FIX scanner URL encoding: urllib.parse.quote pour URLs non-ASCII (méthodologie/adhérence retournaient DOWN alors que 200 en curl) · root cause urllib ne encode pas automatiquement

screens-health-check.py

@@ -5,6 +5,7 @@ import json, os, sys, time, re, tempfile
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from urllib.request import Request, urlopen
 from urllib.error import URLError, HTTPError
+from urllib.parse import quote, urlparse, urlunparse
 
 CARTO_FILE = "/var/www/html/cartographie-screens.html"
 OUT_FILE = "/var/www/html/api/screens-health.json"
@@ -49,6 +50,14 @@ def classify(code, elapsed_ms):
 def check_one(entry):
     url = entry.get("url")
     name = entry.get("name")
+    # Percent-encode non-ASCII chars in path (fixes DOWN on URLs like méthodologie, adhérence)
+    try:
+        p = urlparse(url)
+        if any(ord(c) > 127 for c in p.path):
+            encoded_path = quote(p.path, safe="/")
+            url = urlunparse((p.scheme, p.netloc, encoded_path, p.params, p.query, p.fragment))
+    except Exception:
+        pass
     t0 = time.time()
     code = 0
     try: