Dockerfile

# Browser-Only Version (No User Authentication)
# Use Python 3.11 slim image for better compatibility
FROM python:3.11-slim

# Set environment variables
ENV PYTHONUNBUFFERED=1
ENV DEBIAN_FRONTEND=noninteractive
ENV PORT=7860

# Install system dependencies
RUN apt-get update && apt-get install -y \
    git \
    libmagic1 \
    libmagic-dev \
    file \
    gcc \
    g++ \
    libc6-dev \
    libffi-dev \
    libjpeg-dev \
    libpng-dev \
    libtiff-dev \
    libwebp-dev \
    zlib1g-dev \
    && rm -rf /var/lib/apt/lists/*

# Set work directory
WORKDIR /app

# Clone the repository
RUN git clone https://github.com/wjbmattingly/VLAMy.git .

# Install Python dependencies
RUN pip install --no-cache-dir -r requirements.txt

# Create necessary directories
RUN mkdir -p /app/media/imports /app/staticfiles /app/logs

# Create custom settings for no-auth mode
RUN echo 'from .settings import *\n\
\n\
# Disable authentication requirements but keep auth apps for model dependencies\n\
REST_FRAMEWORK = {\n\
    "DEFAULT_AUTHENTICATION_CLASSES": [],\n\
    "DEFAULT_PERMISSION_CLASSES": [\n\
        "rest_framework.permissions.AllowAny",\n\
    ],\n\
    "DEFAULT_PAGINATION_CLASS": "rest_framework.pagination.PageNumberPagination",\n\
    "PAGE_SIZE": 20,\n\
}\n\
\n\
# Use in-memory database for simplicity\n\
DATABASES = {\n\
    "default": {\n\
        "ENGINE": "django.db.backends.sqlite3",\n\
        "NAME": ":memory:",\n\
    }\n\
}\n\
\n\
# Enhanced cache settings for browser-only mode\n\
CACHES = {\n\
    "default": {\n\
        "BACKEND": "django.core.cache.backends.locmem.LocMemCache",\n\
        "LOCATION": "vlamy-cache",\n\
        "TIMEOUT": 86400,  # 24 hours\n\
        "OPTIONS": {\n\
            "MAX_ENTRIES": 1000,\n\
        }\n\
    }\n\
}\n\
\n\
# Keep auth middleware but disable auth requirements\n\
MIDDLEWARE = [\n\
    "corsheaders.middleware.CorsMiddleware",\n\
    "django.middleware.security.SecurityMiddleware",\n\
    "django.contrib.sessions.middleware.SessionMiddleware",\n\
    "django.middleware.common.CommonMiddleware",\n\
    "django.middleware.csrf.CsrfViewMiddleware",\n\
    "django.contrib.auth.middleware.AuthenticationMiddleware",\n\
    "django.contrib.messages.middleware.MessageMiddleware",\n\
    # Removed XFrameOptionsMiddleware to allow iframe embedding\n\
]\n\
\n\
# Keep all apps installed to avoid model dependency issues\n\
INSTALLED_APPS = [\n\
    "django.contrib.admin",\n\
    "django.contrib.auth",\n\
    "django.contrib.contenttypes",\n\
    "django.contrib.sessions",\n\
    "django.contrib.messages",\n\
    "django.contrib.staticfiles",\n\
    "rest_framework",\n\
    "rest_framework.authtoken",\n\
    "corsheaders",\n\
    "ocr_app",\n\
]\n\
\n\
# Allow all CORS origins for browser-only mode\n\
CORS_ALLOW_ALL_ORIGINS = True\n\
CORS_ALLOW_CREDENTIALS = False\n\
\n\
# Static files configuration for no-auth mode\n\
import os\n\
STATIC_URL = "/static/"\n\
STATIC_ROOT = os.path.join(BASE_DIR, "staticfiles")\n\
STATICFILES_DIRS = [\n\
    os.path.join(BASE_DIR, "static"),\n\
]\n\
\n\
# Media files configuration\n\
MEDIA_URL = "/media/"\n\
MEDIA_ROOT = os.path.join(BASE_DIR, "media")\n\
\n\
# Enable static file serving in development\n\
DEBUG = True\n\
\n\
# Session settings for browser storage\n\
SESSION_ENGINE = "django.contrib.sessions.backends.cache"\n\
SESSION_CACHE_ALIAS = "default"\n\
SESSION_COOKIE_AGE = 86400  # 24 hours\n\
\n\
# Disable file logging to avoid permission issues\n\
LOGGING = {\n\
    "version": 1,\n\
    "disable_existing_loggers": False,\n\
    "handlers": {\n\
        "console": {\n\
            "level": "INFO",\n\
            "class": "logging.StreamHandler",\n\
        },\n\
    },\n\
    "loggers": {\n\
        "ocr_app": {\n\
            "handlers": ["console"],\n\
            "level": "INFO",\n\
            "propagate": True,\n\
        },\n\
    },\n\
}\n\
\n\
# Create anonymous user middleware reference\n\
MIDDLEWARE.insert(0, "vlamy_ocr.middleware.AnonymousUserMiddleware")\n\
\n\
# Settings for HuggingFace Spaces iframe embedding\n\
X_FRAME_OPTIONS = "ALLOWALL"  # Allow iframe embedding\n\
SECURE_CROSS_ORIGIN_OPENER_POLICY = None\n\
SECURE_REFERRER_POLICY = "same-origin"\n\
\n\
# Additional security settings for iframe embedding\n\
SECURE_CONTENT_TYPE_NOSNIFF = False\n\
SECURE_BROWSER_XSS_FILTER = False\n\
\n\
# Allow HuggingFace Spaces in CSP\n\
CSP_FRAME_ANCESTORS = ["*"]\n\
CSP_DEFAULT_SRC = ["*"]\n\
\n\
# Disable Django security warnings for iframe embedding\n\
import warnings\n\
warnings.filterwarnings("ignore", module="django.security")\n\
\n\
' > /app/vlamy_ocr/settings_no_auth.py

# Create a custom URL configuration for no-auth mode
RUN echo 'from django.urls import path, include\n\
from django.conf import settings\n\
from django.conf.urls.static import static\n\
from django.views.generic import TemplateView\n\
from django.contrib.staticfiles.urls import staticfiles_urlpatterns\n\
import os\n\
\n\
urlpatterns = [\n\
    # API endpoints (no auth required - permissions handled by settings)\n\
    path("api/", include("ocr_app.urls")),\n\
    \n\
    # Frontend routes (single-page application)\n\
    path("", TemplateView.as_view(template_name="index.html"), name="home"),\n\
    path("app/", TemplateView.as_view(template_name="index.html"), name="app"),\n\
]\n\
\n\
# Always serve static and media files (essential for no-auth mode)\n\
urlpatterns += static(settings.STATIC_URL, document_root=settings.STATIC_ROOT)\n\
urlpatterns += static(settings.MEDIA_URL, document_root=settings.MEDIA_ROOT)\n\
urlpatterns += staticfiles_urlpatterns()\n\
\n\
# Debug info\n\
print(f"Static URL: {settings.STATIC_URL}")\n\
print(f"Static Root: {settings.STATIC_ROOT}")\n\
static_dirs = getattr(settings, "STATICFILES_DIRS", [])\n\
print(f"Static Dirs: {static_dirs}")\n\
' > /app/vlamy_ocr/urls_no_auth.py

# Ensure static directories exist and collect static files
RUN mkdir -p /app/static /app/staticfiles && \
    DJANGO_SETTINGS_MODULE=vlamy_ocr.settings_no_auth python manage.py collectstatic --noinput --verbosity=2

# Replace the main urls.py with no-auth version at build time
RUN cp /app/vlamy_ocr/urls_no_auth.py /app/vlamy_ocr/urls.py

# Create anonymous user middleware for no-auth mode
RUN echo 'import os\n\
from django.contrib.auth.models import User, AnonymousUser\n\
from django.core.management import execute_from_command_line\n\
from django.db import connection\n\
from django.apps import apps\n\
\n\
class AnonymousUserMiddleware:\n\
    """Middleware that automatically handles migrations and creates an anonymous user for all requests"""\n\
    \n\
    def __init__(self, get_response):\n\
        self.get_response = get_response\n\
        self._anonymous_user = None\n\
        self._migrations_applied = False\n\
        self._setup_complete = False\n\
    \n\
    def _ensure_migrations_and_user(self):\n\
        """Ensure migrations are applied and anonymous user exists"""\n\
        if self._setup_complete:\n\
            return\n\
            \n\
        try:\n\
            # Check if migrations need to be applied\n\
            if not self._migrations_applied:\n\
                print("Applying database migrations...")\n\
                os.environ.setdefault("DJANGO_SETTINGS_MODULE", "vlamy_ocr.settings_no_auth")\n\
                execute_from_command_line(["manage.py", "migrate", "--noinput"])\n\
                self._migrations_applied = True\n\
                print("Migrations applied successfully")\n\
            \n\
            # Create anonymous user\n\
            if not self._anonymous_user:\n\
                self._anonymous_user, created = User.objects.get_or_create(\n\
                    username="anonymous",\n\
                    defaults={\n\
                        "email": "[email protected]",\n\
                        "first_name": "Anonymous",\n\
                        "last_name": "User"\n\
                    }\n\
                )\n\
                \n\
                # Ensure user profile exists and is approved\n\
                if hasattr(self._anonymous_user, "profile"):\n\
                    if not self._anonymous_user.profile.is_approved:\n\
                        self._anonymous_user.profile.is_approved = True\n\
                        self._anonymous_user.profile.save()\n\
                else:\n\
                    from ocr_app.models import UserProfile\n\
                    profile, profile_created = UserProfile.objects.get_or_create(\n\
                        user=self._anonymous_user,\n\
                        defaults={"is_approved": True}\n\
                    )\n\
                \n\
                print(f"Anonymous user ready: created={created}, user_id={self._anonymous_user.id}")\n\
                \n\
            self._setup_complete = True\n\
            \n\
        except Exception as e:\n\
            print(f"Error in middleware setup: {e}")\n\
            # Fallback to Django AnonymousUser\n\
            self._anonymous_user = AnonymousUser()\n\
    \n\
    def __call__(self, request):\n\
        # Ensure setup is complete\n\
        self._ensure_migrations_and_user()\n\
        \n\
        # Always assign the anonymous user to the request\n\
        request.user = self._anonymous_user\n\
        \n\
        response = self.get_response(request)\n\
        return response\n\
' > /app/vlamy_ocr/middleware.py

# Create a startup script for no-auth mode
RUN echo '#!/bin/bash\n\
\n\
# Set the custom settings module\n\
export DJANGO_SETTINGS_MODULE=vlamy_ocr.settings_no_auth\n\
\n\
echo "Starting VLAMy in browser-only mode (no authentication required)"\n\
echo "Anonymous user will be created automatically on first request"\n\
\n\
# Start the Django development server (migrations will run automatically)\n\
python manage.py runserver 0.0.0.0:$PORT\n\
' > /app/start.sh

# Set proper permissions for all files and ensure startup script is executable  
RUN chmod -R 755 /app && \
    chmod +x /app/start.sh && \
    chown -R root:root /app

# Environment variables for production (no-auth mode)
ENV DEBUG=True
ENV SECRET_KEY="browser-only-key-no-sensitive-data"
ENV ALLOWED_HOSTS="*"
ENV DJANGO_SETTINGS_MODULE="vlamy_ocr.settings_no_auth"

# Expose port
EXPOSE 7860

# Start the application
CMD ["/app/start.sh"]