gh-112388: Fix an error that was causing the parser to try to overwrite tokenizer errors

[pablogsal]

• Issue: Assertion failure in Objects/call.c:342: !_PyErr_Occurred(tstate) failed #112388

[carsonRadtke]

Maybe scope creep, but is there an opportunity for an improvement here?

_PyPegen_raise_error always returns NULL - maybe it should be void.

[bedevere-bot]

🤖 New build scheduled with the buildbot fleet by @sunmy2019 for commit cef5ae9 🤖

If you want to schedule another build, you need to add the 🔨 test-with-buildbots label again.

[sunmy2019]

Looks good

[sunmy2019]

_PyPegen_raise_error always returns NULL - maybe it should be void.

We need a placeholder here. If you track the caller, it's like

void *a;
if (all_good)
    a = ...;
else
    a = _PyPegen_raise_error(...);

[lysnikolaou]

LGTM!

[lysnikolaou]

_PyPegen_raise_error always returns NULL - maybe it should be void.

Most of the time we assign the return value of _PyPegen_raise_error to the variable returned by the rule. This is a "hack" to set a specific error from a grammar action, but still signify to the parser that the rule has failed and it should backtrack all the way.

[miss-islington-app]

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

[bedevere-app]

GH-112466 is a backport of this pull request to the 3.12 branch.

[bedevere-app]

GH-112467 is a backport of this pull request to the 3.11 branch.