V2.2.0 beta1

.gitignore

@@ -16,3 +16,4 @@ build/
 node_modules
 datafog_debug.log
 sotu_2023.txt
+/examples/*

.pre-commit-config.yaml

@@ -1,8 +1,9 @@
 repos:
   - repo: https://github.com/PyCQA/isort
-    rev: 5.13.2
+    rev: 5.12.0
     hooks:
       - id: isort
+        args: [--profile=black]
 
   - repo: https://github.com/psf/black
     rev: 24.2.0

requirements.txt

@@ -1,6 +1,13 @@
-'https://huggingface.co/beki/en_spacy_pii_fast/resolve/main/en_spacy_pii_fast-any-py3-none-any.whl'
-pandas==2.2.1
+# spacy==3.4.4
+# thinc==8.1.0
 presidio_analyzer==2.2.353
+pandas==2.2.1
 pytest==8.0.2
 Requests==2.31.0
-spacy
+aiohttp==3.8.2
+yarl==1.8.1
+frozenlist==1.3.1
+en_spacy_pii_fast
+
+
+

setup.py

@@ -1,15 +1,14 @@
 #  package using pip, navigate to the directory that contains the setup.py file and type pip install .
 
 from setuptools import setup
-from src.datafog.__about__ import __version__
 
 # Read README for the long description
 with open("README.md", "r") as f:
     long_description = f.read()
 
 setup(
     name="datafog",
-    version=__version__,
+    version="2.2.0",
     author="Sid Mohan",
     author_email="sid@datafog.ai",
     description="Scan, redact, and manage PII in your documents before they get uploaded to a Retrieval Augmented Generation (RAG) system.",
@@ -21,9 +20,9 @@
         "pytest==8.0.2",
         "Requests==2.31.0",
         "spacy==3.4.4",
-        'en_spacy_pii_fast @ https://huggingface.co/beki/en_spacy_pii_fast/resolve/main/en_spacy_pii_fast-any-py3-none-any.whl',
-
+        "en_spacy_pii_fast",
     ],
+    python_requires=">=3.10",
     classifiers=[
         "Programming Language :: Python :: 3.10",
         "License :: OSI Approved :: MIT License",

src/datafog/__about__.py

@@ -1,2 +1,2 @@
 # SSOT for the package version
-__version__ = "2.2.0-beta1"
+__version__ = "2.2.0"

src/datafog/pii_tools/PresidioEngine/analyzer.py

@@ -7,7 +7,6 @@
 
 
 class CustomSpacyRecognizer(LocalRecognizer):
-
     ENTITIES = [
         "LOCATION",
         "PERSON",

tox.ini

@@ -1,12 +1,13 @@
 [tox]
-envlist = py3
+envlist = py310, py311
 isolated_build = True
 
 [testenv]
 deps =
     pytest
     pytest-codeblocks
     pytest-cov
+    -rrequirements.txt
 extras = all
 commands =
     pytest {posargs} --codeblocks