Remove PyOS_ReadlineFunctionPointer from the stable ABI list

[encukou]

BPO	43868
Nosy	@pfmoore, @vstinner, @tjguk, @encukou, @zware, @zooba
PRs	#25442

^{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 2021-04-27.14:17:45.393>
created_at = <Date 2021-04-16.14:48:28.426>
labels = ['expert-C-API', '3.10', 'OS-windows']
title = 'Remove PyOS_ReadlineFunctionPointer from the stable ABI list'
updated_at = <Date 2021-04-27.14:17:45.393>
user = 'https://github.com/encukou'

bugs.python.org fields:

activity = <Date 2021-04-27.14:17:45.393>
actor = 'vstinner'
assignee = 'none'
closed = True
closed_date = <Date 2021-04-27.14:17:45.393>
closer = 'vstinner'
components = ['Windows', 'C API']
creation = <Date 2021-04-16.14:48:28.426>
creator = 'petr.viktorin'
dependencies = []
files = []
hgrepos = []
issue_num = 43868
keywords = ['patch']
message_count = 3.0
messages = ['391206', '391237', '391690']
nosy_count = 6.0
nosy_names = ['paul.moore', 'vstinner', 'tim.golden', 'petr.viktorin', 'zach.ware', 'steve.dower']
pr_nums = ['25442']
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue43868'
versions = ['Python 3.10']

[encukou]

The inclusion of PyOS_ReadlineFunctionPointer in python3dll.c (*) was a mistake. According to PEP-384:

functions expecting FILE* are not part of the ABI, to avoid depending on a specific version of the Microsoft C runtime DLL on Windows.

The situation may have changed and it might be reasonable to revisit this decision, but that would call for a larger discussion. There are FILE*-taking functions that are probably much ore useful than this one. (But, I think it's a good idea to limit the stable ABI to file-like Python objects anyway.)

I see PEP-384 as being definitive (where it's not ambiguous). The python3dll.c list and public/private headers do not actually define the stable ABI.

So, I'd like to remove the function from the list.

---

(*) it was actually PC/python3.def in 3.2

[vstinner]

Yes, please remove it from the limited C API.

PyOS_ReadlineFunctionPointer has a bad API. I fixed recently PyOS_StdioReadline because it used the Python C API without holding the GIL. To acquire the GIL, you need to access a *private* _PyOS_ReadlineTState variable which is kind of unfortunate :-(

commit c353764
Author: Victor Stinner <vstinner@python.org>
Date: Mon Jun 1 20:59:35 2020 +0200

bpo-40826: Fix GIL usage in PyOS_Readline() (GH-20579)

Fix GIL usage in PyOS_Readline(): lock the GIL to set an exception.

Pass tstate to my_fgets() and _PyOS_WindowsConsoleReadline(). Cleanup
these functions.

See also:

commit fa7ab6a
Author: Victor Stinner <vstinner@python.org>
Date: Wed Jun 3 14:39:59 2020 +0200

bpo-40826: Add _PyOS_InterruptOccurred(tstate) function (GH-20599)

my_fgets() now calls _PyOS_InterruptOccurred(tstate) to check for
pending signals, rather calling PyOS_InterruptOccurred().

my_fgets() is called with the GIL released, whereas
PyOS_InterruptOccurred() must be called with the GIL held.

test_repl: use text=True and avoid SuppressCrashReport in
test_multiline_string_parsing().

Fix my_fgets() on Windows: fgets(fp) does crash if fileno(fp) is closed.

[encukou]

New changeset 91b69b7 by Petr Viktorin in branch 'master':
bpo-43868: Remove PyOS_ReadlineFunctionPointer from the stable ABI list (GH-25442)
91b69b7