open-code-review-toolkit 0.14.5__tar.gz → 0.14.6__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/PKG-INFO +1 -1
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/pyproject.toml +1 -1
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/_version.py +2 -2
- open_code_review_toolkit-0.14.6/src/ocr_toolkit/common/literal_context.py +147 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/redaction.py +28 -9
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/value_boundaries.py +7 -1
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/admission.py +112 -11
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/credentials.py +27 -13
- open_code_review_toolkit-0.14.6/src/ocr_toolkit/context/dlp/email.py +64 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/markup.py +91 -10
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/presentation.py +10 -1
- open_code_review_toolkit-0.14.6/src/ocr_toolkit/context/dlp/projection.py +212 -0
- open_code_review_toolkit-0.14.6/src/ocr_toolkit/context/dlp/result_prose.py +186 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/native_capabilities.py +1 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/ocr_result.py +57 -1
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/preflight.py +2 -1
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/local.py +4 -8
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/model.py +3 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/result.py +1 -34
- open_code_review_toolkit-0.14.6/src/ocr_toolkit/reporting/warnings.py +41 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_runner.py +173 -41
- open_code_review_toolkit-0.14.5/src/ocr_toolkit/context/dlp/email.py +0 -50
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/.gitignore +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/LICENSE +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/README.md +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/cli.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/filesystem.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/git.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/language.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/common/markdown.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/config_writer.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/configure.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/broker.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/ci_outcomes.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/code_text.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/datetimes.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/decoding.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/dlp/numeric.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/mcp.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/policy.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/recognizers.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/context/store.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/admission.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/collector.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/lifecycle.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/mcp.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/persistence.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/diagrams/sources.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/__main__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/cli.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/corpus.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/limits.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/proof.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/report.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/scanner.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/state.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/dlp_audit/worker_process.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/__main__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/actions.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/artifacts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/categorize.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collect.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collectors/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collectors/graphs.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collectors/orchestration.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collectors/projections.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collectors/registry.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/collectors/sources.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/coverage.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/ansible/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/ansible/requirements.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/ansible/topology.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/go.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/javascript.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/php.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/ecosystems/python.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/detection.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/providers/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/providers/frontend.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/providers/go.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/providers/php.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/providers/python.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/registry.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/schema.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/frameworks/templates.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/gitlink_projection.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/gitlinks.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/infrastructure.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/invocation.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/mcp.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/model.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/opaque_code.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/decisions.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/guidance.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/registry.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/schema.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/policy/scopes.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/project.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/repository.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/review_context.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/review_range.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/source_reader.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/source_syntax.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/store/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/store/atomic.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/store/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/store/core.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/store/readback.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/store/values.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/text_contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/text_excerpts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/evidence/text_metadata.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/admission.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/contracts.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/gateway.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/registry.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/relay.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/runtime.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/federation/transport.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/local_chatgpt.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/mcp_config.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/__main__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/approval.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/comments.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/formatting.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/freshness.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/gitlab.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/gitlab_approval.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/markers.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/payloads.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/reconciliation.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/result.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/settings.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/snapshot.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/source_suggestions.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/suggestions.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/transaction.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/posting/workflow.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/pre_execution.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/provider_config.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/provider_failure.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/gitlab.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/gitlab_ci.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/gitlab_context.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/gitlab_discussions.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/gitlab_identity.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/gitlab_remediation.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/providers/local_config.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/py.typed +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/__init__.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/diagram.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/diagram_layout.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/dlp.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/federation.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/metadata.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/outcome.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/sections.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/text.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/reporting/usage.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/repository_coverage.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/result_contract.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/result_usage.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_debug.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_identity.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_progress.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_range.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_receipt.py +0 -0
- {open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/review_trace.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.5
|
|
2
2
|
Name: open-code-review-toolkit
|
|
3
|
-
Version: 0.14.
|
|
3
|
+
Version: 0.14.6
|
|
4
4
|
Summary: Unofficial GitLab CI integration layer for Open Code Review
|
|
5
5
|
Project-URL: Homepage, https://github.com/xeonvs/open-code-review-toolkit
|
|
6
6
|
Project-URL: Repository, https://github.com/xeonvs/open-code-review-toolkit
|
{open_code_review_toolkit-0.14.5 → open_code_review_toolkit-0.14.6}/src/ocr_toolkit/_version.py
RENAMED
|
@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
|
|
|
18
18
|
commit_id: str | None
|
|
19
19
|
__commit_id__: str | None
|
|
20
20
|
|
|
21
|
-
__version__ = version = '0.14.
|
|
22
|
-
__version_tuple__ = version_tuple = (0, 14,
|
|
21
|
+
__version__ = version = '0.14.6'
|
|
22
|
+
__version_tuple__ = version_tuple = (0, 14, 6)
|
|
23
23
|
|
|
24
24
|
__commit_id__ = commit_id = None
|
|
@@ -0,0 +1,147 @@
|
|
|
1
|
+
"""Bounded closed-data proof, without executing or qualifying source code."""
|
|
2
|
+
|
|
3
|
+
from __future__ import annotations
|
|
4
|
+
|
|
5
|
+
import ast
|
|
6
|
+
import json
|
|
7
|
+
import re
|
|
8
|
+
|
|
9
|
+
_MAX_CHARS = 2_000_000
|
|
10
|
+
_MAX_TOKENS = 32_768
|
|
11
|
+
_MAX_DEPTH = 20
|
|
12
|
+
_PUNCTUATION = re.compile(r"[^\w\s]|\d+", re.UNICODE)
|
|
13
|
+
_DATA_SCALARS = frozenset({"true", "false", "null", "none"})
|
|
14
|
+
_DATA_NODES = (
|
|
15
|
+
ast.Expression,
|
|
16
|
+
ast.Module,
|
|
17
|
+
ast.Expr,
|
|
18
|
+
ast.Assign,
|
|
19
|
+
ast.AnnAssign,
|
|
20
|
+
ast.Constant,
|
|
21
|
+
ast.List,
|
|
22
|
+
ast.Tuple,
|
|
23
|
+
ast.Set,
|
|
24
|
+
ast.Dict,
|
|
25
|
+
ast.Load,
|
|
26
|
+
ast.Store,
|
|
27
|
+
ast.UnaryOp,
|
|
28
|
+
ast.UAdd,
|
|
29
|
+
ast.USub,
|
|
30
|
+
)
|
|
31
|
+
|
|
32
|
+
|
|
33
|
+
class LiteralContext:
|
|
34
|
+
"""Cache one bounded quote inventory and a complete data-grammar proof.
|
|
35
|
+
|
|
36
|
+
Only literal values, containers and named literal fields qualify. Unknown
|
|
37
|
+
calls, references, operators or literal syntaxes never acquire authority
|
|
38
|
+
from the number of recognized strings. This is not a source-language proof.
|
|
39
|
+
"""
|
|
40
|
+
|
|
41
|
+
def __init__(self, text: str) -> None:
|
|
42
|
+
self.text = text
|
|
43
|
+
self._quotes: dict[tuple[int, int], bool] | None = None
|
|
44
|
+
self._data: bool | None = None
|
|
45
|
+
self._valid = True
|
|
46
|
+
|
|
47
|
+
def proves_carrier(self, start: int, end: int) -> bool:
|
|
48
|
+
if self._quotes is None:
|
|
49
|
+
self._inventory()
|
|
50
|
+
assert self._quotes is not None
|
|
51
|
+
if not self._valid or (start, end) not in self._quotes:
|
|
52
|
+
return False
|
|
53
|
+
if self._data is None:
|
|
54
|
+
self._data = self._literal_data()
|
|
55
|
+
return self._data
|
|
56
|
+
|
|
57
|
+
def _inventory(self) -> None:
|
|
58
|
+
self._quotes = {}
|
|
59
|
+
text = self.text
|
|
60
|
+
if len(text) > _MAX_CHARS:
|
|
61
|
+
self._valid = False
|
|
62
|
+
return
|
|
63
|
+
cursor = 0
|
|
64
|
+
while cursor < len(text):
|
|
65
|
+
quote = text[cursor]
|
|
66
|
+
if quote not in "\"'":
|
|
67
|
+
cursor += 1
|
|
68
|
+
continue
|
|
69
|
+
start = cursor
|
|
70
|
+
cursor += 1
|
|
71
|
+
while cursor < len(text) and text[cursor] != quote:
|
|
72
|
+
cursor += 2 if text[cursor] == "\\" else 1
|
|
73
|
+
if cursor >= len(text) or text.startswith(quote * 3, start):
|
|
74
|
+
self._valid = False
|
|
75
|
+
return
|
|
76
|
+
cursor += 1
|
|
77
|
+
self._quotes[start, cursor] = text[start + 1 : cursor - 1] not in {"", "***"}
|
|
78
|
+
if len(self._quotes) > _MAX_TOKENS:
|
|
79
|
+
self._valid = False
|
|
80
|
+
return
|
|
81
|
+
|
|
82
|
+
def _literal_data(self) -> bool:
|
|
83
|
+
depth = 0
|
|
84
|
+
quotes = iter(self._quotes or ())
|
|
85
|
+
quoted = next(quotes, None)
|
|
86
|
+
for count, match in enumerate(_PUNCTUATION.finditer(self.text), 1):
|
|
87
|
+
if count > _MAX_TOKENS:
|
|
88
|
+
return False
|
|
89
|
+
while quoted is not None and match.start() >= quoted[1]:
|
|
90
|
+
quoted = next(quotes, None)
|
|
91
|
+
if quoted is not None and quoted[0] <= match.start() < quoted[1]:
|
|
92
|
+
continue
|
|
93
|
+
# No parser-discarded comments or unbounded numeric conversion.
|
|
94
|
+
if match[0] == "#" or len(match[0]) > 4096:
|
|
95
|
+
return False
|
|
96
|
+
if match[0] in "([{":
|
|
97
|
+
depth += 1
|
|
98
|
+
if depth > _MAX_DEPTH:
|
|
99
|
+
return False
|
|
100
|
+
elif match[0] in ")]}":
|
|
101
|
+
depth -= 1
|
|
102
|
+
if depth < 0:
|
|
103
|
+
return False
|
|
104
|
+
try:
|
|
105
|
+
json.loads(self.text)
|
|
106
|
+
return True
|
|
107
|
+
except (ValueError, RecursionError):
|
|
108
|
+
pass
|
|
109
|
+
try:
|
|
110
|
+
try:
|
|
111
|
+
tree = ast.parse(self.text.strip(), mode="eval")
|
|
112
|
+
except SyntaxError:
|
|
113
|
+
tree = ast.parse(self.text.strip(), mode="exec")
|
|
114
|
+
except (SyntaxError, ValueError, RecursionError, UnicodeError):
|
|
115
|
+
return False
|
|
116
|
+
targets: set[int] = set()
|
|
117
|
+
if isinstance(tree, ast.Module):
|
|
118
|
+
for statement in tree.body:
|
|
119
|
+
if isinstance(statement, ast.Assign):
|
|
120
|
+
if len(statement.targets) != 1 or not isinstance(
|
|
121
|
+
statement.targets[0], ast.Name
|
|
122
|
+
):
|
|
123
|
+
return False
|
|
124
|
+
targets.add(id(statement.targets[0]))
|
|
125
|
+
elif isinstance(statement, ast.AnnAssign):
|
|
126
|
+
if statement.value is not None or not isinstance(statement.target, ast.Name):
|
|
127
|
+
return False
|
|
128
|
+
targets.add(id(statement.target))
|
|
129
|
+
elif not isinstance(statement, ast.Expr):
|
|
130
|
+
return False
|
|
131
|
+
strings = 0
|
|
132
|
+
for count, node in enumerate(ast.walk(tree), 1):
|
|
133
|
+
if count > _MAX_TOKENS:
|
|
134
|
+
return False
|
|
135
|
+
if isinstance(node, ast.Name):
|
|
136
|
+
if id(node) not in targets and node.id.casefold() not in _DATA_SCALARS:
|
|
137
|
+
return False
|
|
138
|
+
continue
|
|
139
|
+
if not isinstance(node, _DATA_NODES):
|
|
140
|
+
return False
|
|
141
|
+
if isinstance(node, ast.UnaryOp) and (
|
|
142
|
+
not isinstance(node.operand, ast.Constant)
|
|
143
|
+
or type(node.operand.value) not in {int, float, complex}
|
|
144
|
+
):
|
|
145
|
+
return False
|
|
146
|
+
strings += isinstance(node, ast.Constant) and isinstance(node.value, (str, bytes))
|
|
147
|
+
return strings == len(self._quotes or ())
|
|
@@ -10,6 +10,7 @@ from bisect import bisect_right
|
|
|
10
10
|
from collections.abc import Iterator
|
|
11
11
|
from typing import Any
|
|
12
12
|
|
|
13
|
+
from ocr_toolkit.common.literal_context import LiteralContext
|
|
13
14
|
from ocr_toolkit.common.value_boundaries import complete_carrier as _complete_credential_carrier
|
|
14
15
|
from ocr_toolkit.common.value_boundaries import complete_value
|
|
15
16
|
|
|
@@ -225,7 +226,9 @@ def redact_env_secret_values(text: str, *, values: tuple[str, ...] | None = None
|
|
|
225
226
|
return redacted
|
|
226
227
|
|
|
227
228
|
|
|
228
|
-
def _complete_credential_value(
|
|
229
|
+
def _complete_credential_value(
|
|
230
|
+
text: str, end: int, *, start: int | None = None, context: LiteralContext | None = None
|
|
231
|
+
) -> bool:
|
|
229
232
|
"""Require a terminal value, not a prefix of concatenation or a call.
|
|
230
233
|
|
|
231
234
|
One unambiguous matching outer quote may close its carrier. Arbitrary
|
|
@@ -246,7 +249,7 @@ def _complete_credential_value(text: str, end: int, *, start: int | None = None)
|
|
|
246
249
|
text[end - 1] not in "\"'" and not any(c in prefix for c in (quote, "\\", "\r", "\n"))
|
|
247
250
|
)
|
|
248
251
|
if carrier and single and text[end - 1] != quote:
|
|
249
|
-
return _complete_credential_carrier(text, cursor + 1)
|
|
252
|
+
return _complete_credential_carrier(text, cursor + 1, start=opening, context=context)
|
|
250
253
|
return complete_value(text, end)
|
|
251
254
|
|
|
252
255
|
|
|
@@ -255,12 +258,13 @@ def masked_credential_spans(text: str) -> tuple[tuple[int, int], ...]:
|
|
|
255
258
|
|
|
256
259
|
if "***" not in text:
|
|
257
260
|
return ()
|
|
261
|
+
context = LiteralContext(text)
|
|
258
262
|
spans = {
|
|
259
263
|
match.span("value")
|
|
260
264
|
for pattern, _replacement in (*_AUTH_RULES, *_ASSIGNMENT_RULES)
|
|
261
265
|
for match in pattern.finditer(text)
|
|
262
266
|
if match["value"] == "***"
|
|
263
|
-
and _complete_credential_value(text, match.end(), start=match.start())
|
|
267
|
+
and _complete_credential_value(text, match.end(), start=match.start(), context=context)
|
|
264
268
|
}
|
|
265
269
|
# URI carriers have their own proven boundaries. A closing carrier quote
|
|
266
270
|
# must not require granting the same authority to arbitrary quote runs.
|
|
@@ -294,13 +298,14 @@ def _credential_material(text: str) -> bool:
|
|
|
294
298
|
if parts:
|
|
295
299
|
parts.append(text[previous:])
|
|
296
300
|
text = "".join(parts)
|
|
301
|
+
context = LiteralContext(text)
|
|
297
302
|
safe_fields: set[tuple[int, int]] = set()
|
|
298
303
|
for pattern, _replacement in (*_AUTH_RULES, *_ASSIGNMENT_RULES):
|
|
299
304
|
if "suffix" not in pattern.groupindex:
|
|
300
305
|
continue
|
|
301
306
|
for match in pattern.finditer(text):
|
|
302
307
|
if match["value"] not in {"", "***"} or not _complete_credential_value(
|
|
303
|
-
text, match.end(), start=match.start()
|
|
308
|
+
text, match.end(), start=match.start(), context=context
|
|
304
309
|
):
|
|
305
310
|
return True
|
|
306
311
|
safe_fields.add(match.span())
|
|
@@ -313,7 +318,7 @@ def _credential_material(text: str) -> bool:
|
|
|
313
318
|
if index >= 0 and match.end() <= ordered_fields[index][1]:
|
|
314
319
|
continue
|
|
315
320
|
if match["value"] in {"", "***"} and _complete_credential_value(
|
|
316
|
-
text, match.end(), start=match.start()
|
|
321
|
+
text, match.end(), start=match.start(), context=context
|
|
317
322
|
):
|
|
318
323
|
continue
|
|
319
324
|
return True
|
|
@@ -334,6 +339,7 @@ _SOURCE_BREAK = re.compile(r"[\s\"'<>]")
|
|
|
334
339
|
def _material_queries(text: str) -> Iterator[tuple[re.Match[str], int, bool, bool]]:
|
|
335
340
|
"""Walk complete sensitive values without granting authority to their tails."""
|
|
336
341
|
|
|
342
|
+
context = LiteralContext(text)
|
|
337
343
|
cursor = 0
|
|
338
344
|
while (query := _QUERY_KEY_RE.search(text, cursor)) is not None:
|
|
339
345
|
cursor = query.end()
|
|
@@ -366,7 +372,9 @@ def _material_queries(text: str) -> Iterator[tuple[re.Match[str], int, bool, boo
|
|
|
366
372
|
or _URI_SCHEME.search(window, token_start) is not None
|
|
367
373
|
)
|
|
368
374
|
closing = text.find(carrier["quote"], start) if carrier is not None else -1
|
|
369
|
-
if closing == start and not _complete_credential_carrier(
|
|
375
|
+
if closing == start and not _complete_credential_carrier(
|
|
376
|
+
text, closing + 1, start=opening, context=context
|
|
377
|
+
):
|
|
370
378
|
# An opening query-value quote cannot also prove the carrier's
|
|
371
379
|
# end. Ambiguous nested quoting retains ordinary checks.
|
|
372
380
|
closing = -1
|
|
@@ -380,7 +388,9 @@ def _material_queries(text: str) -> Iterator[tuple[re.Match[str], int, bool, boo
|
|
|
380
388
|
closed = closing >= 0 and end == closing
|
|
381
389
|
empty = start == end or _EMPTY_QUERY_VALUE_RE.fullmatch(text, start, end) is not None
|
|
382
390
|
empty = empty and not ambiguous_carrier
|
|
383
|
-
if closed and not _complete_credential_carrier(
|
|
391
|
+
if closed and not _complete_credential_carrier(
|
|
392
|
+
text, end + 1, start=opening, context=context
|
|
393
|
+
):
|
|
384
394
|
empty = False
|
|
385
395
|
if (
|
|
386
396
|
end < len(text)
|
|
@@ -484,6 +494,9 @@ def sanitize_ocr_value(value: Any) -> Any:
|
|
|
484
494
|
def redact_url_userinfo_only(value: str) -> str:
|
|
485
495
|
"""Redact only URL userinfo, leaving path/query text unchanged."""
|
|
486
496
|
|
|
497
|
+
if "://" not in value or "@" not in value:
|
|
498
|
+
return value
|
|
499
|
+
|
|
487
500
|
def replace(match: re.Match[str]) -> str:
|
|
488
501
|
prefix, userinfo = match.group(1), match.group(2)
|
|
489
502
|
if not _is_url_userinfo_candidate(userinfo):
|
|
@@ -491,7 +504,10 @@ def redact_url_userinfo_only(value: str) -> str:
|
|
|
491
504
|
return f"{prefix}***@"
|
|
492
505
|
|
|
493
506
|
return re.sub(
|
|
494
|
-
|
|
507
|
+
# Attempt once per scheme-character run, not at every letter in a long
|
|
508
|
+
# near match. A non-letter prefix stays byte-for-byte in the replacement,
|
|
509
|
+
# just as it stayed outside the old unanchored match.
|
|
510
|
+
r"(?i)(?<![a-z0-9+.-])([0-9+.-]*[a-z][a-z0-9+.-]*://)([^/?#@]+)@"
|
|
495
511
|
r"(?=[^/?#\s@]+(?:[/?#]|[\s,;.!)]|$))",
|
|
496
512
|
replace,
|
|
497
513
|
value,
|
|
@@ -505,7 +521,10 @@ def _is_url_userinfo_candidate(userinfo: str) -> bool:
|
|
|
505
521
|
char for char in userinfo if unicodedata.category(char) not in {"Cc", "Cf", "Zl", "Zp"}
|
|
506
522
|
)
|
|
507
523
|
decoded = urllib.parse.unquote(normalized)
|
|
508
|
-
|
|
524
|
+
# The URI grammar excludes literal delimiters from this capture. Encoded
|
|
525
|
+
# delimiters remain valid userinfo data and cannot prove absence of a
|
|
526
|
+
# credential, including for dotless/internal host names.
|
|
527
|
+
if not decoded:
|
|
509
528
|
return False
|
|
510
529
|
if any(char.isspace() for char in userinfo):
|
|
511
530
|
return ":" in decoded
|
|
@@ -4,6 +4,8 @@ from __future__ import annotations
|
|
|
4
4
|
|
|
5
5
|
import re
|
|
6
6
|
|
|
7
|
+
from ocr_toolkit.common.literal_context import LiteralContext
|
|
8
|
+
|
|
7
9
|
MAX_TAIL_CHARS = 4096
|
|
8
10
|
_LINE_END = re.compile(r"[\r\n\x85\u2028\u2029]")
|
|
9
11
|
_FIELD_HEAD = re.compile(
|
|
@@ -81,9 +83,13 @@ def complete_field(text: str, end: int) -> bool:
|
|
|
81
83
|
)
|
|
82
84
|
|
|
83
85
|
|
|
84
|
-
def complete_carrier(
|
|
86
|
+
def complete_carrier(
|
|
87
|
+
text: str, end: int, *, start: int, context: LiteralContext | None = None
|
|
88
|
+
) -> bool:
|
|
85
89
|
"""Check a source carrier through trivia, closers and trailing separators."""
|
|
86
90
|
|
|
91
|
+
if not (context or LiteralContext(text)).proves_carrier(start, end):
|
|
92
|
+
return False
|
|
87
93
|
limit = min(len(text), end + MAX_TAIL_CHARS)
|
|
88
94
|
cursor = end
|
|
89
95
|
while cursor < limit:
|
|
@@ -34,6 +34,11 @@ from ocr_toolkit.context.dlp.numeric import (
|
|
|
34
34
|
phone_markup_view,
|
|
35
35
|
)
|
|
36
36
|
from ocr_toolkit.context.dlp.presentation import canonical_whitespace, has_unsafe_controls
|
|
37
|
+
from ocr_toolkit.context.dlp.projection import (
|
|
38
|
+
UnsafeDestination,
|
|
39
|
+
decoded_markup_material,
|
|
40
|
+
project_prose,
|
|
41
|
+
)
|
|
37
42
|
from ocr_toolkit.result_contract import FORBIDDEN_SOURCE_CLASSES
|
|
38
43
|
|
|
39
44
|
MARKDOWN_DEST_RE = re.compile(
|
|
@@ -234,6 +239,17 @@ def _literal_contact_view(value: str, *, visible: bool, formatting: bool = True)
|
|
|
234
239
|
return _source_normalize(normalized)
|
|
235
240
|
|
|
236
241
|
|
|
242
|
+
def _markup_phone(value: str, *, literal: bool) -> bool:
|
|
243
|
+
markup = interpret_markdown(phone_markup_view(value, formatting=False))
|
|
244
|
+
# Visible formatting is interpreted by the parser. Hidden HTML material
|
|
245
|
+
# needs an independent sweep; stripping it or joining its digits would
|
|
246
|
+
# silently make an obfuscated contact look like an opaque identifier.
|
|
247
|
+
return (not literal and _contains_phone(_literal_normalize(markup.contacts))) or any(
|
|
248
|
+
_contains_phone(phone_markup_view(_literal_normalize(fragment)))
|
|
249
|
+
for fragment in markup.hidden
|
|
250
|
+
)
|
|
251
|
+
|
|
252
|
+
|
|
237
253
|
_P = ParamSpec("_P")
|
|
238
254
|
|
|
239
255
|
|
|
@@ -249,7 +265,7 @@ def _bounded_check(function: Callable[_P, DLPResult]) -> Callable[_P, DLPResult]
|
|
|
249
265
|
|
|
250
266
|
|
|
251
267
|
@_bounded_check
|
|
252
|
-
def
|
|
268
|
+
def _check_material(
|
|
253
269
|
value: object,
|
|
254
270
|
*,
|
|
255
271
|
budgets: TextBudgets,
|
|
@@ -323,6 +339,19 @@ def check_text(
|
|
|
323
339
|
("source", source, source_transform(decoded_semantic)),
|
|
324
340
|
)
|
|
325
341
|
)
|
|
342
|
+
for text in decoded_markup_material(normalized):
|
|
343
|
+
views.append(("decoded_markup", text, _semantic_secret_text(text)))
|
|
344
|
+
compatible_markup = bounded_nfkc(text)
|
|
345
|
+
if compatible_markup != text:
|
|
346
|
+
# Composed decoding/normalization carries no source delimiter
|
|
347
|
+
# proof. It remains an independent complete material view.
|
|
348
|
+
views.append(
|
|
349
|
+
(
|
|
350
|
+
"compatible_markup",
|
|
351
|
+
compatible_markup,
|
|
352
|
+
_semantic_secret_text(compatible_markup),
|
|
353
|
+
)
|
|
354
|
+
)
|
|
326
355
|
mermaid = _mermaid_decode(decoded)
|
|
327
356
|
has_mermaid = mermaid != decoded
|
|
328
357
|
if has_mermaid:
|
|
@@ -389,16 +418,21 @@ def check_text(
|
|
|
389
418
|
if candidate in seen_pii:
|
|
390
419
|
continue
|
|
391
420
|
seen_pii.add(candidate)
|
|
392
|
-
|
|
393
|
-
|
|
394
|
-
|
|
395
|
-
|
|
396
|
-
|
|
397
|
-
|
|
398
|
-
|
|
399
|
-
|
|
400
|
-
|
|
421
|
+
email_transform = (
|
|
422
|
+
bounded_nfkc
|
|
423
|
+
if representation == "compatible"
|
|
424
|
+
else _html_decode
|
|
425
|
+
if representation == "decoded"
|
|
426
|
+
else None
|
|
427
|
+
)
|
|
428
|
+
if contains_email(
|
|
429
|
+
candidate,
|
|
430
|
+
markdown=not literal and (candidate == normalized or email_transform is not None),
|
|
431
|
+
original=normalized if email_transform is not None else None,
|
|
432
|
+
transform=email_transform,
|
|
401
433
|
):
|
|
434
|
+
return DLPResult(False, None, "pii", f"email:{representation}")
|
|
435
|
+
if _contains_phone(candidate) or _markup_phone(candidate, literal=literal):
|
|
402
436
|
return DLPResult(False, None, "pii", f"phone:{representation}")
|
|
403
437
|
# Secret/email deobfuscation remains independent of literal rendering.
|
|
404
438
|
# Strong contact signals also retain the stricter arbitrary-format sweep.
|
|
@@ -427,13 +461,24 @@ def check_text(
|
|
|
427
461
|
return DLPResult(False, None, "pii", "phone:literal_markup")
|
|
428
462
|
matcher = forbidden_matcher or ForbiddenMatcher.compile(forbidden)
|
|
429
463
|
matcher_reason, source_classes = matcher.match_details(normalized)
|
|
464
|
+
for name, candidate, _secret in views:
|
|
465
|
+
if name == "decoded_markup":
|
|
466
|
+
reason, classes = matcher.match_details(candidate)
|
|
467
|
+
matcher_reason = matcher_reason or reason
|
|
468
|
+
source_classes = tuple(sorted(set(source_classes) | set(classes)))
|
|
430
469
|
if has_mermaid:
|
|
431
470
|
mermaid_reason, mermaid_sources = matcher.match_details(mermaid)
|
|
432
471
|
matcher_reason = matcher_reason or mermaid_reason
|
|
433
472
|
source_classes = tuple(sorted(set(source_classes) | set(mermaid_sources)))
|
|
434
473
|
if matcher_reason:
|
|
435
474
|
return DLPResult(False, None, matcher_reason, "forbidden_matcher", source_classes)
|
|
436
|
-
if
|
|
475
|
+
if (
|
|
476
|
+
publication
|
|
477
|
+
and not (allow_opaque_code and sink == TextSink.LITERAL_CODE)
|
|
478
|
+
and has_unsafe_controls(decoded, display=True)
|
|
479
|
+
):
|
|
480
|
+
return DLPResult(False, None, "laundering", "display_control")
|
|
481
|
+
if publication and sink == TextSink.LITERAL_CODE and not allow_opaque_code:
|
|
437
482
|
laundering = (
|
|
438
483
|
"markdown_destination"
|
|
439
484
|
if MARKDOWN_DEST_RE.search(normalized)
|
|
@@ -451,3 +496,59 @@ def check_text(
|
|
|
451
496
|
if laundering is not None:
|
|
452
497
|
return DLPResult(False, None, "laundering", laundering)
|
|
453
498
|
return DLPResult(True, normalized, "admitted")
|
|
499
|
+
|
|
500
|
+
|
|
501
|
+
@_bounded_check
|
|
502
|
+
def check_text(
|
|
503
|
+
value: object,
|
|
504
|
+
*,
|
|
505
|
+
budgets: TextBudgets,
|
|
506
|
+
publication: bool = False,
|
|
507
|
+
forbidden: tuple[str | ForbiddenValue, ...] = (),
|
|
508
|
+
forbidden_matcher: ForbiddenMatcher | None = None,
|
|
509
|
+
allow_horizontal_tabs: bool = False,
|
|
510
|
+
allow_opaque_code: bool = False,
|
|
511
|
+
code_context: CodeTextContext | None = None,
|
|
512
|
+
sink: TextSink = TextSink.PROSE,
|
|
513
|
+
enabled: bool | None = None,
|
|
514
|
+
) -> DLPResult:
|
|
515
|
+
"""Admit all material, prepare prose, and independently admit the prepared text."""
|
|
516
|
+
|
|
517
|
+
checked = _check_material(
|
|
518
|
+
value,
|
|
519
|
+
budgets=budgets,
|
|
520
|
+
publication=publication,
|
|
521
|
+
forbidden=forbidden,
|
|
522
|
+
forbidden_matcher=forbidden_matcher,
|
|
523
|
+
allow_horizontal_tabs=allow_horizontal_tabs,
|
|
524
|
+
allow_opaque_code=allow_opaque_code,
|
|
525
|
+
code_context=code_context,
|
|
526
|
+
sink=sink,
|
|
527
|
+
enabled=enabled,
|
|
528
|
+
)
|
|
529
|
+
if (
|
|
530
|
+
not checked.admitted
|
|
531
|
+
or checked.reason == "disabled"
|
|
532
|
+
or not publication
|
|
533
|
+
or sink != TextSink.PROSE
|
|
534
|
+
):
|
|
535
|
+
return checked
|
|
536
|
+
assert checked.text is not None
|
|
537
|
+
try:
|
|
538
|
+
prepared = project_prose(checked.text, budgets)
|
|
539
|
+
except UnsafeDestination:
|
|
540
|
+
return DLPResult(False, None, "laundering", "markdown_destination")
|
|
541
|
+
if prepared != checked.text:
|
|
542
|
+
material = _check_material(
|
|
543
|
+
prepared,
|
|
544
|
+
budgets=budgets,
|
|
545
|
+
publication=True,
|
|
546
|
+
forbidden=forbidden,
|
|
547
|
+
forbidden_matcher=forbidden_matcher,
|
|
548
|
+
allow_horizontal_tabs=allow_horizontal_tabs,
|
|
549
|
+
sink=TextSink.PROSE,
|
|
550
|
+
enabled=enabled,
|
|
551
|
+
)
|
|
552
|
+
if not material.admitted:
|
|
553
|
+
return material
|
|
554
|
+
return DLPResult(True, prepared, checked.reason)
|
|
@@ -4,8 +4,9 @@ from __future__ import annotations
|
|
|
4
4
|
|
|
5
5
|
import re
|
|
6
6
|
|
|
7
|
+
from ocr_toolkit.common.literal_context import LiteralContext
|
|
7
8
|
from ocr_toolkit.common.redaction import SENSITIVE_NAMED_KEY_PATTERN
|
|
8
|
-
from ocr_toolkit.common.value_boundaries import complete_field
|
|
9
|
+
from ocr_toolkit.common.value_boundaries import complete_carrier, complete_field
|
|
9
10
|
from ocr_toolkit.context.dlp.decoding import RepresentationLimit, _technical_decode, bounded_nfkc
|
|
10
11
|
|
|
11
12
|
MAX_CREDENTIAL_FIELDS = 4096
|
|
@@ -26,10 +27,11 @@ _ABSENT_FIELD = re.compile(
|
|
|
26
27
|
rf"[ \t]*[:=][ \t]*(?:{_REFERENCE}|\*\*\*|"
|
|
27
28
|
rf"(?P<value_quote>['\"])(?:{_REFERENCE}|\*\*\*|)(?P=value_quote))[ \t]*"
|
|
28
29
|
)
|
|
30
|
+
_CARRIER_HINT = re.compile(SENSITIVE_NAMED_KEY_PATTERN, re.IGNORECASE)
|
|
29
31
|
|
|
30
32
|
|
|
31
|
-
def
|
|
32
|
-
"""Qualify
|
|
33
|
+
def _absent_carriers(value: str) -> tuple[tuple[int, int, str], ...]:
|
|
34
|
+
"""Qualify complete ordinary quoted field carriers in either representation.
|
|
33
35
|
|
|
34
36
|
Decoding a quoted encoded field as a whole can manufacture a quote run by
|
|
35
37
|
joining its value quotes to the outer carrier. No filename or language
|
|
@@ -38,9 +40,16 @@ def _encoded_absent_carriers(value: str) -> tuple[tuple[int, int, str], ...]:
|
|
|
38
40
|
qualify. Surrounding source remains subject to ordinary continuation checks.
|
|
39
41
|
"""
|
|
40
42
|
|
|
41
|
-
if
|
|
43
|
+
if not any(quote in value for quote in "\"'") or (
|
|
44
|
+
"=" not in value and ":" not in value and not any(c in value for c in "#&")
|
|
45
|
+
):
|
|
46
|
+
return ()
|
|
47
|
+
if _CARRIER_HINT.search(value) is None and (
|
|
48
|
+
_CARRIER_HINT.search(bounded_nfkc(_technical_decode(value))) is None
|
|
49
|
+
):
|
|
42
50
|
return ()
|
|
43
51
|
spans = []
|
|
52
|
+
context = LiteralContext(value)
|
|
44
53
|
cursor = 0
|
|
45
54
|
while cursor < len(value):
|
|
46
55
|
quote = value[cursor]
|
|
@@ -62,18 +71,20 @@ def _encoded_absent_carriers(value: str) -> tuple[tuple[int, int, str], ...]:
|
|
|
62
71
|
if cursor < len(value) and value[cursor] == quote:
|
|
63
72
|
continue
|
|
64
73
|
payload = value[start:end]
|
|
65
|
-
if "#" not in payload and "&" not in payload:
|
|
66
|
-
continue
|
|
67
74
|
decoded = bounded_nfkc(_technical_decode(payload))
|
|
68
75
|
field = _ABSENT_FIELD.fullmatch(decoded)
|
|
69
|
-
if
|
|
76
|
+
if field is not None:
|
|
70
77
|
if len(spans) >= MAX_CREDENTIAL_FIELDS:
|
|
71
78
|
raise RepresentationLimit("credential carrier budget")
|
|
72
79
|
key = field["key"][1:-1] if field["key_quote"] else field["key"]
|
|
73
|
-
#
|
|
74
|
-
#
|
|
75
|
-
|
|
76
|
-
|
|
80
|
+
# Removing a binding is safe only after the original complete
|
|
81
|
+
# carrier proves absence. Unproved contexts retain the binding.
|
|
82
|
+
replacement = (
|
|
83
|
+
quote + key + quote
|
|
84
|
+
if complete_carrier(value, cursor, start=start - 1, context=context)
|
|
85
|
+
else "'" + key + '="***"' + "'"
|
|
86
|
+
)
|
|
87
|
+
spans.append((start - 1, cursor, replacement))
|
|
77
88
|
return tuple(spans)
|
|
78
89
|
|
|
79
90
|
|
|
@@ -100,10 +111,13 @@ def credential_reference_view(value: str) -> str:
|
|
|
100
111
|
parts = []
|
|
101
112
|
previous = 0
|
|
102
113
|
references = credential_reference_spans(value)
|
|
103
|
-
carriers =
|
|
114
|
+
carriers = _absent_carriers(value)
|
|
104
115
|
if len(references) + len(carriers) > MAX_CREDENTIAL_FIELDS:
|
|
105
116
|
raise RepresentationLimit("credential semantic span budget")
|
|
106
|
-
|
|
117
|
+
# A proved absent field stays an empty value. Preserve enclosing data
|
|
118
|
+
# grammar and its key; do not introduce a new masked carrier that would
|
|
119
|
+
# require reconstructing the proof after lossy display transformations.
|
|
120
|
+
spans = sorted((*((a, b, '""') for a, b in references), *carriers))
|
|
107
121
|
for start, end, replacement in spans:
|
|
108
122
|
if start < previous:
|
|
109
123
|
continue
|
|
@@ -0,0 +1,64 @@
|
|
|
1
|
+
"""Bounded internationalized address candidates, not registration validation."""
|
|
2
|
+
|
|
3
|
+
from __future__ import annotations
|
|
4
|
+
|
|
5
|
+
from collections.abc import Callable
|
|
6
|
+
|
|
7
|
+
import regex
|
|
8
|
+
|
|
9
|
+
from ocr_toolkit.context.dlp.markup import interpret_markdown
|
|
10
|
+
|
|
11
|
+
_LOCAL = r"""(?:[a-z0-9.!#$%&'*+/=?^_`{|}~\p{L}\p{N}\p{M}-]{1,64}|"(?:[^"\\\r\n]|\\.){1,64}")"""
|
|
12
|
+
_LABEL = r"[\p{L}\p{N}](?:[\p{L}\p{N}\p{M}-]{0,61}[\p{L}\p{N}\p{M}])?"
|
|
13
|
+
# SMTP literal forms are dotted numeric addresses or tagged address literals,
|
|
14
|
+
# not arbitrary bracketed text (e.g. a regex character class after @ref).
|
|
15
|
+
# Keep malformed numeric/tagged contacts conservative; no DNS/IANA lookup.
|
|
16
|
+
_DOMAIN_LITERAL = (
|
|
17
|
+
r"(?:[0-9]{1,3}(?:\.[0-9]{1,3}){3}|[A-Za-z0-9][A-Za-z0-9-]{0,31}:[^\[\]\\\s]{1,80})"
|
|
18
|
+
)
|
|
19
|
+
EMAIL_RE = regex.compile(
|
|
20
|
+
rf"(?i)(?<![\w.+-])(?P<local>{_LOCAL})@"
|
|
21
|
+
rf"(?:(?:{_LABEL}\.)+(?:[\p{{L}}\p{{M}}]{{1,63}}|xn--[a-z0-9-]{{1,59}})|"
|
|
22
|
+
rf"\[{_DOMAIN_LITERAL}\])(?![\w-])",
|
|
23
|
+
)
|
|
24
|
+
|
|
25
|
+
|
|
26
|
+
def contains_email(
|
|
27
|
+
value: str,
|
|
28
|
+
*,
|
|
29
|
+
markdown: bool = False,
|
|
30
|
+
original: str | None = None,
|
|
31
|
+
transform: Callable[[str], str] | None = None,
|
|
32
|
+
) -> bool:
|
|
33
|
+
if "@" not in value:
|
|
34
|
+
return False
|
|
35
|
+
openings: dict[int, int] | None = None
|
|
36
|
+
for match in EMAIL_RE.finditer(value, timeout=0.05):
|
|
37
|
+
local = match["local"]
|
|
38
|
+
if markdown and local and not local.strip("`"):
|
|
39
|
+
if openings is None:
|
|
40
|
+
source = value if original is None else original
|
|
41
|
+
proven = interpret_markdown(source).code_openings
|
|
42
|
+
openings = {start + width: start for start, width in proven}
|
|
43
|
+
if original is not None and transform is not None:
|
|
44
|
+
# Transform disjoint source slices once, never search for
|
|
45
|
+
# a repeated lexeme. Backticks form normalization boundaries.
|
|
46
|
+
# Equality verifies compositionality before adopting offsets.
|
|
47
|
+
parts: list[str] = []
|
|
48
|
+
mapped: dict[int, int] = {}
|
|
49
|
+
cursor = position = 0
|
|
50
|
+
for start, width in proven:
|
|
51
|
+
part = transform(source[cursor:start])
|
|
52
|
+
parts.append(part)
|
|
53
|
+
position += len(part)
|
|
54
|
+
mapped[position + width] = position
|
|
55
|
+
cursor = start
|
|
56
|
+
parts.append(transform(source[cursor:]))
|
|
57
|
+
openings = mapped if "".join(parts) == value else {}
|
|
58
|
+
# Only the actual opening delimiter can masquerade as a local
|
|
59
|
+
# part. Repetition and the rest of the command do not change it.
|
|
60
|
+
opening = openings.get(match.start() + len(local))
|
|
61
|
+
if opening is not None and opening <= match.start():
|
|
62
|
+
continue
|
|
63
|
+
return True
|
|
64
|
+
return False
|