cap str/bytes node expansion in container and dict.fromkeys inference

[kali834x]

Type of Changes

	Type
✓	🐛 Bug fix

Description

_container_generic_transform (for list/set/tuple/frozenset) and infer_dict_fromkeys build one Const node per character when their argument infers to a str/bytes constant, with no size cap. the multiply guard in const_infer_binary_op bounds "x" * n, but string concatenation ("a" + "b") is not size-bounded and neither is a large literal, so list(("a" * 10 ** 8) + "b") walks the whole string and materializes ~100 million Const nodes (tens of GB) during inference of otherwise untrusted source. cap both sites the same way _multiply_seq_by_int already caps sequence repetition: fall back to the default inference once the string exceeds 1e8 characters, before the nodes are built. small calls (list("abc"), dict.fromkeys("ab")) keep inferring their exact contents.

[codspeed-hq]

Merging this PR will not alter performance

✅ 3 untouched benchmarks
⏩ 1 skipped benchmark¹

---

_{Comparing kali834x:container-str-node-expansion (198e6aa) with main (faf2884)}

Footnotes

1. 1 benchmark was skipped, so the baseline result was used instead. If it was deleted from the codebase, click here and archive it to remove it from the performance reports. ↩

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.62%. Comparing base (faf2884) to head (198e6aa).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3127   +/-   ##
=======================================
  Coverage   93.62%   93.62%           
=======================================
  Files          92       92           
  Lines       11381    11385    +4     
=======================================
+ Hits        10655    10659    +4     
  Misses        726      726           

Flag	Coverage Δ
linux	93.49% <100.00%> (+<0.01%)	⬆️
pypy	93.62% <100.00%> (+<0.01%)	⬆️
windows	93.59% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
astroid/brain/brain_builtin_inference.py	93.36% <100.00%> (+0.05%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[DanielNoord]

Please add a test for this :)

[kali834x]

Added tests in tests/brain/test_brain.py covering list/set/tuple/frozenset and dict.fromkeys: the oversized concat case declines (Instance / empty Dict) and the small-string case still infers its exact contents.

[DanielNoord]

Can you fix merge conflicts and the pre-commit check?

[kali834x]

Rebased onto main, the conflict was just the ChangeLog entry. pre-commit passes locally with the updated hooks.