Threats affecting pkg:brew/molecule

### Summary On Windows, attacker-controlled content in a wheel's `.dist-info/namespace_packages.txt` can cause setuptools' legacy wheel installation path to create an `__init__.py` outside the requested `destination_eggdir`. The attacker can influence the destination path, but not the contents of the created file. ### Details Tested setuptools commit: `6ead555c5fb29bc57fe6105b1bffc163f56fd558` Relevant source: `setuptools/wheel.py` SHA-256: `359b7e0d5dd8c3135653abe08a8ed5dc0352eadfd7df22cb5e2b5303b7c3cafd` The relevant call path is: `Wheel(wheel_path).install_as_egg(destination)` → `_install_as_egg` → `_convert_metadata` → `_fix_namespace_packages` A wheel can provide attacker-controlled content in: `<distribution>.dist-info/namespace_packages.txt` The namespace value is eventually used approximately as: ```python mod_dir = os.path.join(destination_eggdir, *mod.split('.')) mod_init = os.path.join(mod_dir, '__init__.py') If mod is a Windows drive-qualified absolute path containing no dots, mod.split('.') preserves it as a single component. On Windows, os.path.join(destination_eggdir, absolute_component) discards the original base directory and resolves to the attacker-supplied absolute path. The existing wheel filename and ZIP traversal checks do not validate this metadata value before it is subsequently interpreted as a filesystem path. ```markdown ### PoC The reproduction was performed entirely inside a disposable temporary directory. Layout: ```text sandbox/ destination/ outside/ input/ canary.txt ``` A synthetic wheel was created containing: `demo_pkg-1.0.dist-info/namespace_packages.txt` whose content was the absolute path of: `sandbox\outside` The real acquired setuptools implementation was then invoked: ```python setuptools.wheel.Wheel(synthetic_wheel).install_as_egg( sandbox_destination ) ``` The operation completed without exception. Expected metadata was created under: `sandbox\destination\EGG-INFO` but setuptools also created: `sandbox\outside\__init__.py` The file did not exist before execution. Its contents were: ```python __import__('pkg_resources').declare_namespace(__name__) ``` The unrelated canary remained byte-identical. The sandbox was deleted after recording the evidence. ``` ### Impact A crafted wheel can cause setuptools to create a fixed-content `__init__.py` outside the intended installation directory at a filesystem path influenced by the wheel, with the permissions of the process performing the installation. The demonstrated reproduction required the target parent directory to already exist.

## Summary When building a source distribution (`python -m build --sdist` / `setup.py sdist`), setuptools' `FileList` applies `MANIFEST.in` directives (`exclude`, `global-exclude`, `recursive-exclude`, `prune`) by matching a compiled glob against on-disk file names **byte-for-byte, with no Unicode normalization**. On normalization-preserving filesystems (notably macOS APFS and HFS+), a file written in NFD and a `MANIFEST.in` rule written in NFC refer to the same file but are byte-distinct, so the exclusion silently fails to match. A file the maintainer intended to exclude is then packed into the `.tar.gz` and, if published, uploaded to the public, immutable PyPI index. ## Details File names in `FileList.files` come from `os.walk` (`setuptools/_distutils/filelist.py`, `_find_all_simple`), so on APFS a file written NFD is offered to the matcher in NFD, while the `MANIFEST.in` pattern carries the author's editor form (typically NFC). The matching path performs no canonicalization: ```python # setuptools/command/egg_info.py (FileList.global_exclude) def global_exclude(self, pattern): match = translate_pattern(os.path.join('**', pattern)) # fnmatch.translate -> regex, no NFC/NFD return self._remove_files(match.match) # byte-level regex over raw os.walk names ``` A rule written NFC (`café` = `63 61 66 c3 a9`) does not match an on-disk name written NFD (`café` = `63 61 66 65 cc 81`), even though the filesystem treats the two as one file. A `unicodedata.normalize('NFD', ...)` helper exists in `setuptools/unicode_utils.py` (`decompose()`), but it is **never called in the manifest matching path**, so neither the pattern nor the walked path is normalized before matching. The only normalization in this area, `EggInfoCommand._manifest_normalize`, uses `filesys_decode` (bytes→str decode only, no NFC/NFD) and runs when writing `SOURCES.txt`, after matching has already occurred. ## Impact `MANIFEST.in` exclusions are the documented mechanism maintainers use to keep secrets, local configs, and private fixtures out of the published sdist. A non-ASCII excluded file may be published to the public, immutable PyPI index despite the rule — an irreversible disclosure with no visual cue (NFC and NFD forms render identically). Exposure is filesystem-dependent and most relevant on macOS APFS/HFS+, where many maintainers build and publish. Pure-ASCII rules are unaffected. ## Proof of concept With a project containing `MANIFEST.in`: ``` global-include *.txt *.json global-exclude secret_café.txt # rule saved NFC ``` and an on-disk file `secret_café.txt` written in NFD, `python -m build --sdist` packs the secret file into the resulting `.tar.gz`, while an ASCII control file excluded by the same directive is correctly dropped — isolating the bypass to the NFC-pattern vs. NFD-name mismatch. Reproduced on macOS APFS with setuptools 82.0.1. ## Remediation Normalize both the walked path and each `MANIFEST.in` pattern to a single canonical form before matching, in both `setuptools/command/egg_info.py` (`FileList`) and the vendored `setuptools/_distutils/filelist.py`. For an exclusion list, err toward excluding more, and document that `MANIFEST.in` matching is normalization-insensitive on macOS. ## Credit Reported by Tomas Illuminati. Coordinated via CERT/CC VINCE VU#604762.