mloda-pii 0.2.1__tar.gz

mloda_pii-0.2.1/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

mloda_pii-0.2.1/PKG-INFO

@@ -0,0 +1,29 @@
+Metadata-Version: 2.4
+Name: mloda-pii
+Version: 0.2.1
+Summary: PII detection and redaction plugins for mloda
+Author-email: Tom Kaltofen <info@mloda.ai>
+License-Expression: Apache-2.0
+Keywords: mloda,mloda-plugin,pii,privacy,redaction,anonymization
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+License-File: LICENSE
+Requires-Dist: mloda[pyarrow]<0.18.0,>=0.17.0
+Requires-Dist: tomli>=1.0.0; python_version < "3.11"
+Provides-Extra: hook
+Requires-Dist: mloda-enterprise[anonymizer]>=0.11.0; extra == "hook"
+Provides-Extra: dev
+Requires-Dist: mloda-testing[binary-model]>=0.11.0; extra == "dev"
+Requires-Dist: mloda-enterprise>=0.11.0; extra == "dev"
+Requires-Dist: tox; extra == "dev"
+Requires-Dist: tox-uv; extra == "dev"
+Requires-Dist: uv>=0.13.4; extra == "dev"
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: ruff; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: bandit; extra == "dev"
+Dynamic: license-file

mloda_pii-0.2.1/README.md

@@ -0,0 +1,136 @@
+[![License](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](https://github.com/mloda-ai/mloda-plugin-template/blob/main/LICENSE)
+[![mloda](https://img.shields.io/badge/built%20with-mloda-blue.svg)](https://github.com/mloda-ai/mloda)
+[![Python](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/)
+[![Tests](https://github.com/mloda-ai/mloda-plugin-template/actions/workflows/test.yml/badge.svg?branch=main)](https://github.com/mloda-ai/mloda-plugin-template/actions/workflows/test.yml?query=branch%3Amain)
+[![Scaffold](https://github.com/mloda-ai/mloda-plugin-template/actions/workflows/scaffold-test.yml/badge.svg?branch=main)](https://github.com/mloda-ai/mloda-plugin-template/actions/workflows/scaffold-test.yml?query=branch%3Amain)
+
+# mloda-plugin-template
+
+> **A GitHub template for creating standalone mloda plugins.** Part of the [mloda](https://github.com/mloda-ai/mloda) ecosystem for open data access. Visit [mloda.ai](https://mloda.ai) for an overview and business context, the [GitHub repository](https://github.com/mloda-ai/mloda) for technical context, or the [documentation](https://mloda-ai.github.io/mloda/) for detailed guides.
+
+Create your own FeatureGroups, ComputeFrameworks, and Extenders as standalone packages. This repository serves two audiences:
+
+- **Plugin authors**: click *Use this template* on GitHub to scaffold a new plugin repository, then follow the [Use this template](#use-this-template) section below.
+- **Template contributors**: improving the scaffold itself? See [CONTRIBUTING.md](CONTRIBUTING.md) and the [Contribute to this template](#contribute-to-this-template) section.
+
+## mloda-pii
+
+- **`<col>__pii_anonymized`**: a license-free FeatureGroup that tokenizes a string column with the anonymizer-rs binary (keyed HMAC-SHA256). Build the binary with `bin/build-anonymizer-rs.sh`.
+- **Claude Code PII hook**: pseudonymizes identifiers in tool traffic through the `mloda run` CLI, licensed (default) or local. Setup, license and limits: [mloda_pii/user_interface/hook/README.md](mloda_pii/user_interface/hook/README.md); install with `uv tool install "mloda-pii[hook]"`, then `mloda-pii init <invite>` in your project. Design: [docs/pii-hook-mvp-spec.md](docs/pii-hook-mvp-spec.md).
+- The hook needs mloda 0.17.0 or later, the first release with the `mloda run` CLI.
+
+## Related Repositories
+
+- **[mloda](https://github.com/mloda-ai/mloda)**: The core library for open data access. Declaratively define what data you need, not how to get it. mloda handles feature resolution, dependency management, and compute framework abstraction automatically.
+
+- **[mloda-registry](https://github.com/mloda-ai/mloda-registry)**: The central hub for discovering and sharing mloda plugins. Browse community-contributed FeatureGroups, find integration guides, and publish your own plugins for others to use.
+
+## Use this template
+
+Click *Use this template* on GitHub to scaffold a new plugin repository. See [docs/getting-started.md](docs/getting-started.md) for the GitHub template-creation walkthrough; once your repository is in place, follow the steps below to customize the scaffold for your organization.
+
+### Structure
+
+```
+placeholder/
+├── feature_groups/
+│   ├── manifest.py               # Entry-point manifest: FEATURE_GROUPS list
+│   └── anonymizer_rs/            # <col>__pii_anonymized via the anonymizer-rs binary, no license
+│       ├── __init__.py           # Package exports
+│       ├── base.py               # Public API: name pattern, options
+│       ├── binary.py             # Binary resolution and Arrow IPC, no mloda imports
+│       ├── pyarrow_anonymizer_rs.py
+│       └── tests/
+└── user_interface/               # mloda-pii command: hook, license, stats
+```
+
+Each plugin kind ships an example test next to its implementation, put yours in the same
+place.
+
+### Plugin discovery
+
+mloda auto-discovers installed plugins via `PluginLoader.all()`, no manual import needed. Each
+`manifest.py` lists its concrete classes (`FEATURE_GROUPS`, `COMPUTE_FRAMEWORKS`, `EXTENDERS`) and
+`pyproject.toml` wires them up under `[project.entry-points."mloda.*"]`. Add a plugin by appending
+it to the relevant list.
+
+For an optional backend, do not wrap manifest imports in try/except. Give each optional extra its own
+manifest and entry point, keep the base manifest free of optional imports, and add a
+`mloda.optional_dependencies` marker entry named like the entry it protects (needed unless the backend
+is one of core's built-in roots, such as pandas or polars). The marker module must import without the
+backend. mloda then skips the entry when the backend is missing, lists it in
+`PluginLoader.skipped_plugins()`, and still raises on your plugin's own import errors. See
+[guide 04: Optional Backends](https://github.com/mloda-ai/mloda-registry/blob/main/docs/guides/04-create-plugin-package.md#optional-backends).
+
+### Key files
+
+- `placeholder/` - Root namespace (rename to your organization's name)
+- `pyproject.toml` - Package config (edit directly, not auto-generated)
+- `.github/workflows/test.yml` - CI workflow running pytest
+
+### Setup Your Plugin
+
+#### 1. Run the customization script
+
+```bash
+./bin/customize.sh <your-package-name> \
+  --author "Your Name" \
+  --email you@example.com \
+  --description "Your plugin description" \
+  --repository-url https://github.com/<your-org>/<your-repo>
+```
+
+This renames `placeholder/` to `<your-package-name>/`, updates `pyproject.toml` (`name`, `authors`, `description`, `keywords`, `packages.find.include`, `pytest.testpaths`), updates `.releaserc.yaml` (`message`, `repositoryUrl`), rewrites the `[project.entry-points."mloda.*"]` manifest paths, and rewrites `from placeholder.` imports across the package. Then add your plugin's domain terms to `keywords` so it shows up in PyPI search.
+
+The package name must be a valid Python identifier (lowercase letters, digits, underscores; must start with a letter). The names `mloda` and `mloda_plugins` are reserved: the core `mloda` package is a shared PEP 420 namespace, so a plugin that shipped `mloda/__init__.py` would make mloda unimportable for everyone who installs it. Both `bin/customize.sh` and a `tox`/CI check reject these names. All option flags are optional; if you omit them you can edit the corresponding fields by hand later.
+
+#### 2. Verify setup
+
+```bash
+uv venv && source .venv/bin/activate && uv sync --all-extras && tox
+```
+
+After `tox` passes, confirm `pyproject.toml` no longer contains the template's placeholder strings:
+
+```bash
+tox -e placeholders
+```
+
+This is also enforced in CI: the `test.yml` workflow fails on scaffolded plugins until the template's placeholder strings are removed from `pyproject.toml`. The check covers the placeholder-bearing fields from step 1: `name`, `authors` (name and email), `description`, `tool.setuptools.packages.find.include`, and `tool.pytest.ini_options.testpaths`. It is skipped on the `mloda-ai/mloda-plugin-template` repository itself, where the placeholders are intentional.
+
+#### 3. Remove template-only files
+
+After `tox` passes, remove the files that only exist to support the template itself:
+
+```bash
+rm CONTRIBUTING.md bin/customize.sh tests/test_customize_script.py
+```
+
+`CONTRIBUTING.md` describes how to contribute to the template repo; `bin/customize.sh` is a one-shot scaffold script that has nothing left to do. `tests/test_customize_script.py` tests that script and copies it into a scratch scaffold, so it must go with it — leaving it behind fails `tox` with a `FileNotFoundError` on every subsequent run. The other test files (`test_entry_points.py`, `test_mloda_imports.py`, `test_reserved_namespace.py`) are rename-safe and should stay.
+
+Also delete the `## First-time setup` section from `CLAUDE.md` and `AGENTS.md`. Those instructions only apply to fresh-template repos.
+
+The remaining baseline files apply to your plugin out of the box and can be edited to match your conventions:
+
+- `AGENTS.md` and `CLAUDE.md`: toolchain and project practices for the same `tox`/ruff/mypy/bandit pipeline you inherit. Tune the bullets if you change the toolchain.
+- `CODE_OF_CONDUCT.md`: short, plain-English baseline. Update the contact (`conduct@mloda.ai` → your address) if you want enforcement to come to you.
+- `.github/ISSUE_TEMPLATE/issue.yml`: unified issue form. `bin/customize.sh` already repoints its example file paths at your renamed package; edit the field wording if you want examples drawn from your own plugin.
+
+You may also want to replace this `README.md` with one that describes your plugin.
+
+### Where to next
+
+- **[mloda-registry/docs/guides/](https://github.com/mloda-ai/mloda-registry/tree/main/docs/guides/)**: full plugin development walkthrough (FeatureGroups, ComputeFrameworks, Extenders, packaging, publishing).
+- **[mloda](https://github.com/mloda-ai/mloda)**: core framework reference.
+- **[Claude Code skills](https://github.com/mloda-ai/mloda-registry/tree/main/.claude/skills/)**: pattern guidance and best practices for AI-assisted plugin development.
+- **[docs/github-workflows.md](docs/github-workflows.md)**: CI/CD setup and required secrets for the included workflows.
+- **[docs/github-repository-settings.md](docs/github-repository-settings.md)**: repository-side settings (secrets, branch protection, required checks, dependabot reviewer) you should configure after scaffolding.
+- **[docs/pulling-template-updates.md](docs/pulling-template-updates.md)**: how to pull a later template fix into an already-scaffolded repo, and how to resolve the conflicts scaffolding causes.
+
+## Contribute to this template
+
+This section is for people improving the scaffold itself (CI workflows, dev tooling, docs, examples). See [CONTRIBUTING.md](CONTRIBUTING.md) for the full contributor guide. Quick pointers:
+
+- [Code of Conduct](CODE_OF_CONDUCT.md)
+- [AGENTS.md](AGENTS.md): agent guidance, project practices, issue creation
+- [Issue template](.github/ISSUE_TEMPLATE/issue.yml)

mloda_pii-0.2.1/mloda_pii/feature_groups/__init__.py

@@ -0,0 +1 @@
+"""Feature groups namespace package."""

mloda_pii-0.2.1/mloda_pii/feature_groups/anonymizer_rs/__init__.py

@@ -0,0 +1,13 @@
+"""Local, license-free FeatureGroup over the anonymizer-rs binary."""
+
+from mloda_pii.feature_groups.anonymizer_rs.base import PII_KEY_FILE_ENV, LocalAnonymizerFeatureGroup
+from mloda_pii.feature_groups.anonymizer_rs.binary import AnonymizerRsError, derive_anon_key
+from mloda_pii.feature_groups.anonymizer_rs.pyarrow_anonymizer_rs import PyArrowLocalAnonymizerFeatureGroup
+
+__all__ = [
+    "PII_KEY_FILE_ENV",
+    "AnonymizerRsError",
+    "LocalAnonymizerFeatureGroup",
+    "PyArrowLocalAnonymizerFeatureGroup",
+    "derive_anon_key",
+]

mloda_pii-0.2.1/mloda_pii/feature_groups/anonymizer_rs/base.py

@@ -0,0 +1,68 @@
+"""Framework-independent API of the local, license-free anonymizer-rs FeatureGroup."""
+
+from __future__ import annotations
+
+import os
+from abc import abstractmethod
+from typing import Any, ClassVar
+
+from mloda.provider import DefaultOptionKeys, FeatureChainParserMixin, FeatureGroup, FeatureSet, property_spec
+from mloda.user import DataType, Feature
+
+from mloda_pii.feature_groups.anonymizer_rs.binary import AnonymizerRsError
+
+PII_KEY_FILE_ENV = "pii_key_file_env"
+
+
+class LocalAnonymizerFeatureGroup(FeatureChainParserMixin, FeatureGroup):
+    """Keyed HMAC-SHA256 token (64 hex) of one string column by the anonymizer-rs binary; no license check.
+
+    String form: ``Feature("email__pii_anonymized", Options(context={"pii_key_file_env": "KEY_FILE"}))``.
+    ``pii_key_file_env`` names the env var holding the path to the master key file (at least 32 bytes).
+    Same tokens as the licensed ``__hmac_sha256_pseudonymized`` group under the key derived by ``derive_anon_key``.
+    """
+
+    PREFIX_PATTERN = r".+__pii_anonymized$"
+    RECOGNITION_ONLY_PATTERN = True
+    MIN_IN_FEATURES = 1
+    MAX_IN_FEATURES = 1
+    BINARY_COMMAND_OVERRIDE: ClassVar[list[str] | None] = None
+    BINARY_TIMEOUT_SECONDS: ClassVar[float | None] = 60.0
+
+    PROPERTY_MAPPING: ClassVar = {
+        # Required and not strict, so no validator can echo it.
+        PII_KEY_FILE_ENV: property_spec("Name of the env var holding the path to the master key file"),
+        DefaultOptionKeys.in_features: property_spec("Source column to anonymize"),
+    }
+
+    @classmethod
+    def _validate_string_match(cls, _feature_name: str, _operation_config: str, _in_feature: str) -> bool:
+        return cls.IN_FEATURE_SEPARATOR not in _in_feature
+
+    @classmethod
+    def return_data_type_rule(cls, feature: Feature) -> DataType | None:
+        return DataType.STRING
+
+    @classmethod
+    def calculate_feature(cls, data: Any, features: FeatureSet) -> Any:
+        by_key_file: dict[str, list[tuple[str, str]]] = {}
+        for feature in features.features:
+            key_file = cls._key_file(feature.options.get(PII_KEY_FILE_ENV))
+            source = cls._extract_single_source_feature(feature)
+            by_key_file.setdefault(key_file, []).append((source, str(feature.name)))
+        for key_file, mappings in by_key_file.items():
+            data = cls._anonymize(data, mappings, key_file)
+        return data
+
+    @classmethod
+    def _key_file(cls, env_name: object) -> str:
+        """The key file path from the named variable; errors never echo the name's value."""
+        path = os.environ.get(env_name, "").strip() if isinstance(env_name, str) and env_name else ""
+        if not path:
+            raise AnonymizerRsError(f"{PII_KEY_FILE_ENV} must name a set env var holding the master key file path")
+        return path
+
+    @classmethod
+    @abstractmethod
+    def _anonymize(cls, data: Any, mappings: list[tuple[str, str]], key_file: str) -> Any:
+        """Append one token column per (source, output) pair; abstract so mloda never selects the base."""

mloda_pii-0.2.1/mloda_pii/feature_groups/anonymizer_rs/binary.py

@@ -0,0 +1,124 @@
+"""Runs the anonymizer-rs ``anonymizer`` binary over Arrow IPC; no mloda imports and no license check."""
+
+from __future__ import annotations
+
+import hashlib
+import hmac
+import json
+import os
+import shutil
+import subprocess  # nosec B404
+import sys
+from collections.abc import Sequence
+from pathlib import Path
+
+import pyarrow as pa
+
+BINARY_ENV = "ANONYMIZER_BINARY"
+BINARY_NAME = "anonymizer"
+MIN_MASTER_KEY_BYTES = 32
+_ANON_INFO = b"anonymizer-rs-anon"
+_MESSAGE_LIMIT = 1024
+
+
+class AnonymizerRsError(ValueError):
+    """The binary was missing, failed, or broke its output contract; never carries cell values."""
+
+    def __init__(self, message: str, code: int | None = None) -> None:
+        super().__init__(message)
+        self.code = code
+
+
+def derive_anon_key(master_key: bytes) -> str:
+    """The 64-hex HMAC key anonymizer-rs derives from a master key (HKDF-SHA256, no salt)."""
+    if len(master_key) < MIN_MASTER_KEY_BYTES:
+        raise AnonymizerRsError(f"the master key must be at least {MIN_MASTER_KEY_BYTES} bytes, got {len(master_key)}")
+    prk = hmac.new(bytes(32), master_key, hashlib.sha256).digest()
+    return hmac.new(prk, _ANON_INFO + b"\x01", hashlib.sha256).hexdigest()
+
+
+def find_binary(override: Sequence[str] | None = None) -> list[str]:
+    """argv prefix: ``override``, else ``$ANONYMIZER_BINARY``, else ``anonymizer`` on PATH."""
+    if override:
+        return list(override)
+    configured = os.environ.get(BINARY_ENV, "")
+    if configured:
+        path = Path(configured)
+        if not (path.is_absolute() and path.is_file() and os.access(path, os.X_OK)):
+            raise AnonymizerRsError(f"{BINARY_ENV} must be an absolute path to an executable file")
+        return [configured]
+    found = shutil.which(BINARY_NAME)
+    if found is None:
+        raise AnonymizerRsError(
+            f"anonymizer-rs binary not found: set {BINARY_ENV} or put '{BINARY_NAME}' on PATH "
+            "(bin/build-anonymizer-rs.sh builds it)"
+        )
+    return [found]
+
+
+def _environment(key_file: str, columns: str) -> dict[str, str]:
+    env = {"PATH": os.environ.get("PATH", ""), "ANONYMIZER_KEY_FILE": key_file, "ANONYMIZER_COLUMNS": columns}
+    if sys.platform == "win32" and "SYSTEMROOT" in os.environ:
+        env["SYSTEMROOT"] = os.environ["SYSTEMROOT"]
+    return env
+
+
+def _ipc_bytes(table: pa.Table) -> bytes:
+    sink = pa.BufferOutputStream()
+    with pa.ipc.new_stream(sink, table.schema) as writer:
+        writer.write_table(table)
+    return bytes(sink.getvalue())
+
+
+def _binary_message(stderr: bytes) -> str:
+    """The ``error`` field of the binary's JSON stderr line, bounded; raw stderr is never echoed."""
+    for line in reversed(stderr.decode("utf-8", errors="replace").strip().splitlines()):
+        try:
+            parsed = json.loads(line)
+        except ValueError:
+            continue
+        error = parsed.get("error") if isinstance(parsed, dict) else None
+        if isinstance(error, str):
+            return error[:_MESSAGE_LIMIT]
+    return "no error message"
+
+
+def anonymize_columns(
+    argv: Sequence[str], columns: Sequence[pa.ChunkedArray], key_file: str, timeout: float | None
+) -> list[pa.ChunkedArray]:
+    """One token column per input column, in one binary call; columns travel under positional names."""
+    if not columns:
+        return []
+    rows = len(columns[0])
+    if rows == 0:
+        return [pa.chunked_array([], pa.string()) for _ in columns]
+    inputs = [f"c{i}" for i in range(len(columns))]
+    outputs = [f"o{i}" for i in range(len(columns))]
+    table = pa.table(dict(zip(inputs, columns, strict=True)))
+    spec = ",".join(f"{i}:{o}" for i, o in zip(inputs, outputs, strict=True))
+    try:
+        proc = subprocess.run(  # nosec B603
+            list(argv),
+            input=_ipc_bytes(table),
+            capture_output=True,
+            env=_environment(key_file, spec),
+            timeout=timeout,
+            check=False,
+        )
+    except subprocess.TimeoutExpired:
+        raise AnonymizerRsError(f"anonymizer-rs binary timed out after {timeout} s") from None
+    except OSError as exc:
+        raise AnonymizerRsError(f"anonymizer-rs binary could not start: {exc.strerror}") from None
+    if proc.returncode != 0:
+        message = _binary_message(proc.stderr)
+        raise AnonymizerRsError(f"anonymizer-rs binary exited {proc.returncode}: {message}", proc.returncode)
+    try:
+        result = pa.ipc.open_stream(proc.stdout).read_all()
+    except (pa.ArrowException, OSError) as exc:
+        raise AnonymizerRsError(f"anonymizer-rs output is not an Arrow IPC stream: {type(exc).__name__}") from None
+    if sorted(result.column_names) != sorted(outputs) or result.num_rows != rows:
+        raise AnonymizerRsError(
+            f"anonymizer-rs output has columns {sorted(result.column_names)} and {result.num_rows} rows, "
+            f"expected {sorted(outputs)} and {rows}"
+        )
+    return [result.column(name).cast(pa.string()) for name in outputs]

mloda_pii-0.2.1/mloda_pii/feature_groups/anonymizer_rs/pyarrow_anonymizer_rs.py

@@ -0,0 +1,36 @@
+"""PyArrow implementation of the local anonymizer-rs FeatureGroup."""
+
+from __future__ import annotations
+
+from typing import Any
+
+import pyarrow as pa
+from mloda.provider import ComputeFramework
+from mloda_plugins.compute_framework.base_implementations.pyarrow.table import PyArrowTable
+
+from mloda_pii.feature_groups.anonymizer_rs.base import LocalAnonymizerFeatureGroup
+from mloda_pii.feature_groups.anonymizer_rs.binary import AnonymizerRsError, anonymize_columns, find_binary
+
+
+class PyArrowLocalAnonymizerFeatureGroup(LocalAnonymizerFeatureGroup):
+    @classmethod
+    def compute_framework_rule(cls) -> set[type[ComputeFramework]]:
+        return {PyArrowTable}
+
+    @classmethod
+    def _anonymize(cls, data: Any, mappings: list[tuple[str, str]], key_file: str) -> Any:
+        columns = []
+        for source, output in mappings:
+            if source not in data.column_names:
+                raise AnonymizerRsError(f"Source column {source!r} not found; available: {sorted(data.column_names)}.")
+            if output in data.column_names:
+                raise AnonymizerRsError(f"Output column {output!r} already exists.")
+            column = data.column(source)
+            if not (pa.types.is_string(column.type) or pa.types.is_large_string(column.type)):
+                raise AnonymizerRsError(f"Source column {source!r} must be a string column, got {column.type}.")
+            columns.append(column.cast(pa.string()))
+        argv = find_binary(cls.BINARY_COMMAND_OVERRIDE)
+        tokens = anonymize_columns(argv, columns, key_file, cls.BINARY_TIMEOUT_SECONDS)
+        for (_, output), token_column in zip(mappings, tokens, strict=True):
+            data = data.append_column(output, token_column)
+        return data

mloda_pii-0.2.1/mloda_pii/feature_groups/manifest.py

@@ -0,0 +1,8 @@
+"""Entry-point manifest: concrete FeatureGroup classes this package exports to mloda discovery."""
+
+from mloda.provider import FeatureGroup
+
+from mloda_pii.feature_groups.anonymizer_rs import PyArrowLocalAnonymizerFeatureGroup
+
+# Referenced by the "mloda.feature_groups" entry point in pyproject.toml. Concrete classes only.
+FEATURE_GROUPS: list[type[FeatureGroup]] = [PyArrowLocalAnonymizerFeatureGroup]