Add

app.py

@@ -1,6 +1,9 @@
+import base64
 import io
+import urllib.request
 from pathlib import Path
 from typing import List
+from urllib.parse import urlparse
 
 import numpy as np
 import torch
@@ -125,6 +128,41 @@ def predict_tensorflow(image: Image.Image):
     return predicted_class, confidence, all_probs
 
 
+def load_image_from_url(image_url: str):
+    if not image_url or not image_url.strip():
+        raise ValueError("URL vide ou invalide.")
+
+    if image_url.startswith("data:"):
+        try:
+            header, encoded = image_url.split(",", 1)
+            if "base64" in header:
+                image_data = base64.b64decode(encoded)
+            else:
+                image_data = urllib.request.unquote_to_bytes(encoded)
+            return Image.open(io.BytesIO(image_data)).convert("RGB")
+        except Exception as exc:
+            raise ValueError(f"Impossible de lire le data URL: {exc}")
+
+    parsed = urlparse(image_url)
+    if parsed.scheme not in ("http", "https"):
+        raise ValueError("L'URL doit commencer par http:// ou https://")
+
+    try:
+        request = urllib.request.Request(
+            image_url,
+            headers={"User-Agent": "GeoClassifier/1.0"},
+        )
+        with urllib.request.urlopen(request, timeout=15) as response:
+            image_data = response.read()
+    except Exception as exc:
+        raise ValueError(f"Impossible de récupérer l'image depuis l'URL: {exc}")
+
+    try:
+        return Image.open(io.BytesIO(image_data)).convert("RGB")
+    except Exception as exc:
+        raise ValueError(f"Impossible de lire l'image depuis l'URL: {exc}")
+
+
 def classify(image: Image.Image, model_choice: str):
     if model_choice == "pytorch":
         return predict_pytorch(image)
@@ -145,15 +183,28 @@ def health_check():
 
 
 @app.post("/predict")
-def predict(image: UploadFile = File(...), model_choice: str = Form("pytorch")):
-    if image.content_type.split('/')[0] != 'image':
-        raise HTTPException(status_code=400, detail="Le fichier doit être une image.")
-
-    image_data = image.file.read()
-    try:
-        img = Image.open(io.BytesIO(image_data)).convert("RGB")
-    except Exception as exc:
-        raise HTTPException(status_code=400, detail=f"Impossible de lire l'image: {exc}")
+def predict(
+    image: UploadFile = File(None),
+    image_url: str = Form(None),
+    model_choice: str = Form("pytorch")
+):
+    if image is None and not image_url:
+        raise HTTPException(status_code=400, detail="Le fichier ou l'URL est requis.")
+
+    if image is not None:
+        if image.content_type.split('/')[0] != 'image':
+            raise HTTPException(status_code=400, detail="Le fichier doit être une image.")
+
+        image_data = image.file.read()
+        try:
+            img = Image.open(io.BytesIO(image_data)).convert("RGB")
+        except Exception as exc:
+            raise HTTPException(status_code=400, detail=f"Impossible de lire l'image: {exc}")
+    else:
+        try:
+            img = load_image_from_url(image_url)
+        except ValueError as exc:
+            raise HTTPException(status_code=400, detail=str(exc))
 
     predicted_class, confidence, all_probs = classify(img, model_choice)
     return {

index.html

@@ -618,7 +618,11 @@ html::-webkit-scrollbar {
     }
 
     function toggleUrlInput() {
-      document.getElementById('url-input-wrap').classList.toggle('hidden');
+      const wrap = document.getElementById('url-input-wrap');
+      wrap.classList.toggle('hidden');
+      if (wrap.classList.contains('hidden')) {
+        urlInput.value = '';
+      }
     }
 
     function resetAll() {
@@ -650,8 +654,26 @@ html::-webkit-scrollbar {
 
       if (!file && !urlVal) { showError(t('selectImage')); return; }
 
+      // Validation basique de l'URL
+      if (!file && urlVal) {
+        try {
+          const url = new URL(urlVal);
+          if (!['http:', 'https:', 'data:'].includes(url.protocol)) {
+            showError('URL invalide. Utilisez http://, https:// ou data:');
+            return;
+          }
+        } catch (e) {
+          showError('URL invalide. Vérifiez le format.');
+          return;
+        }
+      }
+
       const formData = new FormData();
-      if (file) formData.append('image', file);
+      if (file) {
+        formData.append('image', file);
+      } else {
+        formData.append('image_url', urlVal);
+      }
       formData.append('model_choice', modelChoice);
 
       setProcessing(true);