pyannotators-patterns 0.6.21__tar.gz → 0.7.1__tar.gz

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/.gitignore

@@ -129,9 +129,3 @@ dmypy.json
 .idea/
 .groovylintrc.json
 .emailNotif
-uv.lock
-
-# SBOMs
-**/sbom*.json
-**/trivy*.html
-**/audit*.json

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/.pre-commit-config.yaml

@@ -1,14 +1,17 @@
 repos:
--   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.3.0
+-   repo: https://github.com/ambv/black
+    rev: 19.3b0
     hooks:
-    - id: ruff
-      args: [--fix]
-    - id: ruff-format
+    - id: black
+      language_version: python3.8
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.5.0
+    rev: v2.1.0
     hooks:
     - id: end-of-file-fixer
       exclude: '.bumpversion.cfg'
     - id: trailing-whitespace
       exclude: '.bumpversion.cfg'
+-   repo: https://gitlab.com/pycqa/flake8
+    rev: 3.7.9
+    hooks:
+    - id: flake8

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.12-slim-bookworm
+FROM python:3.8-slim-bookworm
 # Install prerequisites
 RUN apt-get update -y && \
     apt-get install -y \
@@ -11,9 +11,9 @@ RUN apt-get update -y && \
     apt-get update -y && \
     apt-get clean all -y
 
-# Install uv
-RUN pip install uv
+# Enable Installing packages as root
+ENV FLIT_ROOT_INSTALL=1
 
-# Add pyproject.toml + README.md for uv install
+# Add pyproject.toml + README.md for flit install
 ADD pyproject.toml pyproject.toml
 ADD README.md README.md

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/Jenkinsfile

@@ -20,7 +20,7 @@ pipeline {
     JENKINS_UIDGID = '1004:1004'
 
     MAJOR_VERSION = '0'
-    MINOR_VERSION = '6'
+    MINOR_VERSION = '7'
   }
 
   stages {
@@ -58,7 +58,7 @@ pipeline {
         stage('Add credentials') {
           steps {
             script {
-              // Add password file for uv publishing
+              // Add password file for flit publishing
               sh "cp ${PATH_HOME}/.passwd-pypi .env"
             }
           }
@@ -73,7 +73,6 @@ pipeline {
               withCredentials([gitUsernamePassword(credentialsId: 'bitbucket-user', gitToolName: 'git-tool')]) {
                 sh 'git pull'
                 sh "echo '\"\"\"Annotator based on Presidio pattern recognizer\"\"\"' > src/pyannotators_patterns/__init__.py"
-                sh "echo '' >> src/pyannotators_patterns/__init__.py"
                 sh "echo '__version__ = \"${MAJOR_VERSION}.${MINOR_VERSION}.${BUILD_ID}\"' >> src/pyannotators_patterns/__init__.py"
                 sh 'git commit src/pyannotators_patterns/__init__.py -m "[Jenkins CI] Commit on version files" || echo "No changes to commit"'
                 sh 'git push'
@@ -104,11 +103,13 @@ pipeline {
       }
 
       stages {
-        stage('Install dependencies') {
+        stage('Install flit & flake8') {
           steps {
-            sh 'rm -f uv.lock'
-            sh 'pip install uv'
-            sh 'uv sync --no-cache --extra test'
+            // remove any previous tox env
+            sh 'rm -rf .tox'
+            sh 'python -m pip install pip==22.0.3'
+            sh 'pip install --no-cache-dir flit==3.2.0 flake8==3.9.2 flakehell tox'
+            sh 'flit install'
           }
         }
 
@@ -116,20 +117,20 @@ pipeline {
           steps {
             // remove any previous results.xml file
             sh "rm -f ${TEST_REPORT_DIR}/results.xml"
-            sh 'uv run ruff check .'
-            sh 'uv run ruff format --check .'
-            sh "uv run pytest --junit-xml=${TEST_REPORT_DIR}/results.xml"
+            sh 'tox'
           }
         }
 
         stage('Publish on PyPI') {
           environment {
-            UV_PUBLISH_USERNAME = getUserName '.env'
-            UV_PUBLISH_PASSWORD = getUserPass '.env'
+            FLIT_USERNAME = getUserName '.env'
+            FLIT_PASSWORD = getUserPass '.env'
           }
           steps {
             // remove any previous folder dist
             sh 'rm -rf dist'
+            // create (as root) folder dist
+            sh 'mkdir dist'
             // pull recent updates of file __init__.py
             withCredentials([gitUsernamePassword(credentialsId: 'bitbucket-user', gitToolName: 'git-tool')]) {
               sh 'git config --global pull.rebase false'
@@ -142,85 +143,18 @@ pipeline {
             sh "chown ${JENKINS_UIDGID} src/pyannotators_patterns/__init__.py"
             // get git status
             sh 'git status'
-            // build and publish on PyPI
+            // publish on PyPI
             sh '''
                export COMMIT_VERSION=$( cat src/pyannotators_patterns/__init__.py|grep version|cut -d '"' -f2|tr -s '[:blank:]' )
                export BUILD_VERSION="${MAJOR_VERSION}"."${MINOR_VERSION}"."${BUILD_ID}"
-               if [ "${COMMIT_VERSION}" = "${BUILD_VERSION}" ] ; then uv build && uv publish ; fi
+               if [ "${COMMIT_VERSION}" = "${BUILD_VERSION}" ] ; then flit publish ; fi
                '''
             // remove current folder dist
             sh 'rm -rf dist'
             // remove current folder .hypothesis
             sh 'rm -rf .hypothesis'
-          }
-        }
-      }
-    }
-
-    stage('Prepare SBOM') {
-      when {
-        beforeAgent true
-        environment name: 'SKIP_JOB', value: '0'
-      }
-      options {
-        timeout(time: 120, unit: 'SECONDS')
-      }
-
-      agent {
-        docker {
-          image 'alpine:3.19'
-          label 'built-in'
-          customWorkspace "${PATH_HOME}/${JOB_NAME}"
-          args "-u 0"
-        }
-      }
-
-      stages {
-        stage('Generate SBOM') {
-          steps {
-            sh '''
-               apk add --no-cache curl
-
-               curl -sSfL https://raw.githubusercontent.com/anchore/syft/main/install.sh | sh -s -- -b . v${SYFT_VERSION}
-
-               ./syft . \
-                --exclude '**/syft' \
-                --exclude '**/.pytest_cache' \
-                --exclude '**/.ruff_cache' \
-                --exclude '**/sbom.cdx.json' \
-                --exclude '**/sbom.spdx.json' \
-                --source-name "${JOB_NAME}" \
-                --source-version "${BUILD_NUMBER}" \
-                -o cyclonedx-json=sbom.cdx.json \
-                -o spdx-json=sbom.spdx.json
-
-               rm -f ./syft
-               '''
-          }
-        }
-
-        stage('Analyse SBOM') {
-          steps {
-            sh '''
-               curl -sfL https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/install.sh | sh -s -- -b . v${TRIVY_VERSION}
-               curl -L https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/html.tpl -o trivy-html-template.tpl
-
-               ./trivy sbom --download-db-only
-
-               ./trivy sbom sbom.cdx.json \
-                --skip-db-update \
-                --format template \
-                --template "@trivy-html-template.tpl" \
-                -o trivy-report.html
-
-               rm -f ./trivy
-               '''
-          }
-        }
-
-        stage('Archive SBOM & analysis') {
-          steps {
-            archiveArtifacts artifacts: 'sbom*.json, trivy*.html', fingerprint: true
+            // remove current folder .tox
+            sh 'rm -rf .tox'
           }
         }
       }
@@ -305,7 +239,7 @@ pipeline {
   }
 }
 
-// return UV_PUBLISH_USERNAME from given file (reads FLIT_USERNAME key)
+// return FLIT_USERNAME from given file
 def getUserName(path) {
   def USERNAME = sh(
                  script: "grep FLIT_USERNAME ${path}|cut -d '=' -f2",
@@ -314,7 +248,7 @@ def getUserName(path) {
   return USERNAME
 }
 
-// return UV_PUBLISH_PASSWORD from given file (reads FLIT_PASSWORD key)
+// return FLIT_PASSWORD from given file
 def getUserPass(path) {
   def USERPASS = sh(
                  script: "grep FLIT_PASSWORD ${path}|cut -d '=' -f2",
@@ -475,6 +409,6 @@ def analyseBuildCause() {
     println 'Skipping build because last commit has been done by CI'
     env.SKIP_JOB = '1'
     switchEmailNotif(false, 0)
-    currentBuild.result = 'NOT_BUILT'
+    //currentBuild.result = 'NOT_BUILT'
   }
 }

pyannotators_patterns-0.7.1/PKG-INFO

@@ -0,0 +1,97 @@
+Metadata-Version: 2.4
+Name: pyannotators-patterns
+Version: 0.7.1
+Summary: Annotator based on Presidio pattern recognizer
+Home-page: https://github.com/oterrier/pyannotators_patterns/
+Keywords: 
+Author: Olivier Terrier
+Author-email: olivier.terrier@kairntech.com
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Classifier: Intended Audience :: Information Technology
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: System Administrators
+Classifier: Operating System :: OS Independent
+Classifier: Topic :: Software Development :: Libraries :: Application Frameworks
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Topic :: Software Development
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Development Status :: 4 - Beta
+Classifier: Programming Language :: Python :: 3.8
+License-File: LICENSE
+Requires-Dist: pymultirole-plugins>=0.7.0,<0.8.0
+Requires-Dist: tldextract==5.1.2
+Requires-Dist: spacy==3.4.4
+Requires-Dist: spacy[lookups]==3.4.4
+Requires-Dist: log-with-context
+Requires-Dist: collections_extended
+Requires-Dist: unidecode
+Requires-Dist: presidio-analyzer==2.2.354
+Requires-Dist: flit ; extra == "dev"
+Requires-Dist: pre-commit ; extra == "dev"
+Requires-Dist: bump2version ; extra == "dev"
+Requires-Dist: sphinx ; extra == "docs"
+Requires-Dist: sphinx-rtd-theme ; extra == "docs"
+Requires-Dist: m2r2 ; extra == "docs"
+Requires-Dist: pytest ; extra == "test"
+Requires-Dist: pytest-cov ; extra == "test"
+Requires-Dist: pytest-flake8 ; extra == "test"
+Requires-Dist: pytest-black ; extra == "test"
+Requires-Dist: pytest_check ; extra == "test"
+Requires-Dist: flake8==3.9.2 ; extra == "test"
+Requires-Dist: tox ; extra == "test"
+Requires-Dist: dirty-equals ; extra == "test"
+Provides-Extra: dev
+Provides-Extra: docs
+Provides-Extra: test
+
+# pyannotators_patterns
+
+[![license](https://img.shields.io/github/license/oterrier/pyannotators_patterns)](https://github.com/oterrier/pyannotators_patterns/blob/master/LICENSE)
+[![tests](https://github.com/oterrier/pyannotators_patterns/workflows/tests/badge.svg)](https://github.com/oterrier/pyannotators_patterns/actions?query=workflow%3Atests)
+[![codecov](https://img.shields.io/codecov/c/github/oterrier/pyannotators_patterns)](https://codecov.io/gh/oterrier/pyannotators_patterns)
+[![docs](https://img.shields.io/readthedocs/pyannotators_patterns)](https://pyannotators_patterns.readthedocs.io)
+[![version](https://img.shields.io/pypi/v/pyannotators_patterns)](https://pypi.org/project/pyannotators_patterns/)
+[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/pyannotators_patterns)](https://pypi.org/project/pyannotators_patterns/)
+
+Annotator based on Facebook's Patterns
+
+## Installation
+
+You can simply `pip install pyannotators_patterns`.
+
+## Developing
+
+### Pre-requesites
+
+You will need to install `flit` (for building the package) and `tox` (for orchestrating testing and documentation building):
+
+```
+python3 -m pip install flit tox
+```
+
+Clone the repository:
+
+```
+git clone https://github.com/oterrier/pyannotators_patterns
+```
+
+### Running the test suite
+
+You can run the full test suite against all supported versions of Python (3.8) with:
+
+```
+tox
+```
+
+### Building the documentation
+
+You can build the HTML documentation with:
+
+```
+tox -e docs
+```
+
+The built documentation is available at `docs/_build/index.html.
+

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/README.md

@@ -7,82 +7,42 @@
 [![version](https://img.shields.io/pypi/v/pyannotators_patterns)](https://pypi.org/project/pyannotators_patterns/)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/pyannotators_patterns)](https://pypi.org/project/pyannotators_patterns/)
 
-Annotator based on Presidio regex pattern recognizers.
+Annotator based on Facebook's Patterns
 
 ## Installation
 
-```
-pip install pyannotators-patterns
-```
+You can simply `pip install pyannotators_patterns`.
 
 ## Developing
 
-### Prerequisites
+### Pre-requesites
 
-You will need [uv](https://github.com/astral-sh/uv) (package manager) and Python 3.12+.
-
-Clone the repository:
+You will need to install `flit` (for building the package) and `tox` (for orchestrating testing and documentation building):
 
 ```
-git clone https://github.com/oterrier/pyannotators_patterns
-cd pyannotators_patterns
+python3 -m pip install flit tox
 ```
 
-Install dependencies (including test extras):
+Clone the repository:
 
 ```
-uv sync --extra test
+git clone https://github.com/oterrier/pyannotators_patterns
 ```
 
 ### Running the test suite
 
-```
-uv run pytest
-```
-
-### Linting and formatting
+You can run the full test suite against all supported versions of Python (3.8) with:
 
 ```
-uv run ruff check .
-uv run ruff format --check .
-```
-
-To auto-fix formatting:
-
-```
-uv run ruff format .
+tox
 ```
 
 ### Building the documentation
 
-```
-uv run --extra docs sphinx-build docs docs/_build
-```
-
-The built documentation is available at `docs/_build/index.html`.
-
-## SBOM & vulnerability check
-
-Install the SBOM dependencies:
-
-```
-uv sync --extra sbom
-```
-
-Generate a CycloneDX SBOM from the current environment:
+You can build the HTML documentation with:
 
 ```
-uv run cyclonedx-py environment -o sbom.cdx.json --output-format json
+tox -e docs
 ```
 
-Audit dependencies for known vulnerabilities:
-
-```
-uv run pip-audit --format json --output audit-report.json
-```
-
-To fail on any known vulnerability (useful in CI):
-
-```
-uv run pip-audit --strict
-```
+The built documentation is available at `docs/_build/index.html.

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/docs/conf.py

@@ -18,6 +18,7 @@ import os
 
 import pyannotators_patterns
 
+
 # -- Project information -----------------------------------------------------
 
 project = "pyannotators_patterns"

pyannotators_patterns-0.7.1/pyproject.toml

@@ -0,0 +1,90 @@
+[build-system]
+requires = ["flit_core >=3.2,<4"]
+#requires = ["flit_core >=2,<3"]
+build-backend = "flit_core.buildapi"
+
+[tool.flit.metadata]
+module = "pyannotators_patterns"
+author = "Olivier Terrier"
+author-email = "olivier.terrier@kairntech.com"
+home-page = "https://github.com/oterrier/pyannotators_patterns/"
+requires-python=">=3.8"
+description-file="README.md"
+keywords = ""
+classifiers = [
+    "Intended Audience :: Information Technology",
+    "Intended Audience :: Developers",
+    "Intended Audience :: System Administrators",
+    "Operating System :: OS Independent",
+    "Topic :: Software Development :: Libraries :: Application Frameworks",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+    "Topic :: Software Development :: Libraries",
+    "Topic :: Software Development",
+    "License :: OSI Approved :: MIT License",
+    "Development Status :: 4 - Beta",
+    "Programming Language :: Python :: 3.8",
+]
+requires = [
+    "pymultirole-plugins>=0.7.0,<0.8.0",
+    "tldextract==5.1.2",
+    "spacy==3.4.4",
+    "spacy[lookups]==3.4.4",
+    "log-with-context",
+    "collections_extended",
+    "unidecode",
+    "presidio-analyzer==2.2.354"
+]
+dist-name = "pyannotators-patterns"
+
+[tool.flit.entrypoints."pyannotators.plugins"]
+patterns = "pyannotators_patterns.patterns:PatternsAnnotator"
+
+[tool.flit.metadata.requires-extra]
+test = [
+    "pytest",
+    "pytest-cov",
+    "pytest-flake8",
+    "pytest-black",
+    "pytest_check",
+    "flake8==3.9.2",
+    "tox",
+    "dirty-equals"
+]
+docs = [
+    "sphinx",
+    "sphinx-rtd-theme",
+    "m2r2",  # markdown support
+    # "sphinxcontrib.apidoc",  # run sphinx-apidoc when building docs
+    # "jupyter_sphinx",   # for execution of code snippets in the documentation
+]
+dev = [
+    "flit",
+    "pre-commit",
+    "bump2version",
+]
+
+[tool.flakehell]
+exclude = ["README.md"]
+format = "colored"
+#format = "junit-xml"
+max_line_length = 120
+show_source = true
+#whitelist = "../../allowlist.txt"
+
+[tool.flakehell.plugins]
+flake8-bandit = ["+*", "-S322"]
+flake8-bugbear = ["+*"]
+flake8-builtins = ["+*"]
+flake8-comprehensions = ["+*"]
+#flake8-darglint = ["+*"]
+flake8-docstrings = ["+*"]
+flake8-eradicate = ["+*"]
+flake8-isort = ["+*"]
+flake8-mutable = ["+*"]
+flake8-pytest-style = ["+*"]
+flake8-spellcheck = ["+*"]
+mccabe = ["+*"]
+pep8-naming = ["+*"]
+pycodestyle = ["+*"]
+pyflakes = ["+*"]
+pylint = ["+*"]

pyannotators_patterns-0.7.1/setup.py

@@ -0,0 +1,54 @@
+#!/usr/bin/env python
+# setup.py generated by flit for tools that don't yet use PEP 517
+
+from distutils.core import setup
+
+packages = \
+['pyannotators_patterns']
+
+package_data = \
+{'': ['*']}
+
+package_dir = \
+{'': 'src'}
+
+install_requires = \
+['pymultirole-plugins>=0.7.0,<0.8.0',
+ 'tldextract==5.1.2',
+ 'spacy==3.4.4',
+ 'spacy[lookups]==3.4.4',
+ 'log-with-context',
+ 'collections_extended',
+ 'unidecode',
+ 'presidio-analyzer==2.2.354']
+
+extras_require = \
+{'dev': ['flit', 'pre-commit', 'bump2version'],
+ 'docs': ['sphinx', 'sphinx-rtd-theme', 'm2r2'],
+ 'test': ['pytest',
+          'pytest-cov',
+          'pytest-flake8',
+          'pytest-black',
+          'pytest_check',
+          'flake8==3.9.2',
+          'tox',
+          'dirty-equals']}
+
+entry_points = \
+{'pyannotators.plugins': ['patterns = '
+                          'pyannotators_patterns.patterns:PatternsAnnotator']}
+
+setup(name='pyannotators-patterns',
+      version='0.7.1',
+      description='Annotator based on Presidio pattern recognizer',
+      author='Olivier Terrier',
+      author_email='olivier.terrier@kairntech.com',
+      url='https://github.com/oterrier/pyannotators_patterns/',
+      packages=packages,
+      package_data=package_data,
+      package_dir=package_dir,
+      install_requires=install_requires,
+      extras_require=extras_require,
+      entry_points=entry_points,
+      python_requires='>=3.8',
+     )

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/src/pyannotators_patterns/__init__.py

@@ -1,3 +1,2 @@
 """Annotator based on Presidio pattern recognizer"""
-
-__version__ = "0.6.21"
+__version__ = "0.7.1"

{pyannotators_patterns-0.6.21 → pyannotators_patterns-0.7.1}/src/pyannotators_patterns/named_pattern_recognizer.py

@@ -1,41 +1,33 @@
 import re
+from typing import List, Optional
 
-from presidio_analyzer import EntityRecognizer, Pattern, PatternRecognizer, RecognizerResult
+from presidio_analyzer import Pattern, PatternRecognizer, RecognizerResult, EntityRecognizer
 from presidio_analyzer.nlp_engine import NlpArtifacts
 
 
 class NamedPatternRecognizer(PatternRecognizer):
     def __init__(
-        self,
-        supported_entity: str,
-        name: str = None,
-        supported_language: str = "en",
-        patterns: list[Pattern] = None,
-        deny_list: list[str] = None,
-        context: list[str] = None,
-        deny_list_score: float = 1.0,
-        global_regex_flags: int | None = re.DOTALL | re.MULTILINE,
-        version: str = "0.0.1",
+            self,
+            supported_entity: str,
+            name: str = None,
+            supported_language: str = "en",
+            patterns: List[Pattern] = None,
+            deny_list: List[str] = None,
+            context: List[str] = None,
+            deny_list_score: float = 1.0,
+            global_regex_flags: Optional[int] = re.DOTALL | re.MULTILINE,
+            version: str = "0.0.1",
     ):
-        super().__init__(
-            supported_entity,
-            name,
-            supported_language,
-            patterns,
-            deny_list,
-            context,
-            deny_list_score,
-            global_regex_flags,
-            version,
-        )
+        super().__init__(supported_entity, name, supported_language, patterns, deny_list, context, deny_list_score,
+                         global_regex_flags, version)
 
     def analyze(
-        self,
-        text: str,
-        entities: list[str],  # noqa: ARG002
-        nlp_artifacts: NlpArtifacts | None = None,  # noqa: ARG002
-        regex_flags: int | None = None,
-    ) -> list[RecognizerResult]:
+            self,
+            text: str,
+            entities: List[str],
+            nlp_artifacts: Optional[NlpArtifacts] = None,
+            regex_flags: Optional[int] = None,
+    ) -> List[RecognizerResult]:
         """
         Analyzes text to detect PII using regular expressions or deny-lists.
 
@@ -53,7 +45,9 @@ class NamedPatternRecognizer(PatternRecognizer):
 
         return results
 
-    def __analyze_patterns(self, text: str, flags: int = None) -> list[RecognizerResult]:
+    def __analyze_patterns(
+            self, text: str, flags: int = None
+    ) -> List[RecognizerResult]:
         """
         Evaluate all patterns in the provided text.
 
@@ -99,7 +93,7 @@ class NamedPatternRecognizer(PatternRecognizer):
                     end=end,
                     score=score,
                     analysis_explanation=description,
-                    recognition_metadata=recognition_metadata,
+                    recognition_metadata=recognition_metadata
                 )
 
                 if validation_result is not None: