ericvsmith]

Unfortunately I'm not going to have time to implement this before 3.6 beta 1, so I'm un-assigning it to myself. I'll post to python-dev and hopefully someone else can get to it. Apologies for waiting so long.

[Rosuav]

Here's a first-cut patch. No docs, no tests, and applies only to decimal formatting. It involves redefining the thousands_separators flag from being a boolean to being a three-state flag (none, comma, or underscore), and (ab)uses LT_*_LOCALE to carry that same information around.

[Rosuav]

Hmm, adding bin/oct/hex support didn't turn out that hard. Although it feels like this code is getting hackish. Definitely needs code review!

[Rosuav]

Hmm, strange. Comma formatting never seems to have had tests added. So I've added a couple of simple tests of comma formatting at the same time as adding underscore formatting tests. Also, test_long.py currently has a comment "# octal" preceding a bunch of tests of *binary* formatting, and no tests of octal. So I've added those too.

In any case, here's a patch with full functionality and tests.

[Rosuav]

Another couple of tests, per comments.

[brettcannon]

Latest patch LGTM. If Georg isn't able to get to issue bpo-26331 then I'll commit to tomorrow. Do you want me to commit yours on your behalf right after, Chris to make sure this hits b1?

[brettcannon]

Actually, Chris doesn't have commit privileges so I just will. :)

[Rosuav]

Thanks Brett! Sounds like a plan.

[ericvsmith]

Since Brett is working on the parsing part, I'll work on this. I might take a stab at the documentation first, but in any event I'll commit it tonight.