API PDF upload from Dash->Localstack

README.md

@@ -20,13 +20,13 @@ EU Fact Force is a collaborative platform developed by [EUPHA](https://www.eupha
 ###  Use Case
 
 > **Marie**, a health communicator at a national public health association, sees a viral post claiming "vaccines cause autism." She needs to respond quickly with solid evidence.
-> 
+>
 > She searches **"vaccines autism"** on EU Fact Force and immediately sees:
 > - An interactive graph showing 15+ peer-reviewed articles that refute this claim
 > - The scientific consensus: **"Refuted with high confidence"**
 > - Current disinformation trends: 1,200 mentions this week, peak in France/Belgium
 > - Key evidence to cite in her response
-> 
+>
 > **Time to find relevant evidence: <30 seconds**
 
 ## Key Features
@@ -112,7 +112,7 @@ uv run pytest
 
 ### Déploiement de l'application
 
-L'application se compose d'un serveur Django, d'une base PostgreSQL (avec pgvector) et de LocalStack pour le stockage S3. 
+L'application se compose d'un serveur Django, d'une base PostgreSQL (avec pgvector) et de LocalStack pour le stockage S3.
 Pour déployer et utiliser l'application en local :
 
 **1. Prérequis**
@@ -176,3 +176,44 @@ AWS_S3_REGION_NAME=eu-west-1
 ```
 
 Sans ces variables, l'application utilise le stockage fichier local par défaut.
+
+
+**7. Démarrer la web-app d'ingestion vers le S3 local**
+
+Pour rapatrier l'upload d'un couple PDF/métadatas :
+
+***Lancer le containeur Docker***
+```bash
+docker compose up -d
+```
+
+Cela démarre PostgreSQL (port 5432) et LocalStack S3 (port 4566) et écoute sur
+ce port.
+Le bucket configuré est créé automatiquement au démarrage de LocalStack.
+
+***Installer les dépendances et appliquer les migrations***
+
+```bash
+uv sync
+uv run python manage.py migrate
+```
+
+***Lancer le pipeline Dash-app > API > Localstack***
+
+***Démarrer le serveur Django :***
+
+```bash
+uv run python manage.py runserver
+```
+
+***Démarrer le script FastAPI***
+
+```bash
+uv run python ingestion/front_upload/api_front_upload.py
+```
+
+***Démarrer la webapp Dash***
+
+```bash
+uv run python ingestion/front_upload/api_front_upload.py
+```

docker-compose.yml

@@ -48,20 +48,22 @@ services:
       retries: 5
 
   localstack:
-    image: localstack/localstack:latest
-    ports:
-      - 4566
-    environment:
-      SERVICES: s3
-      PERSISTENCE: 1
-      AWS_DEFAULT_REGION: ${AWS_S3_REGION_NAME:-eu-west-1}
-      AWS_STORAGE_BUCKET_NAME: ${AWS_STORAGE_BUCKET_NAME:-eu-fact-force-files}
-      AWS_ACCESS_KEY_ID: ${AWS_ACCESS_KEY_ID:-test}
-      AWS_SECRET_ACCESS_KEY: ${AWS_SECRET_ACCESS_KEY:-test}
-      DEBUG: ${DEBUG:-0}
-    volumes:
-      - ./s3:/var/lib/localstack
-      - ./docker/localstack/init-ready.d:/etc/localstack/init/ready.d:ro
+      image: localstack/localstack:3.3 #latest conflicted w/ FastAPI
+      ports:
+        - "4566:4566" #required listening port for FastAPI
+      environment:
+        SERVICES: s3
+        PERSISTENCE: 1
+        AWS_DEFAULT_REGION: ${AWS_S3_REGION_NAME:-eu-west-1}
+        AWS_STORAGE_BUCKET_NAME: ${AWS_STORAGE_BUCKET_NAME:-eu-fact-force-files}
+        AWS_ACCESS_KEY_ID: ${AWS_ACCESS_KEY_ID:-test}
+        AWS_SECRET_ACCESS_KEY: ${AWS_SECRET_ACCESS_KEY:-test}
+        DEBUG: ${DEBUG:-0}
+        LOCALSTACK_ACKNOWLEDGE_ACCOUNT_REQUIREMENT: 1 #required licensing
+      volumes:
+        - ./s3:/var/lib/localstack
+        - ./docker/localstack/init-ready.d:/etc/localstack/init/ready.d:ro
+        - /var/run/docker.sock:/var/run/docker.sock
 
 volumes:
   postgres_data:

eu_fact_force/dash-app/api/upload_to_s3.py

@@ -0,0 +1,95 @@
+import os
+import json
+from fastapi import FastAPI, File, UploadFile, Form, HTTPException
+from botocore.client import Config
+from fastapi.middleware.cors import CORSMiddleware
+import boto3
+import uvicorn
+from dotenv import load_dotenv
+
+# 1. Environment var loading
+load_dotenv()
+
+app = FastAPI(title="EUFactForce API")
+
+# 2. Dash-app URL
+origins = [
+    "http://localhost:8050",
+]
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=origins,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# 3. S3 instancing
+s3_client = boto3.client(
+    "s3",
+    endpoint_url=os.getenv("AWS_S3_ENDPOINT_URL"),
+    aws_access_key_id=os.getenv("AWS_ACCESS_KEY_ID"),
+    aws_secret_access_key=os.getenv("AWS_SECRET_ACCESS_KEY"),
+    region_name=os.getenv("AWS_REGION"),
+    config=Config(s3={'addressing_style': 'path'}) # <-- Ajoute cette ligne impérativement
+)
+
+BUCKET_NAME = os.getenv("AWS_STORAGE_BUCKET_NAME")
+
+# API print debugging
+try:
+    s3_client.create_bucket(Bucket=BUCKET_NAME)
+    print(f"Bucket '{BUCKET_NAME}' créé ou déjà existant.")
+except Exception as e:
+    print(f"Note: Le bucket existe peut-être déjà : {e}")
+
+@app.get("/")
+async def root():
+    return {"message": "API EUFactForce opérationnelle"}
+
+# 4. Upload routine
+@app.post("/upload/")
+async def upload_file(
+    file: UploadFile = File(...),
+    metadata: str = Form(...)
+):
+    try:
+        # filename cleanup
+        filename = file.filename
+        json_filename = f"{os.path.splitext(filename)[0]}.json"
+
+        # A. PDF upload on S3
+        file_content = await file.read()
+        s3_client.put_object(
+            Bucket=BUCKET_NAME,
+            Key=filename,
+            Body=file_content,
+            ContentType="application/pdf"
+        )
+
+        # B. JSON Metadatas S3 upload
+        # Json type check
+        try:
+            json_data = json.loads(metadata)
+        except json.JSONDecodeError:
+            raise HTTPException(status_code=400, detail="Métadonnées JSON invalides")
+
+        s3_client.put_object(
+            Bucket=BUCKET_NAME,
+            Key=json_filename,
+            Body=json.dumps(json_data),
+            ContentType="application/json"
+        )
+
+        return {
+            "status": "success",
+            "message": f"Fichiers {filename} et {json_filename} téléchargés avec succès."
+        }
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+if __name__ == "__main__":
+    # Uvicorn server exposition on 8001
+    uvicorn.run(app, host="0.0.0.0", port=8001)

eu_fact_force/dash-app/app.py

@@ -5,9 +5,12 @@
 import plotly.io as pio
 import plotly.graph_objects as go
 
+# PDF ingestion
 import base64
 import io
 import json
+from pathlib import Path
+import requests
 import uuid
 
 from utils.colors import EUPHAColors
@@ -16,7 +19,7 @@
 from pages import readme, ingest, graph
 
 # Plotly template
-with open("assets/template.json", "r") as f:
+with open(Path(__file__).parent / "assets/template.json", "r") as f:
     debate_template = json.load(f)
 pio.templates["app_template"] = go.layout.Template(debate_template)
 pio.templates.default = "app_template"
@@ -368,6 +371,9 @@ def lock_authors(is_correct, ids):
 @app.callback(
     Output('final-output', 'children'),
     Input('btn-final-upload', 'n_clicks'),
+    # AJOUT DES DEUX STATES MANQUANTS ICI
+    State('upload-pdf', 'contents'),
+    State('upload-pdf', 'filename'),
     State('input-doi', 'value'),
     State('input-abstract', 'value'),
     State('input-journal', 'value'),
@@ -381,14 +387,20 @@ def lock_authors(is_correct, ids):
     State({'type': 'auth-email', 'index': ALL}, 'value'),
     prevent_initial_call=True
 )
-def finalize_and_display_json(n_clicks, doi, abstract, journal, date, link, category, study_type, title, names, surnames, emails):
+def finalize_and_send(n_clicks, pdf_base64, filename, doi, abstract, journal, date, link, category, study_type, title, names, surnames, emails):
+    # Dash vérifie maintenant que pdf_base64 reçoit bien 'contents' et filename reçoit 'filename'
+
+    if not n_clicks or pdf_base64 is None:
+        return no_update
+
+    print(f"Tentative d'envoi pour : {filename}") # DEBUG CLI
 
     authors_list = [
         {"name": n, "surname": s, "email": e}
         for n, s, e in zip(names, surnames, emails) if n or s
     ]
 
-    metadata_json = {
+    metadata_payload = {
         "title": title,
         "category": category,
         "study_type": study_type,
@@ -400,11 +412,31 @@ def finalize_and_display_json(n_clicks, doi, abstract, journal, date, link, cate
         "authors": authors_list
     }
 
-    return html.Div([
-        dbc.Alert("Successfully contributed, thank you!", color="success"),
-        html.H4("Metadata JSON"),
-        html.Pre(json.dumps(metadata_json, indent=4), style={'backgroundColor': '#f8f9fa', 'padding': '15px', 'borderRadius': '8px', 'border': '1px solid #dee2e6'})
-                    ])
+    try:
+        # Décodage propre
+        content_type, content_string = pdf_base64.split(',')
+        pdf_bytes = base64.b64decode(content_string)
+
+        url = "http://localhost:8001/upload/"
+
+        files = {
+            'file': (filename, pdf_bytes, 'application/pdf')
+        }
+        data = {
+            'metadata': json.dumps(metadata_payload)
+        }
+
+        # Timeout ajouté pour éviter que Dash ne freeze si FastAPI est éteint
+        response = requests.post(url, files=files, data=data, timeout=10)
+
+        if response.status_code == 200:
+            return dbc.Alert(f"Succès ! {filename} est sur S3.", color="success")
+        else:
+            return dbc.Alert(f"Erreur API : {response.text}", color="danger")
+
+    except Exception as e:
+        print(f"Erreur détaillée : {str(e)}") # Visible dans ton terminal Dash
+        return dbc.Alert(f"Erreur lors de l'envoi : {str(e)}", color="danger")
 
 if __name__ == "__main__":
     app.run(debug=True)

pyproject.toml

@@ -17,6 +17,9 @@ dependencies = [
     "dash-bootstrap-components>=2.0.4",
     "dash-cytoscape>=1.0.2",
     "pymupdf>=1.27.1",
+    "fastapi>=0.135.2",
+    "uvicorn>=0.42.0",
+    "python-multipart>=0.0.22",
 ]
 
 [tool.pytest.ini_options]