chore(deploy): slim production image and fix SVG MIME type

.dockerignore

@@ -1,6 +1,10 @@
 # Local reports, captures and scratch files
 .user/
 
+# Test sources are mounted into the Compose test container instead of shipped
+# in the production image.
+vectorizing/tests/
+
 ### Python template
 # Byte-compiled / optimized / DLL files
 __pycache__/
@@ -164,7 +168,6 @@ cython_debug/
 .idea/
 
 # Other files
-heroku.yml
 docker-compose.yml
 .pre-commit-config.yaml
 .vscode/

Dockerfile

@@ -1,17 +1,10 @@
 # Use an official Python runtime as a parent image
 FROM python:3.11-slim
 
-# Install necessary packages
-RUN apt-get update -y && apt-get install -y \
-    wget \
-    build-essential \
-    python3-dev \
-    libagg-dev \
-    libpotrace-dev \
-    pkg-config \
-    libgl1 \
-    libcairo2 \
-    --no-install-recommends && \
+# Install system dependencies shared with the development container.
+COPY scripts/install_system_dependencies.sh /tmp/install_system_dependencies.sh
+RUN bash /tmp/install_system_dependencies.sh && \
+    rm /tmp/install_system_dependencies.sh && \
     rm -rf /var/lib/apt/lists/*
 
 # Create a non-root user and group with specific IDs for consistency
@@ -21,8 +14,10 @@ RUN addgroup --gid 1001 appuser && \
 # Set the working directory
 WORKDIR /app
 
-# Copy only requirements to leverage Docker cache
-COPY requirements/dev.txt /app/requirements.txt
+# Copy only requirements to leverage Docker cache. The test Compose service
+# overrides this with the development requirements.
+ARG REQUIREMENTS_FILE=prod.txt
+COPY requirements/${REQUIREMENTS_FILE} /app/requirements.txt
 
 # Install Python dependencies
 RUN pip install --no-cache-dir -r /app/requirements.txt

README.md

@@ -40,7 +40,7 @@ The first time the execution can take few minutes, as it is pulling the dev cont
 
 	This will compile dependencies and environments, ensuring a consistent development workflow and deployment.
 
-4. If you want to add or remove **system** dependencies, update the script: [`scripts/install_system_dependencies.sh`](scripts/install_system_dependencies.sh). This is used both in CI and at dev container creation, to keep them consistent.
+4. If you want to add or remove **system** dependencies, update the shared script: [`scripts/install_system_dependencies.sh`](scripts/install_system_dependencies.sh). It is used by the production Docker image and at dev container creation, to keep them consistent. The Compose test image uses development Python requirements and mounts `vectorizing/tests/` at runtime because test sources are excluded from the production build context.
 
 ## Linting and formatting
 
@@ -74,12 +74,16 @@ python scripts/benchmark_vectorization.py --output .user/benchmark.json --case a
 - Add `--baseline-root /path/to/baseline` to benchmark two trusted checkouts in alternating order, with exact-output checks. `--repeats` defaults to 3 and `--warmups` to 1 per process. Output mismatches exit `1` but still save the report; timings have no pass/fail threshold.
 - Reports include runtime/source fingerprints and raw samples. `--label` can record commit IDs. Checkouts share the installed dependencies; this is not a dependency-isolated or production/concurrency benchmark. Use only trusted source roots.
 
-For Docker, build the current scripts and mount only the output directory:
+For Docker, build the current scripts with development dependencies and mount the
+fixture images and output directory:
 
 ```bash
 mkdir -p .user
-docker build -t vectorizing:tools .
-docker run --rm --user "$(id -u):$(id -g)" -v "$PWD/.user:/results" vectorizing:tools \
+docker build --build-arg REQUIREMENTS_FILE=dev.txt -t vectorizing:tools .
+docker run --rm --user "$(id -u):$(id -g)" \
+  -v "$PWD/.user:/results" \
+  -v "$PWD/vectorizing/tests/images:/app/vectorizing/tests/images:ro" \
+  vectorizing:tools \
   python scripts/vectorization_outputs.py capture /results/candidate
 ```
 

docker-compose.yml

@@ -48,12 +48,16 @@ services:
 
   test:
     <<: *app
+    image: vectorizing:test
+    build:
+      context: .
+      args:
+        REQUIREMENTS_FILE: dev.txt
     profiles: [testing]
     environment:
       <<: *local-aws
     volumes:
-      - ./vectorizing/tests/results:/app/vectorizing/tests/results
-      - ./vectorizing/tests/diff_output:/app/vectorizing/tests/diff_output
+      - ./vectorizing/tests:/app/vectorizing/tests
     depends_on:
       moto-init:
         condition: service_completed_successfully

envs/dev.yaml

@@ -44,7 +44,6 @@ dependencies:
       - faiss-cpu==1.7.4
       - filelock==3.18.0
       - flask==2.3.3
-      - flask-cors==4.0.2
       - gunicorn==21.2.0
       - identify==2.6.12
       - idna==3.10
@@ -54,11 +53,9 @@ dependencies:
       - jinja2==3.1.6
       - jmespath==1.0.1
       - lazy-loader==0.4
-      - llvmlite==0.40.1
       - markupsafe==3.0.2
       - networkx==3.4.2
       - nodeenv==1.9.1
-      - numba==0.57.1
       - numpy==1.24.4
       - opencv-python-headless==4.8.0.76
       - packaging==25.0

envs/prod.yaml

@@ -34,18 +34,15 @@ dependencies:
       - cuid==0.4
       - faiss-cpu==1.7.4
       - flask==2.3.3
-      - flask-cors==4.0.2
       - gunicorn==21.2.0
       - idna==3.10
       - imageio==2.37.0
       - itsdangerous==2.2.0
       - jinja2==3.1.6
       - jmespath==1.0.1
       - lazy-loader==0.4
-      - llvmlite==0.40.1
       - markupsafe==3.0.2
       - networkx==3.4.2
-      - numba==0.57.1
       - numpy==1.24.4
       - opencv-python-headless==4.8.0.76
       - packaging==25.0

requirements/dev.txt

@@ -54,10 +54,6 @@ faiss-cpu==1.7.4
 filelock==3.18.0
     # via virtualenv
 flask==2.3.3
-    # via
-    #   -r /workspaces/vectorizing/requirements/prod.txt
-    #   flask-cors
-flask-cors==4.0.2
     # via -r /workspaces/vectorizing/requirements/prod.txt
 gunicorn==21.2.0
     # via -r /workspaces/vectorizing/requirements/prod.txt
@@ -90,10 +86,6 @@ lazy-loader==0.4
     # via
     #   -r /workspaces/vectorizing/requirements/prod.txt
     #   scikit-image
-llvmlite==0.40.1
-    # via
-    #   -r /workspaces/vectorizing/requirements/prod.txt
-    #   numba
 markupsafe==3.0.2
     # via
     #   -r /workspaces/vectorizing/requirements/prod.txt
@@ -105,13 +97,10 @@ networkx==3.4.2
     #   scikit-image
 nodeenv==1.9.1
     # via pre-commit
-numba==0.57.1
-    # via -r /workspaces/vectorizing/requirements/prod.txt
 numpy==1.24.4
     # via
     #   -r /workspaces/vectorizing/requirements/prod.txt
     #   imageio
-    #   numba
     #   opencv-python-headless
     #   pypotrace
     #   pywavelets

requirements/prod.in

@@ -3,9 +3,7 @@
 cuid~=0.4
 boto3~=1.35.2
 flask~=2.3.3
-flask-cors~=4.0.0
 sentry-sdk~=1.30.0
-numba~=0.57.1
 pillow~=10.0.0
 faiss-cpu~=1.7.4
 pypotrace~=0.3

requirements/prod.txt

@@ -25,10 +25,6 @@ cuid==0.4
 faiss-cpu==1.7.4
     # via -r /workspaces/vectorizing/scripts/../requirements/prod.in
 flask==2.3.3
-    # via
-    #   -r /workspaces/vectorizing/scripts/../requirements/prod.in
-    #   flask-cors
-flask-cors==4.0.2
     # via -r /workspaces/vectorizing/scripts/../requirements/prod.in
 gunicorn==21.2.0
     # via -r /workspaces/vectorizing/scripts/../requirements/prod.in
@@ -46,20 +42,15 @@ jmespath==1.0.1
     #   botocore
 lazy-loader==0.4
     # via scikit-image
-llvmlite==0.40.1
-    # via numba
 markupsafe==3.0.2
     # via
     #   jinja2
     #   werkzeug
 networkx==3.4.2
     # via scikit-image
-numba==0.57.1
-    # via -r /workspaces/vectorizing/scripts/../requirements/prod.in
 numpy==1.24.4
     # via
     #   imageio
-    #   numba
     #   opencv-python-headless
     #   pypotrace
     #   pywavelets

scripts/install_system_dependencies.sh

@@ -1,5 +1,23 @@
 #!/bin/bash
-# this script install system dependencies for vectorizing
+# Install the system dependencies required by the production image and devcontainer.
 
-sudo apt-get update
-sudo apt-get install -y build-essential libagg-dev libpotrace-dev pkg-config libffi-dev libcairo2-dev
+set -euo pipefail
+
+if [[ "$(id -u)" -eq 0 ]]; then
+  APT_GET=(apt-get)
+else
+  APT_GET=(sudo apt-get)
+fi
+
+"${APT_GET[@]}" update
+"${APT_GET[@]}" install -y --no-install-recommends \
+  wget \
+  build-essential \
+  python3-dev \
+  libagg-dev \
+  libpotrace-dev \
+  pkg-config \
+  libgl1 \
+  libcairo2 \
+  libffi-dev \
+  libcairo2-dev

vectorizing/__init__.py

@@ -130,7 +130,7 @@ def index() -> str | Response | tuple[Response, int]:
             timer.end_timer()
 
             if raw:
-                return markup
+                return Response(markup, mimetype="image/svg+xml")
 
             timer.start_timer("Markup Upload")
             cuid_str = upload_markup(markup, S3_BUCKET)

vectorizing/tests/test_api_contracts.py

@@ -84,6 +84,7 @@ def test_raw_requests_crop_before_vectorizing_and_do_not_upload(
         },
     )
     assert response.status_code == 200
+    assert response.mimetype == "image/svg+xml"
     assert response.data.startswith(b"<svg ")
     assert b'width="8" height="8"' in response.data
     chosen, unused = (binary, color) if solver == 0 else (color, binary)