gh-69919: Catch all compile errors in the code module, pyrepl and IDLE

[serhiy-storchaka]

compile() can raise more than SyntaxError, OverflowError and ValueError: MemoryError ("Parser stack overflowed") for a few thousand chained unary operators, RecursionError ("Stack overflow during compilation") for tens of thousands of chained binary operators, and a bare MemoryError when memory really runs out.

code.InteractiveInterpreter.runsource() let those propagate, so code.interact() died. The new REPL survives through its outer loop, but prints a bare MemoryError without the message, or a RecursionError traceback through its own internals. In IDLE's Shell the exception escaped into the Tk callback, no new prompt was issued, and the next Enter re-compiled the same text: the Shell was stuck until the input was deleted.

Based on Terry's 2016 patch from the issue: runsource() in code and _pyrepl, _pyrepl's incomplete-input check, the startup-file compile in pyshell.py, runscript.checksyntax(), and pdb's find_function() and the check of condition/display/ignore expressions catch Exception. The SyntaxError-specific code in IDLE (marking the position in the text) runs only for SyntaxError; other exceptions are reported through the base class, i.e. by traceback. IDLE now also names the actual exception, so an IndentationError is no longer reported as SyntaxError.

The compile() documentation now lists the exceptions it can raise (including OverflowError for a too large source), and notes the 3.12 change for null bytes from ValueError to SyntaxError.

An alternative that fixes this in compile() itself, for main only, is #157586.

🤖 Generated with Claude Code

• Issue: Code and IDLE should catch all compile errors. #69919

[read-the-docs-community]

Documentation build overview

📚 cpython-previews | 🛠️ Build #34585425 | 📁 Comparing e5c5118 against main (82952e3)

  🔍 Preview build  

3 files changed

± builtins/functions.html
± library/code.html
± whatsnew/changelog.html

[terryjreedy]

I learned something new: mock.mock_open.

[johnslavik]

@serhiy-storchaka I think we should also catch SystemError. This could then supersede #151480, see issue #150732. Feel free to steal commits from my PR.

[serhiy-storchaka]

No, SystemError should not be raised at first place.

[terryjreedy]

The change to the non-idlelib .py files is to replace an inadequate and not consistent tuple of exception classes with Exception. This is the same as done in idlelib and should not need further review. The non-idlelib doc changes look good to me.

[miss-islington-app]

Thanks @serhiy-storchaka for the PR 🌮🎉.. I'm working now to backport this PR to: 3.13, 3.14, 3.15.
🐍🍒⛏🤖

[miss-islington-app]

Sorry, @serhiy-storchaka, I could not cleanly backport this to 3.14 due to a conflict.
Please backport using cherry_picker on command line.

cherry_picker 4bc392c13462096213777c6b2bdc470028c41536 3.14

[miss-islington-app]

Sorry, @serhiy-storchaka, I could not cleanly backport this to 3.13 due to a conflict.
Please backport using cherry_picker on command line.

cherry_picker 4bc392c13462096213777c6b2bdc470028c41536 3.13

[bedevere-app]

GH-158025 is a backport of this pull request to the 3.15 branch.

[bedevere-app]

GH-158029 is a backport of this pull request to the 3.14 branch.

[bedevere-app]

GH-158030 is a backport of this pull request to the 3.13 branch.