PEP 3121, 384 refactoring applied to dbm module

[RobinSchreiber]

BPO	15650
Nosy	@loewis, @vstinner, @asvetlov
Superseder	bpo-1635741: Py_Finalize() doesn't clear all Python objects at exit
Files	_dbmmodule_pep3121-384_v0.patch Issue15650_v2.diff

^{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 2020-12-18.00:43:53.569>
created_at = <Date 2012-08-14.18:14:33.875>
labels = ['extension-modules']
title = 'PEP 3121, 384 refactoring applied to dbm module'
updated_at = <Date 2020-12-18.00:43:53.568>
user = 'https://bugs.python.org/RobinSchreiber'

bugs.python.org fields:

activity = <Date 2020-12-18.00:43:53.568>
actor = 'vstinner'
assignee = 'none'
closed = True
closed_date = <Date 2020-12-18.00:43:53.569>
closer = 'vstinner'
components = ['Extension Modules']
creation = <Date 2012-08-14.18:14:33.875>
creator = 'Robin.Schreiber'
dependencies = []
files = ['26804', '27774']
hgrepos = []
issue_num = 15650
keywords = ['pep3121']
message_count = 7.0
messages = ['168216', '172310', '172320', '172323', '172644', '176650', '383279']
nosy_count = 4.0
nosy_names = ['loewis', 'vstinner', 'asvetlov', 'Robin.Schreiber']
pr_nums = []
priority = 'normal'
resolution = 'duplicate'
stage = 'resolved'
status = 'closed'
superseder = '1635741'
type = None
url = 'https://bugs.python.org/issue15650'
versions = ['Python 3.4']

[RobinSchreiber]

Changes proposed in PEP-3121 and PEP-384 have now been applied to the dbm module!

[asvetlov]

Robin, why do you increment refcounter for type object on every construction of dbm instance?

Also I would to see macros in uppercase if possible.

[RobinSchreiber]

PEP-384 demands, among other things, that the TypeObjects themselves are transformed to Heaptypes. This means that the Typeobjects, that are created from this new stable ABI, reside within the heap and therefore have to be managed by the Python GC. This is of course done by appropriately incref- and decrefing of the specific Heaptype.

[asvetlov]

From my perspective type object is referenced by module state.
While module is present it's state hold all types exposed by module.
After module has deleted it's state has cleaned up and types has gone away — thats look good to me.

Instances of python classes (not extensions) hold ownership for it's class. I guess it done because python class can be constructed inside function and instance of this class should work after exit from that function like closure does.

In case of extension modules enough to hold type reference into module state itself.

Thoughts?

[asvetlov]

Robin, after thinking I would to agree with your decision to hold reference to type into type instance.

Please, can you describe your check like:

if((void *)type->tp_dealloc == (void *)dbm_dealloc) {
    Py_DECREF(type);
}

Why you decref only if type->tp_dealloc points to module's dealloc? What are you protected from? What other values also possible?

[RobinSchreiber]

Take a look at the discussion in bpo-15653.

[vstinner]

Fixed by:

commit bf69a8f
Author: Dong-hee Na <donghee.na92@gmail.com>
Date: Tue Jun 16 01:20:54 2020 +0900

bpo-1635741: Port _dbm module to multiphase initialization (GH-20848)