gh-107972: Argument Clinic: Ensure a C basename is provided after 'as'

[erlend-aasland]

• Issue: Argument Clinic silently allow empty custom C basenames #107972

[AlexWaygood]

+1 to Adam's suggestions, but otherwise this LGTM. It appears we dofail() currently for at least the second case you're adding a test for, but the error message is pretty nonsensical currently:

Couldn't find existing function 'foo2'! 

[erlend-aasland]

+1 to Adam's suggestions, but otherwise this LGTM. It appears we dofail() currently for at least the second case you're adding a test for, but the error message is pretty nonsensical currently:

Couldn't find existing function 'foo2'! 

Well, if you'd defined foo2 first, you would not get that failure; it would silently accept the weird line, but still produce a valid C basename (just using the full name).

[AlexWaygood]

Well, if you'd defined foo2 first, you would not get that failure; it would silently accept the weird line, but still produce a valid C basename (just using the full name).

Yup, I gathered. Maybe we should do that in the test, actually, so that the only erroneous thing in the test case is the one we're explicitly testing that clinic catches?

[erlend-aasland]

Try this weird input; clinic accepts it on main:

/*[clinic input]output everything blockfoo2[clinic start generated code]*//*[clinic input]output everything blockfoo as = foo2[clinic start generated code]*/

[erlend-aasland]

Try this weird input; clinic accepts it on main:

And with that in mind, we should probably adjust the test case to look exactly like that.

[erlend-aasland]

Yup, I gathered. Maybe we should do that in the test, actually, so that the only erroneous thing in the test case is the one we're explicitly testing that clinic catches?

I did not read this until now; we came to the same conclusion :)

[erlend-aasland]

If it won't backport cleanly, I'm inclined to not backport it.

[erlend-aasland]

Thanks for the reviews!

[miss-islington]

Thanks @erlend-aasland for the PR 🌮🎉.. I'm working now to backport this PR to: 3.11, 3.12.
🐍🍒⛏🤖

[miss-islington]

Sorry, @erlend-aasland, I could not cleanly backport this to 3.12 due to a conflict.
Please backport using cherry_picker on command line.
cherry_picker 607f18c89456cdc9064e27f86a7505e011209757 3.12

[miss-islington]

Sorry, @erlend-aasland, I could not cleanly backport this to 3.11 due to a conflict.
Please backport using cherry_picker on command line.
cherry_picker 607f18c89456cdc9064e27f86a7505e011209757 3.11

[bedevere-bot]

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot aarch64 RHEL8 LTO + PGO 3.x has failed when building commit 607f18c.

What do you need to do:

1. Don't panic.
2. Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
3. Go to the page of the buildbot that failed (https://buildbot.python.org/all/#builders/78/builds/5072) and take a look at the build logs.
4. Check if the failure is related to this commit (607f18c) or if it is a false positive.
5. If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.

You can take a look at the buildbot page here:

https://buildbot.python.org/all/#builders/78/builds/5072

Summary of the results of the build (if available):

== Tests result: ENV CHANGED ==

431 tests OK.

10 slowest tests:

• test_concurrent_futures: 2 min 18 sec
• test_multiprocessing_spawn: 1 min 31 sec
• test_signal: 1 min 15 sec
• test_multiprocessing_forkserver: 1 min 8 sec
• test_multiprocessing_fork: 1 min 4 sec
• test_socket: 38.0 sec
• test_io: 37.9 sec
• test_imaplib: 32.0 sec
• test_subprocess: 28.3 sec
• test_xmlrpc: 27.7 sec

1 test altered the execution environment:
test_interpreters

15 tests skipped:
test.test_asyncio.test_windows_events
test.test_asyncio.test_windows_utils test_devpoll test_gdb
test_ioctl test_kqueue test_launcher test_startfile test_tkinter
test_ttk test_winconsoleio test_winreg test_winsound test_wmi
test_zipfile64

Total duration: 2 min 18 sec

Click to see traceback logs

Traceback (most recent call last): File "<frozen importlib._bootstrap>", line 1266, in _find_and_load File "<frozen importlib._bootstrap>", line 1237, in _find_and_load_unlocked File "<frozen importlib._bootstrap>", line 841, in _load_unlocked File "<frozen importlib._bootstrap_external>", line 999, in exec_module File "<frozen importlib._bootstrap_external>", line 1095, in get_code File "<frozen importlib._bootstrap_external>", line 1194, in get_dataTypeError: descriptor 'close' for '_io.BufferedReader' objects doesn't apply to a '_io.FileIO' object Traceback (most recent call last): File "/home/buildbot/buildarea/3.x.cstratak-RHEL8-aarch64.lto-pgo/build/Lib/threading.py", line 1052, in _bootstrap_innerself.run() File "/home/buildbot/buildarea/3.x.cstratak-RHEL8-aarch64.lto-pgo/build/Lib/threading.py", line 989, in runself._target(*self._args, **self._kwargs) File "/home/buildbot/buildarea/3.x.cstratak-RHEL8-aarch64.lto-pgo/build/Lib/test/test_interpreters.py", line 481, in task interp = interpreters.create() ^^^^^^^^^^^^^^^^^^^^^ File "/home/buildbot/buildarea/3.x.cstratak-RHEL8-aarch64.lto-pgo/build/Lib/test/support/interpreters.py", line 25, in createid= _interpreters.create(isolated=isolated) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^RuntimeError: interpreter creation failed k