feat: implement Google Drive uploader and image processing pipeline f…

.env.example

@@ -0,0 +1,3 @@
+SESSIONIZE_API_SLUG=prcjw6ue
+GDRIVE_CREDENTIALS_PATH=path/to/service-account.json
+GDRIVE_FOLDER_ID=your-google-drive-folder-id

.github/workflows/generate-cards.yml

@@ -0,0 +1,40 @@
+name: Generate Speaker Cards
+
+on:
+  schedule:
+    - cron: '0 6 * * *'
+  workflow_dispatch:
+
+jobs:
+  generate-and-upload:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+          cache-dependency-glob: "uv.lock"
+
+      - name: Install Python
+        run: uv python install
+
+      - name: Install dependencies
+        run: uv sync --locked
+
+      - name: Write Google credentials file
+        # We store the JSON string in a GitHub Secret and write it to a temp file
+        # This file is NEVER committed to the repository.
+        run: echo '${{ secrets.GDRIVE_CREDENTIALS_JSON }}' > /tmp/gdrive-creds.json
+
+      - name: Generate cards and upload
+        env:
+          SESSIONIZE_API_SLUG: ${{ secrets.SESSIONIZE_API_SLUG }}
+          GDRIVE_CREDENTIALS_PATH: /tmp/gdrive-creds.json
+          GDRIVE_FOLDER_ID: ${{ secrets.GDRIVE_FOLDER_ID }}
+        run: uv run python src/generate_cards.py --upload
+
+      - name: Cleanup credentials
+        if: always()
+        run: rm -f /tmp/gdrive-creds.json

.gitignore

@@ -1 +1,6 @@
 node_modules/
+__pycache__/
+.venv/
+*.pyc
+.env
+output/

AGENTS.md

@@ -0,0 +1,50 @@
+# AGENTS.md
+
+## 🤖 Persona & Description
+
+**Name:** `card-generator-agent`
+**Description:** Automates the generation of DevBcn speaker cards by parsing REST API data, removing image backgrounds, and compositing text.
+**Persona:** You are a senior Python developer specializing in image processing and automation. You write modular, clean, and type-hinted code using `requests`, `rembg`, and `Pillow`.
+
+## ⚡ Commands
+
+* **Setup environment:** `uv sync`
+* **Run generator:** `uv run python src/generate_cards.py`
+* **Run specific test:** `uv run pytest tests/ -v`
+* **Lint code:** `uv run flake8 src/`
+
+## 🚫 Boundaries
+
+* **Read-Only Assets:** Never modify, resize, or overwrite the base template located at `assets/base_template.jpg`.
+* **Output Restrictions:** Always save generated output images strictly to the `output/` directory.
+* **Secrets:** Never hardcode API URLs or endpoints in the Python scripts. Always use `os.getenv()` and an `.env` file.
+* **Logic Constraints:** Do not attempt to write custom background removal algorithms; always rely on the `rembg` library.
+
+## 📂 Project Structure
+
+* `src/`: Core Python logic (`api_client.py`, `image_processor.py`, `main.py`).
+* `assets/`: Static required files (`base_template.jpg`, `.ttf` fonts).
+* `output/`: Directory for the final generated `.png` or `.jpg` speaker cards.
+* `tests/`: Unit tests for API mocking and image manipulation logic.
+
+## ✍️ Code Style & Examples
+
+Write modular functions with explicit Python type hinting.
+
+**Good Output Example (Typography with Drop Shadow):**
+
+```python
+from PIL import ImageDraw, ImageFont
+
+def draw_text_with_shadow(
+    draw: ImageDraw.ImageDraw, 
+    text: str, 
+    position: tuple[int, int], 
+    font: ImageFont.FreeTypeFont, 
+    text_color: str = "white", 
+    shadow_color: str = "black"
+) -> None:
+    x, y = position
+    draw.text((x + 2, y + 2), text, font=font, fill=shadow_color)
+    draw.text((x, y), text, font=font, fill=text_color)
+```

docs/pipeline.md

@@ -0,0 +1,73 @@
+# Speaker Card Generator Pipeline
+
+This document describes the data flow and logic of the speaker card generation pipeline.
+
+## Overview
+
+The pipeline automates the creation of conference speaker cards by fetching data from Sessionize, removing backgrounds from speaker photos, and compositing them onto a template with word-wrapped title text.
+
+## Modules
+
+### `src/api_client.py`
+
+- Fetches all data in a single request from the `/view/All` Sessionize endpoint.
+- Maps speaker data to `Speaker` dataclasses.
+- Filters and maps session data to `SessionCard` dataclasses.
+- Skips service and plenum sessions.
+
+### `src/image_processor.py`
+
+- **Background Removal**: Uses `rembg` with a shared session to efficiently remove backgrounds from profile pictures.
+- **Normalization**: Crops transparent margins and resizes the subject to a standardized height (`500px`).
+- **Compositing**:
+  - Handles single and dual speaker layouts.
+  - Positions subjects relative to an anchor line.
+  - Uses alpha channels as masks for clean blending.
+- **Text Rendering**:
+  - join speaker names with " & ".
+  - Word-wraps the talk title to fit within safe margins.
+  - Applies a drop-shadow effect for legibility.
+
+### `src/gdrive_uploader.py`
+
+- **Authentication**: Uses Google Service Account JSON credentials.
+- **Upsert Logic**: Searches for existing files by name in the target folder.
+  - If found: Updates the existing file.
+  - If not found: Uploads as a new file.
+- **MIME Type**: Enforces `image/png` for all uploads.
+
+### `src/generate_cards.py`
+
+- Entry point that orchestrates the workflow.
+- Loads configuration from `.env`.
+- Iterates over session cards and saves the final PNGs to the `output/` directory.
+
+## Requirements
+
+- Python 3.11+
+- `uv` for dependency management.
+- ML model for `rembg` (downloaded automatically on first run, ~170MB).
+
+## Usage
+
+1. Place `base_template.png` and your font file in `assets/`.
+2. Configure coordinates in `src/image_processor.py`.
+3. Set `SESSIONIZE_API_SLUG` in `.env`.
+4. Run: `uv run python src/generate_cards.py`.
+5. For upload: `uv run python src/generate_cards.py --upload` (requires `.env` config).
+
+## CI/CD Automation
+
+A GitHub Actions workflow (`.github/workflows/generate-cards.yml`) is configured to run daily at 06:00 UTC.
+
+### Required GitHub Secrets
+
+- `SESSIONIZE_API_SLUG`: The Sessionize API slug.
+- `GDRIVE_CREDENTIALS_JSON`: The full JSON content of the Google Service Account key.
+- `GDRIVE_FOLDER_ID`: The ID of the target Google Drive folder.
+
+### Automation Logic
+
+1. **Caching**: Uses `setup-uv` to cache Python dependencies based on `uv.lock`.
+2. **Ephemeral Credentials**: The `GDRIVE_CREDENTIALS_JSON` secret is written to `/tmp/gdrive-creds.json` during the run and deleted afterward.
+3. **Daily Generation**: Automatically fetches current data from Sessionize and refreshes the cards in Google Drive.

pyproject.toml

@@ -0,0 +1,38 @@
+[project]
+name = "cover-maker"
+version = "0.1.0"
+description = "DevBcn Speaker Card Generator"
+authors = [
+    { name = "DevBcn Team" }
+]
+requires-python = ">=3.11"
+dependencies = [
+    "requests>=2.31",
+    "rembg[cpu]>=2.0",
+    "Pillow>=10.0",
+    "python-dotenv>=1.0",
+    "google-api-python-client>=2.100",
+    "google-auth>=2.23",
+]
+
+[dependency-groups]
+dev = [
+    "pytest>=8.0",
+    "pytest-cov>=5.0",
+    "ruff>=0.4",
+    "responses>=0.25",
+]
+
+[tool.ruff]
+line-length = 100
+target-version = "py311"
+
+[tool.ruff.lint]
+select = ["E", "F", "I", "N", "UP", "ANN"]
+
+[tool.ruff.lint.per-file-ignores]
+"tests/*" = ["ANN"]
+
+[tool.pytest.ini_options]
+addopts = "--cov=src --cov-report=term-missing --cov-fail-under=90"
+testpaths = ["tests"]

src/api_client.py

@@ -0,0 +1,42 @@
+import requests
+
+from models import SessionCard, Speaker
+
+SESSIONIZE_BASE_URL = "https://sessionize.com/api/v2"
+
+
+def fetch_session_cards(api_slug: str) -> list[SessionCard]:
+    url = f"{SESSIONIZE_BASE_URL}/{api_slug}/view/All"
+    response = requests.get(url)
+    response.raise_for_status()
+    data = response.json()
+
+    # Map speakers by ID
+    speaker_lookup: dict[str, Speaker] = {}
+    for speaker_data in data.get("speakers", []):
+        speaker_id = speaker_data.get("id")
+        speaker_lookup[speaker_id] = Speaker(
+            id=speaker_id,
+            full_name=speaker_data.get("fullName", ""),
+            profile_picture_url=speaker_data.get("profilePicture", ""),
+        )
+
+    cards: list[SessionCard] = []
+    for session_data in data.get("sessions", []):
+        # Skip service and plenum sessions
+        if session_data.get("isServiceSession") or session_data.get("isPlenumSession"):
+            continue
+
+        session_speakers = []
+        for speaker_id in session_data.get("speakers", []):
+            if speaker_id in speaker_lookup:
+                session_speakers.append(speaker_lookup[speaker_id])
+
+        if session_speakers:
+            cards.append(
+                SessionCard(
+                    talk_title=session_data.get("title", ""), speakers=tuple(session_speakers)
+                )
+            )
+
+    return cards

src/gdrive_uploader.py

@@ -0,0 +1,73 @@
+import logging
+import os
+from pathlib import Path
+
+from google.oauth2 import service_account
+from googleapiclient.discovery import build
+from googleapiclient.http import MediaFileUpload
+
+# Configure logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+def upload_output_folder(
+    credentials_path: str, folder_id: str, output_dir: Path = Path("output")
+) -> int:
+    """
+    Uploads all PNG files from output_dir to a Google Drive folder.
+    Updates existing files if name matches.
+    """
+    if not output_dir.exists():
+        logger.warning(f"Output directory {output_dir} does not exist. Skipping upload.")
+        return 0
+
+    # 1. Authenticate
+    try:
+        creds = service_account.Credentials.from_service_account_file(
+            credentials_path, scopes=["https://www.googleapis.com/auth/drive.file"]
+        )
+        service = build("drive", "v3", credentials=creds)
+    except Exception as e:
+        logger.error(f"Failed to authenticate with Google Drive: {e}")
+        raise
+
+    # 2. Get existing files in folder to avoid duplicates (upsert logic)
+    query = f"'{folder_id}' in parents and trashed = false"
+    try:
+        results = (
+            service.files()
+            .list(q=query, spaces="drive", fields="files(id, name)")
+            .execute()
+        )
+    except Exception as e:
+        logger.error(f"Failed to list files in Google Drive folder {folder_id}: {e}")
+        raise
+
+    existing_files = {f["name"]: f["id"] for f in results.get("files", [])}
+
+    upload_count = 0
+    # 3. Iterate over PNGs in output/
+    for file_path in output_dir.glob("*.png"):
+        file_name = file_path.name
+        media = MediaFileUpload(str(file_path), mimetype="image/png")
+
+        try:
+            if file_name in existing_files:
+                # Update existing file
+                file_id = existing_files[file_name]
+                logger.info(f"Updating existing file: {file_name} (ID: {file_id})")
+                service.files().update(fileId=file_id, media_body=media).execute()
+            else:
+                # Create new file
+                logger.info(f"Uploading new file: {file_name}")
+                file_metadata = {"name": file_name, "parents": [folder_id]}
+                service.files().create(
+                    body=file_metadata, media_body=media, fields="id"
+                ).execute()
+
+            upload_count += 1
+        except Exception as e:
+            logger.error(f"Error uploading {file_name}: {e}")
+
+    return upload_count