Repository navigation
Warnings operate out of global runtime state. #80918
Description
| BPO | 36737 |
|---|---|
| Nosy | @brettcannon, @ericsnowcurrently, @zooba |
| PRs |
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/ericsnowcurrently'
closed_at = <Date 2019-05-10.17:30:26.460>
created_at = <Date 2019-04-26.22:28:58.200>
labels = ['expert-subinterpreters', '3.8', 'type-bug', '3.7', '3.9']
title = 'Warnings operate out of global runtime state.'
updated_at = <Date 2020-05-15.00:37:35.654>
user = 'https://github.com/ericsnowcurrently'bugs.python.org fields:
activity = <Date 2020-05-15.00:37:35.654>
actor = 'vstinner'
assignee = 'eric.snow'
closed = True
closed_date = <Date 2019-05-10.17:30:26.460>
closer = 'eric.snow'
components = ['Subinterpreters']
creation = <Date 2019-04-26.22:28:58.200>
creator = 'eric.snow'
dependencies = []
files = []
hgrepos = []
issue_num = 36737
keywords = ['patch']
message_count = 4.0
messages = ['340951', '340987', '340988', '342083']
nosy_count = 3.0
nosy_names = ['brett.cannon', 'eric.snow', 'steve.dower']
pr_nums = ['13159']
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue36737'
versions = ['Python 3.7', 'Python 3.8', 'Python 3.9']
Activity
ericsnowcurrently commentedon Apr 26, 2019
(See Include/internal/pycore_warnings.h and Python/_warnings.c.)
The warnings module's state (filters, default action, etc.) is currently stored at the level of the global runtime. That's a problem for the following reasons:
While those are all a problem in a future where interpreters don't share the GIL, that last one is a problem right now for people using subinterpreters.
One of the following should happen:
I could also see use cases for *also* configuring warnings process-wide but that could be handled separately if actually desired.
zooba commentedon Apr 27, 2019
Beyond "warning configuration is inherited by new interpreters", I don't see any reason to have process wide configuration. People using Python directly will get "process" wide from runtime configuration, and embedders don't want anything implicitly process-wide at all.
ericsnowcurrently commentedon Apr 27, 2019
Good point.
Also, the whole idea of inheriting things (settings, some copied objects, etc.) into subinterpreters is interesting. My initial reaction is that folks would appreciate that feature, at least for a handful of things. It's not critical, but is worth adding to the list of deferred ideas in PEP-554. :)
ericsnowcurrently commentedon May 10, 2019
New changeset 86ea581 by Eric Snow in branch 'master':
bpo-36737: Use the module state C-API for warnings. (gh-13159)
86ea581