Add web interface for localhost usage

Dockerfile.webapp

@@ -0,0 +1,30 @@
+FROM python:3.13-slim
+
+# Install LibreOffice for local conversion (optional, can use the separate container)
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+    libreoffice \
+    curl \
+    && rm -rf /var/lib/apt/lists/*
+
+# Set working directory
+WORKDIR /app
+
+# Install UV first
+RUN pip install --no-cache-dir uv
+
+# Copy project files
+COPY pyproject.toml ./
+COPY src/ ./src/
+
+# Install dependencies (without frozen lock to allow updates)
+RUN uv sync
+
+# Set Python path
+ENV PYTHONPATH=/app
+
+# Expose port
+EXPOSE 8000
+
+# Run the web application
+CMD ["uv", "run", "python", "-m", "uvicorn", "src.webapp:app", "--host", "0.0.0.0", "--port", "8000"]

LOCALHOST_GUIDE.md

@@ -0,0 +1,166 @@
+# Running PPT2Desc on Localhost
+
+This guide explains how to run the PPT2Desc web application on your local machine.
+
+## Quick Start
+
+### Option 1: Using Docker Compose (Recommended)
+
+This is the easiest way to get started. Both LibreOffice converter and the web application will run in containers.
+
+1. **Start the services:**
+   ```bash
+   docker compose up -d
+   ```
+
+2. **Access the web interface:**
+   - Open your browser and navigate to: **http://localhost:5001**
+
+3. **Stop the services:**
+   ```bash
+   docker compose down
+   ```
+
+That's it! The web interface will be available at http://localhost:5001, and you can upload PowerPoint files directly through your browser.
+
+### Option 2: Running Locally with UV
+
+If you prefer to run the application directly on your machine without Docker:
+
+1. **Install dependencies:**
+   ```bash
+   uv sync
+   ```
+
+2. **Start the LibreOffice converter (optional, if you want to use Docker-based conversion):**
+   ```bash
+   docker compose up -d libreoffice-converter
+   ```
+
+3. **Run the web application:**
+   ```bash
+   uv run uvicorn src.webapp:app --host 0.0.0.0 --port 5001
+   ```
+
+4. **Access the web interface:**
+   - Open your browser and navigate to: **http://localhost:5001**
+
+## Using the Web Interface
+
+Once the application is running, you can:
+
+1. **Upload a PowerPoint file** (.ppt or .pptx)
+2. **Select an AI provider** (Gemini, OpenAI, Anthropic, etc.)
+3. **Configure model settings** (API keys, model name, etc.)
+4. **Add optional instructions** to customize the output
+5. **Click "Convert Presentation"** to process your file
+
+The results will be displayed directly in the browser, showing detailed descriptions for each slide.
+
+The web service runs on **port 5001** by default.
+
+## Configuration Options
+
+### AI Provider Settings
+
+The web interface supports multiple AI providers:
+
+- **Google Gemini API**: Requires API key
+- **Google Vertex AI**: Requires GCP project ID, region, and service account credentials
+- **OpenAI**: Requires API key
+- **Anthropic Claude**: Requires API key
+- **Azure OpenAI**: Requires API key, endpoint, and deployment name
+- **AWS Bedrock**: Requires access key ID, secret access key, and region
+
+### LibreOffice Configuration
+
+By default, the web application uses the Docker-based LibreOffice converter at `http://libreoffice-converter:2002` (when using Docker Compose) or `http://localhost:2002` (when running locally).
+
+If you have LibreOffice installed locally, you can leave the LibreOffice URL field blank, and the application will attempt to find it in your system PATH.
+
+## API Endpoints
+
+If you want to integrate the service programmatically:
+
+### Health Check
+```bash
+curl http://localhost:5001/health
+```
+
+### Convert Presentation
+```bash
+curl -X POST http://localhost:5001/convert \
+  -F "file=@presentation.pptx" \
+  -F "client=gemini" \
+  -F "api_key=YOUR_API_KEY" \
+  -F "model=gemini-2.5-flash"
+```
+
+## Troubleshooting
+
+### Port Already in Use
+
+If port 5001 is already in use, you can change it:
+
+**Docker Compose:**
+Edit `docker-compose.yml` and change the port mapping:
+```yaml
+ports:
+  - "5002:8000"  # Change 5002 to any available port
+```
+
+**Local Running:**
+```bash
+uv run uvicorn src.webapp:app --host 0.0.0.0 --port 5002
+```
+
+### LibreOffice Connection Issues
+
+If you get errors about LibreOffice conversion:
+
+1. Make sure the LibreOffice converter is running:
+   ```bash
+   docker compose ps
+   ```
+
+2. Check the health of the converter:
+   ```bash
+   curl http://localhost:2002/health
+   ```
+
+3. If using local LibreOffice, ensure it's installed:
+   ```bash
+   which soffice
+   # or
+   which libreoffice
+   ```
+
+### Memory Issues
+
+For large presentations or high rate limits, you may need to increase Docker memory limits. Edit your Docker settings or add resource limits to `docker-compose.yml`.
+
+## Development
+
+To run in development mode with auto-reload:
+
+```bash
+uv run uvicorn src.webapp:app --host 0.0.0.0 --port 5001 --reload
+```
+
+## Environment Variables
+
+You can set default values using environment variables:
+
+```bash
+export GEMINI_API_KEY=your_api_key
+export OPENAI_API_KEY=your_api_key
+export ANTHROPIC_API_KEY=your_api_key
+```
+
+Then you won't need to enter API keys in the web interface each time.
+
+## Next Steps
+
+- Check the main [README.md](README.md) for detailed information about the project
+- Learn about customizing prompts and instructions
+- Explore the CLI version for batch processing

README.md

@@ -10,6 +10,8 @@ ppt2desc is a command-line tool that converts PowerPoint presentations into deta
 
 ## Features
 
+- **Web Interface**: Easy-to-use browser-based interface for converting presentations
+- **CLI Tool**: Command-line interface for batch processing and automation
 - Convert PPT/PPTX files to semantic descriptions
 - Process individual files or entire directories
 - Support for visual elements interpretation (charts, graphs, figures)
@@ -80,6 +82,26 @@ This will create a virtual environment and install all dependencies from `pyproj
 
 ## Usage
 
+### Web Interface (Recommended for Quick Start)
+
+The easiest way to use ppt2desc is through the web interface:
+
+1. **Start the web application:**
+   ```bash
+   docker compose up -d
+   ```
+
+2. **Open your browser and navigate to:**
+   ```
+   http://localhost:5001
+   ```
+
+3. **Upload your PowerPoint file, configure your AI provider, and convert!**
+
+For detailed instructions, see [LOCALHOST_GUIDE.md](LOCALHOST_GUIDE.md).
+
+### Command Line Interface
+
 Basic usage with Gemini API:
 ```bash
 uv run src/main.py \

docker-compose.yml

@@ -1,6 +1,6 @@
 services:
   libreoffice-converter:
-    build: 
+    build:
       context: ./src/libreoffice_docker
       dockerfile: Dockerfile
     ports:
@@ -11,4 +11,21 @@ services:
       test: ["CMD", "curl", "-f", "http://localhost:2002/health"]
       interval: 300s
       timeout: 10s
+      retries: 3
+
+  ppt2desc-web:
+    build:
+      context: .
+      dockerfile: Dockerfile.webapp
+    ports:
+      - "5001:8000"
+    restart: unless-stopped
+    depends_on:
+      - libreoffice-converter
+    environment:
+      - LIBREOFFICE_URL=http://libreoffice-converter:2002
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
+      interval: 30s
+      timeout: 10s
       retries: 3

pyproject.toml

@@ -15,6 +15,7 @@ dependencies = [
     "charset-normalizer==3.4.1",
     "distro==1.9.0",
     "docstring-parser==0.16",
+    "fastapi>=0.115.0",
     "google-ai-generativelanguage==0.6.10",
     "google-api-core==2.24.0",
     "google-api-python-client==2.156.0",
@@ -53,6 +54,7 @@ dependencies = [
     "pymupdf==1.25.1",
     "pyparsing==3.2.1",
     "python-dateutil==2.9.0.post0",
+    "python-multipart>=0.0.12",
     "requests==2.32.3",
     "rsa==4.9",
     "s3transfer==0.10.4",
@@ -63,6 +65,7 @@ dependencies = [
     "typing-extensions==4.12.2",
     "uritemplate==4.1.1",
     "urllib3==2.3.0",
+    "uvicorn>=0.32.0",
     "pytest==8.3.3",
     "pytest-mock==3.14.0",
 ]

src/__init__.py

@@ -0,0 +1 @@
+# Package initialization

src/processor.py

@@ -10,11 +10,19 @@
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from tqdm import tqdm
 
-from llm import LLMClient
-from converters.ppt_converter import convert_pptx_to_pdf
-from converters.pdf_converter import convert_pdf_to_images
-from converters.docker_converter import convert_pptx_via_docker
-from schemas.deck import DeckData, SlideData
+# Support both relative imports (for webapp) and absolute imports (for main.py)
+try:
+    from .llm import LLMClient
+    from .converters.ppt_converter import convert_pptx_to_pdf
+    from .converters.pdf_converter import convert_pdf_to_images
+    from .converters.docker_converter import convert_pptx_via_docker
+    from .schemas.deck import DeckData, SlideData
+except ImportError:
+    from llm import LLMClient
+    from converters.ppt_converter import convert_pptx_to_pdf
+    from converters.pdf_converter import convert_pdf_to_images
+    from converters.docker_converter import convert_pptx_via_docker
+    from schemas.deck import DeckData, SlideData
 
 # Create a type alias for all possible clients
 logger = logging.getLogger(__name__)