Refactor spaCy model installation: remove explicit downloads from CI and documentation, add model dependencies in pyproject.toml

.github/workflows/ci.yml

@@ -71,23 +71,18 @@ jobs:
           - name: 'Analyzer'
             path: 'presidio-analyzer'
             extras: '--all-extras'
-            spacy-models: 'en_core_web_lg en_core_web_sm'
           - name: 'Anonymizer'
             path: 'presidio-anonymizer'
             extras: ''
-            spacy-models: ''
           - name: 'Image Redactor'
             path: 'presidio-image-redactor'
             extras: ''
-            spacy-models: 'en_core_web_lg'
           - name: 'CLI'
             path: 'presidio-cli'
             extras: ''
-            spacy-models: ''
           - name: 'Structured'
             path: 'presidio-structured'
             extras: ''
-            spacy-models: ''
     env:
       POETRY_CACHE_DIR: /mnt/poetry_cache
       COVERAGE_THRESHOLD: 90
@@ -133,14 +128,6 @@ jobs:
         run: |
           poetry run pip install -e ../presidio-analyzer/.
 
-      - name: Download spaCy models
-        if: matrix.component.spacy-models != ''
-        working-directory: ${{ matrix.component.path }}
-        run: |
-          for model in ${{ matrix.component.spacy-models }}; do
-            poetry run python -m spacy download $model
-          done
-
       - name: Run tests with coverage
         working-directory: ${{ matrix.component.path }}
         run: |

docs/analyzer/nlp_engines/transformers.md

@@ -53,8 +53,12 @@ AutoModelForTokenClassification.from_pretrained(transformers_model)
 
 Then, also download a spaCy pipeline/model:
 
+!!! note "Note"
+    The `en_core_web_sm` and `en_core_web_lg` models are installed automatically with `presidio_analyzer`.
+    For other spaCy models, download them manually:
+
 ```sh
-python -m spacy download en_core_web_sm
+python -m spacy download <model_name>
 ```
 
 ### Configuring the NER pipeline

docs/development.md

@@ -57,9 +57,6 @@ Follow these steps when starting to work on a Presidio service with poetry:
     `poetry run`. For example:
     1. `poetry run ruff check`
     2. `poetry run pip freeze`
-    3. `poetry run python -m spacy download en_core_web_lg`
-
-    Command 3 downloads the default spacy model needed for Presidio Analyzer.`
 
 #### Alternatively, activate the virtual environment and use the commands using [this method](https://python-poetry.org/docs/basic-usage/#activating-the-virtual-environment).
 

docs/getting_started/getting_started_text.md

@@ -15,7 +15,6 @@ Using Presidio's modules as Python packages to get started:
         ```sh
         pip install presidio-analyzer
         pip install presidio-anonymizer
-        python -m spacy download en_core_web_lg
         ```
     
     2. Analyze + Anonymize
@@ -52,7 +51,6 @@ Using Presidio's modules as Python packages to get started:
         ```sh
         pip install "presidio-analyzer[transformers]"
         pip install presidio-anonymizer
-        python -m spacy download en_core_web_sm
         ```
     
     2. Analyze + Anonymize

docs/image-redactor/index.md

@@ -31,11 +31,10 @@ Pre-requisites:
         Consider installing the Presidio python packages on a virtual environment like venv or conda.
     
     To get started with Presidio-image-redactor,
-    download the package and the `en_core_web_lg` spaCy model:
+    download the package:
     
     ```sh
     pip install presidio-image-redactor
-    python -m spacy download en_core_web_lg
     ```
 
 === "Using Docker"

docs/installation.md

@@ -32,7 +32,6 @@ with at least one NLP engine (`spaCy`, `transformers` or `stanza`):
     ```
     pip install presidio_analyzer
     pip install presidio_anonymizer
-    python -m spacy download en_core_web_lg
     ```
 
 === "Transformers"
@@ -40,13 +39,12 @@ with at least one NLP engine (`spaCy`, `transformers` or `stanza`):
     ```
     pip install "presidio_analyzer[transformers]"
     pip install presidio_anonymizer
-    python -m spacy download en_core_web_sm
     ```
 
     !!! note "Note"
         
-        When using a transformers NLP engine, Presidio would still use spaCy for other capabilities,
-        therefore a small spaCy model (such as en_core_web_sm) is required. 
+        When using a transformers NLP engine, Presidio would still use spaCy for other capabilities.
+        The spaCy models (`en_core_web_lg` and `en_core_web_sm`) are installed automatically.
         Transformers models would be loaded lazily. To pre-load them, see: [Downloading a pre-trained model](./analyzer/nlp_engines/transformers.md#downloading-a-pre-trained-model)
 
 === "Stanza"
@@ -69,10 +67,6 @@ For PII redaction in images
 
     ```sh
     pip install presidio_image_redactor
-    
-    # Presidio image redactor uses the presidio-analyzer
-    # which requires a spaCy language model:
-    python -m spacy download en_core_web_lg
     ```
 
 2. Install an OCR engine. The default version uses the [Tesseract OCR Engine](https://github.com/tesseract-ocr/tesseract).

docs/tutorial/00_getting_started.md

@@ -8,7 +8,6 @@ Install from PyPI:
 ```sh
 pip install presidio_analyzer
 pip install presidio_anonymizer
-python -m spacy download en_core_web_lg
 ```
 
 ## Simple flow

presidio-analyzer/pyproject.toml

@@ -24,6 +24,8 @@ include = ["conf/*",]
 requires-python = ">=3.10,<3.14"
 dependencies = [
     "spacy (>=3.4.4, !=3.7.0)",
+    "en-core-web-lg @ https://github.com/explosion/spacy-models/releases/download/en_core_web_lg-3.8.0/en_core_web_lg-3.8.0-py3-none-any.whl",
+    "en-core-web-sm @ https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0-py3-none-any.whl",
     "regex",
     "tldextract",
     "pyyaml",

presidio-image-redactor/Dockerfile

@@ -29,9 +29,6 @@ RUN pip install poetry \
     && poetry install --no-root --only=main -E server\
     && rm -rf $(poetry config cache-dir)
 
-# Install spaCy model during build (as root) so it's available to non-root user at runtime
-RUN python -m spacy download en_core_web_lg
-
 COPY . /app/
 
 # Create a non-root user and set ownership