Python3.9 changes col_offset for some ast nodes

[hauntsaninja]

BPO	40349
Nosy	@serhiy-storchaka, @pablogsal, @isidentical, @hauntsaninja

^{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 = <Date 2020-04-21.06:43:12.775>
created_at = <Date 2020-04-21.03:58:08.777>
labels = ['invalid']
title = 'Python3.9 changes col_offset for some ast nodes'
updated_at = <Date 2020-04-21.06:43:12.772>
user = 'https://github.com/hauntsaninja'

bugs.python.org fields:

activity = <Date 2020-04-21.06:43:12.772>
actor = 'serhiy.storchaka'
assignee = 'none'
closed = True
closed_date = <Date 2020-04-21.06:43:12.775>
closer = 'serhiy.storchaka'
components = []
creation = <Date 2020-04-21.03:58:08.777>
creator = 'hauntsaninja'
dependencies = []
files = []
hgrepos = []
issue_num = 40349
keywords = []
message_count = 2.0
messages = ['366909', '366910']
nosy_count = 4.0
nosy_names = ['serhiy.storchaka', 'pablogsal', 'BTaskaya', 'hauntsaninja']
pr_nums = []
priority = 'normal'
resolution = 'not a bug'
stage = 'resolved'
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue40349'
versions = []

[hauntsaninja]

With Python 3.8:

Python 3.8.1 (default, Jan 23 2020, 23:36:06) 
[Clang 11.0.0 (clang-1100.0.33.17)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import ast
>>> ast.parse("(a).x").body[0].value.col_offset
1

With Python 3.9:

Python 3.9.0a5+ (heads/master:799d7d6, Apr  6 2020, 16:05:37) 
[Clang 11.0.0 (clang-1100.0.33.17)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import ast
>>> ast.parse("(a).x").body[0].value.col_offset
0

Maybe related to the new parser? I couldn't find what the environment variable to turn it off was.

For context, I'm trying to get mypy tests to pass with 3.9, but the tests make use of specific col_offsets for error reporting. It would be nice if these continued to match with 3.9

[serhiy-storchaka]

col_offset = 0 is the correct value. There was a bug which is fixed now (see issue39474).