Cleanup: remove unused code, add CI/CD workflow

- Remove flask-sse, Redis (unused)
- Remove ImageService, file_utils (unused)
- Remove /download-posters route (frontend handles download client-side)
- Remove LETTERBOXD_API_URL, secure_filename (dead imports)
- Fix duplicate TMDBService instantiation
- Simplify config.py (remove REDIS_URL)
- Fix gunicorn user home dir for control socket
- Add .gitignore for __pycache__
- Add .gitea/workflows/deploy.yml for tag-based CI/CD

.gitea/workflows/deploy.yml

@@ -0,0 +1,33 @@
+name: Deploy Backend
+
+on:
+  push:
+    tags:
+      - 'v*.*.*'
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Extract version tag
+        run: echo "VERSION_TAG=${GITHUB_REF_NAME}" >> $GITHUB_ENV
+
+      - name: Build Docker image
+        run: |
+          docker build \
+            -t poster-picker-backend:latest \
+            -t poster-picker-backend:${{ env.VERSION_TAG }} .
+
+      - name: Stop old container
+        run: |
+          cd ~/dev-server/poster-picker
+          docker compose stop backend || true
+
+      - name: Start new container
+        run: |
+          cd ~/dev-server/poster-picker
+          docker compose up -d --no-build backend

.gitignore

@@ -1,2 +1,5 @@
 .env
 .venv
+__pycache__/
+*.pyc
+*.pyo

Dockerfile

@@ -0,0 +1,30 @@
+# backend/Dockerfile
+FROM python:3.11-slim
+
+RUN addgroup --system app && adduser --system --ingroup app --home /home/app app
+
+WORKDIR /srv/app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+# Assurer permissions
+RUN chown -R app:app /srv/app
+
+USER app
+
+# Port interne
+EXPOSE 5000
+
+# Variable d'environnement par défaut
+ENV FLASK_ENV=production
+ENV GUNICORN_WORKERS=2
+ENV GUNICORN_THREADS=2
+ENV GUNICORN_WORKER_CLASS=gevent
+ENV GUNICORN_TIMEOUT=300
+
+# Commande de démarrage via gunicorn (attache l'app Flask : app:app)
+# Utilise timeout raisonnable, bind sur 0.0.0.0:5000
+CMD ["sh","-c","gunicorn --workers ${GUNICORN_WORKERS} --threads ${GUNICORN_THREADS} --worker-class ${GUNICORN_WORKER_CLASS} --timeout ${GUNICORN_TIMEOUT} --worker-tmp-dir /dev/shm --bind 0.0.0.0:5000 --access-logfile - --error-logfile - 'main:create_app()'"]

config.py

@@ -1,21 +1,14 @@
 import os
 
 class Config:
-    # Configuration générale
     SECRET_KEY = os.environ.get('SECRET_KEY', 'dev_key')
     DEBUG = False
     TESTING = False
-    REDIS_URL = os.environ.get("REDIS_URL", "redis://localhost:6379")
 
 class DevelopmentConfig(Config):
     DEBUG = True
-    # Configuration pour le développement local
-    REDIS_URL = "redis://localhost:6379"
 
 class ProductionConfig(Config):
-    # Configuration pour la production (Render, etc.)
-    REDIS_URL = os.environ.get("REDIS_URL")
-    # Peut désactiver le debug mode pour la prod
     DEBUG = False
 
 config = {

main.py

@@ -2,7 +2,6 @@ import os
 from flask import Flask
 from flask_cors import CORS
 from src.api.routes import api_bp
-from flask_sse import sse
 from config import config
 
 
@@ -13,17 +12,18 @@ def create_app():
         "http://localhost:3000",
         "http://127.0.0.1:3000"
     ]
+    frontend_url = os.environ.get("FRONTEND_URL")
+    if frontend_url:
+        origins.append(frontend_url)
     CORS(app, supports_credentials=True, origins=origins)
 
-    app.config.update(
-        SESSION_COOKIE_SAMESITE="None",   # necessary for cross-site cookies
-        SESSION_COOKIE_SECURE=False       # True in prod (HTTPS)
-    )
-
     env = os.environ.get('FLASK_ENV', 'default')
+
+    if env == "production":
+        app.config.update(SESSION_COOKIE_SAMESITE="None", SESSION_COOKIE_SECURE=True)
+
     app.config.from_object(config[env])
 
-    app.register_blueprint(sse, url_prefix='/stream')
     app.register_blueprint(api_bp, url_prefix='/api')
 
     return app

src/api/routes.py

@@ -1,21 +1,25 @@
 import io
 import math
 import os
-import zipfile
+import uuid
 
 import requests
-import uuid
-from flask import Blueprint, jsonify, request, session, send_file
-from werkzeug.utils import secure_filename
+from flask import Blueprint, jsonify, request, session
 from src.services.tmdb_service import TMDBService
-from src.services.image_service import ImageService
 import pandas as pd
 from functools import lru_cache
 
 api_bp = Blueprint('api', __name__)
 
 tmdb_service = TMDBService()
-image_service = ImageService()
+
+
+def get_user_id():
+    user_id = request.headers.get('X-User-ID') or session.get('user_id')
+    if not user_id:
+        user_id = str(uuid.uuid4())
+        session['user_id'] = user_id
+    return user_id
 
 ALLOWED_EXTENSIONS = {'csv'}
 
@@ -32,27 +36,45 @@ def get_poster_url(movie_name, movie_year):
 
 @api_bp.route('/movies', methods=['GET'])
 def get_movies():
-    user_id = session.get('user_id')  # Récupérer l'ID de l'utilisateur de la session
+    user_id = get_user_id()
 
     if user_id not in user_csvs:
         return jsonify({'movies': [], 'message': 'No movies found. Please upload a CSV file.'}), 200
 
-    # Lire le DataFrame à partir du cache
     csv_content = user_csvs[user_id]
 
     try:
-        # Si csv_content est une chaîne, il n'est pas nécessaire de décoder
         if isinstance(csv_content, bytes):
             df = pd.read_csv(io.StringIO(csv_content.decode('utf-8')))
         else:
             df = pd.read_csv(io.StringIO(csv_content))
 
-        # Reste de votre code pour la pagination et le traitement...
-        page = int(request.args.get('page', 1))
-        limit = int(request.args.get('limit', 10))
+        month = request.args.get('month')  # e.g. "2024-01"
 
         df = df.iloc[::-1].reset_index(drop=True)
         total_movies = len(df)
+
+        if month:
+            df['_date_parsed'] = pd.to_datetime(df['Watched Date'], errors='coerce')
+            df = df[df['_date_parsed'].dt.strftime('%Y-%m') == month]
+            df = df.sort_values('_date_parsed')
+            df = df.drop(columns=['_date_parsed'])
+            movies = df.to_dict('records')
+            movies_with_posters = [
+                {
+                    'id': i + 1,
+                    'Watched Date': movie['Watched Date'],
+                    'Name': movie['Name'],
+                    'Year': movie['Year'],
+                    'Poster': get_poster_url(movie['Name'], movie['Year']),
+                }
+                for i, movie in enumerate(movies)
+            ]
+            return jsonify({'movies': movies_with_posters, 'total': len(movies_with_posters)})
+
+        page = int(request.args.get('page', 1))
+        limit = int(request.args.get('limit', 10))
+
         total_pages = math.ceil(total_movies / limit)
 
         start_index = (page - 1) * limit
@@ -86,8 +108,6 @@ def get_movies():
 
 @api_bp.route('/posters/<movie_name>/<movie_year>', methods=['GET'])
 def get_movie_posters(movie_name, movie_year):
-    tmdb_service = TMDBService()
-
     movie = tmdb_service.search_movie(movie_name, movie_year)
 
     if movie:
@@ -103,11 +123,7 @@ def allowed_file(filename):
 
 @api_bp.route('/upload-csv', methods=['POST'])
 def upload_csv():
-    user_id = session.get('user_id')
-
-    if not user_id:
-            user_id = str(uuid.uuid4())
-            session['user_id'] = user_id
+    user_id = get_user_id()
 
     if 'file' not in request.files:
         return jsonify({'error': 'No file part'}), 400
@@ -120,11 +136,16 @@ def upload_csv():
 
         try:
             df = pd.read_csv(io.StringIO(content.decode('utf-8')))
-            df = df.rename(columns={
-                'Date': 'Watched Date',
-                'Name': 'Name',
-                'Year': 'Year'
-            })
+
+            if 'Watched Date' in df.columns:
+                pass
+            elif 'Date' in df.columns:
+                df = df.rename(columns={'Date': 'Watched Date'})
+            else:
+                return jsonify({'error': 'CSV must contain a "Watched Date" or "Date" column'}), 400
+
+            if 'Name' not in df.columns or 'Year' not in df.columns:
+                return jsonify({'error': 'CSV must contain "Name" and "Year" columns'}), 400
 
             user_csvs[user_id] = df.to_csv(index=False)
 
@@ -136,7 +157,7 @@ def upload_csv():
 
 @api_bp.route('/check-csv', methods=['GET'])
 def check_csv():
-    user_id = session.get('user_id')
+    user_id = get_user_id()
 
     try:
         if user_id in user_csvs:
@@ -150,7 +171,7 @@ def check_csv():
 
 @api_bp.route('/delete-csv', methods=['DELETE'])
 def delete_csv():
-    user_id = session.get('user_id')
+    user_id = get_user_id()
 
     try:
         if user_id in user_csvs:
@@ -163,106 +184,71 @@ def delete_csv():
         return jsonify({'error': 'An error occurred while deleting the CSV file'}), 500
 
 
-@api_bp.route('/fetch-diary', methods=['POST'])
-def fetch_diary_from_username():
-    """
-    Récupère le diary via l'API publique (api.hugo-pierret.be), transforme en CSV et
-    le stocke dans user_csvs[user_id] pour que le front continue à utiliser le même flux que l'upload.
-    """
+def _parse_letterboxd_rss(xml_bytes):
+    import xml.etree.ElementTree as ET
+    root = ET.fromstring(xml_bytes)
+    entries = []
+    for item in root.findall('.//item'):
+        watched_date = film_title = film_year = None
+        for child in item:
+            tag = child.tag.split('}')[-1] if '}' in child.tag else child.tag
+            if tag == 'watchedDate':
+                watched_date = child.text
+            elif tag == 'filmTitle':
+                film_title = child.text
+            elif tag == 'filmYear':
+                film_year = child.text
+        if watched_date and film_title:
+            entries.append({
+                'Watched Date': watched_date,
+                'Name': film_title,
+                'Year': film_year or '',
+            })
+    return entries
+
+
+@api_bp.route('/sync-rss', methods=['POST'])
+def sync_rss():
     try:
         data = request.get_json() or {}
-        username = data.get('username')
+        username = data.get('username', '').strip()
         if not username:
             return jsonify({'error': 'username required'}), 400
 
-        user_id = session.get('user_id')
-        if not user_id:
-            user_id = str(uuid.uuid4())
-            session['user_id'] = user_id
+        user_id = get_user_id()
 
-        api_url = f"https://api.hugo-pierret.be/letterboxd/diary?username={username}"
-        resp = requests.get(api_url)
+        rss_url = f"https://letterboxd.com/{username}/rss/"
+        resp = requests.get(rss_url, headers={'User-Agent': 'Mozilla/5.0'}, timeout=10)
         if resp.status_code != 200:
-            return jsonify({'error': 'Failed to fetch diary from upstream API'}), 502
+            return jsonify({'error': f'Could not fetch RSS feed (HTTP {resp.status_code})'}), 502
 
-        payload = resp.json()
-        entries = payload.get('entries', [])
-        if not entries:
-            return jsonify({'error': 'No entries returned by upstream API'}), 404
+        rss_entries = _parse_letterboxd_rss(resp.content)
+        if not rss_entries:
+            return jsonify({'error': 'No diary entries found in RSS feed'}), 404
 
-        rows = []
-        import re
-        for e in entries:
-            title = e.get('title') or ""
-            date = e.get('date') or ""
+        if user_id not in user_csvs:
+            df = pd.DataFrame(rss_entries)
+            df = df.sort_values('Watched Date').reset_index(drop=True)
+            user_csvs[user_id] = df.to_csv(index=False)
+            return jsonify({'success': True, 'fresh': True, 'added': len(rss_entries), 'total': len(rss_entries)}), 200
 
-            m = re.match(r"^(?P<name>.+?)\s*\((?P<year>\d{4})\)\s*$", title)
-            if m:
-                name = m.group('name').strip()
-                year = m.group('year')
-            else:
-                year_match = re.search(r"(\d{4})", title)
-                year = year_match.group(1) if year_match else ""
-                name = re.sub(r"\(\d{4}\)", "", title).strip()
-
-            rows.append({
-                "Watched Date": date,
-                "Name": name,
-                "Year": year
-            })
-
-        import pandas as pd, io
-        rows.reverse()
-        df = pd.DataFrame(rows)
-        csv_str = df.to_csv(index=False)
-
-        user_csvs[user_id] = csv_str
-
-        return jsonify({'success': True, 'message': 'Diary fetched and stored', 'total': len(rows)}), 200
-
-    except Exception as e:
-        print("Error in fetch_diary_from_username:", str(e))
-        return jsonify({'error': 'An internal error occurred'}), 500
-
-
-@api_bp.route('/download-posters', methods=['POST'])
-def download_posters():
-    try:
-        data = request.get_json()
-        selected_posters = data.get('posters', [])
-        
-        if not selected_posters:
-            return jsonify({'error': 'No posters selected'}), 400
-
-        zip_buffer = io.BytesIO()
-        with zipfile.ZipFile(zip_buffer, 'w', zipfile.ZIP_DEFLATED) as zip_file:
-            # Télécharger chaque poster
-            for poster in selected_posters:
-                movie_id = poster['movieId']
-                poster_id = poster['posterId']
-                watched_date = poster.get('watchedDate', 'unknown-date')
-
-                poster_url = f"https://image.tmdb.org/t/p/original{poster_id}"
-
-                try:
-                    response = requests.get(poster_url)
-                    if response.status_code == 200:
-                        file_name = f"{watched_date}_{movie_id}.jpg"
-                        zip_file.writestr(file_name, response.content)
-                    else:
-                        print(f"Failed to download poster for movie ID {movie_id}")
-                except Exception as e:
-                    print(f"Error downloading poster for movie ID {movie_id}: {str(e)}")
-
-        # Retourner le fichier ZIP
-        zip_buffer.seek(0)
-        return send_file(
-            zip_buffer,
-            mimetype='application/zip',
-            as_attachment=True,
-            download_name='posters.zip'
+        existing_content = user_csvs[user_id]
+        df_existing = pd.read_csv(
+            io.StringIO(existing_content.decode('utf-8') if isinstance(existing_content, bytes) else existing_content)
         )
 
+        existing_keys = set(zip(df_existing['Name'].astype(str), df_existing['Watched Date'].astype(str)))
+        new_rows = [e for e in rss_entries if (e['Name'], e['Watched Date']) not in existing_keys]
+
+        if not new_rows:
+            return jsonify({'success': True, 'fresh': False, 'added': 0, 'total': len(df_existing)}), 200
+
+        df_merged = pd.concat([df_existing, pd.DataFrame(new_rows)], ignore_index=True)
+        df_merged = df_merged.sort_values('Watched Date').reset_index(drop=True)
+        user_csvs[user_id] = df_merged.to_csv(index=False)
+
+        return jsonify({'success': True, 'fresh': False, 'added': len(new_rows), 'total': len(df_merged)}), 200
+
     except Exception as e:
-        print(f"Error in download_posters: {str(e)}")
-        return jsonify({'error': 'An error occurred while fetching movie posters'}), 500
+        print("Error in sync_rss:", str(e))
+        return jsonify({'error': 'An internal error occurred'}), 500

src/services/file_utils.py

@@ -1,10 +0,0 @@
-import re
-import os
-
-
-def clean_filename(filename):
-    return re.sub(r'[\\/*?:"<>|]', '_', filename)
-
-
-def ensure_directory(directory):
-    os.makedirs(directory, exist_ok=True)

src/services/image_service.py

@@ -1,14 +0,0 @@
-import requests
-from PIL import Image
-import io
-
-
-class ImageService:
-    def download_image(self, image_url, save_path=None):
-        response = requests.get(image_url, stream=True)
-        if response.status_code == 200:
-            image = Image.open(io.BytesIO(response.content))
-            if save_path:
-                image.save(save_path)
-            return image
-        return None

src/services/tmdb_service.py

@@ -24,7 +24,7 @@ class TMDBService:
         return None
     
     def get_movie_posters(self, movie_id):
-        url = f"{self.base_url}/movie/{movie_id}/images?include_image_language=en,fr,null"
+        url = f"{self.base_url}/movie/{movie_id}/images?include_image_language=en,fr,ja,null&language=null"
         headers = {
             "accept": "application/json",
             "Authorization": f"Bearer {self.access_token}"