align_from_ayoub: dedoublonne les domaines case/accent-insensitive

Fusionne 'Flux Libre'/'flux libre', 'Péage'/'peage'/'PeagE' en gardant la
forme propre (avec accents et capitale). Update domain_environments.domain_id
vers le keeper et supprime les doublons.

tools/align_from_ayoub.py

@@ -82,13 +82,25 @@ def slugify(s, maxlen=10):
     return ascii_str[:maxlen] or None
 
 
+def norm_domain_key(s):
+    """Cle de normalisation domaine: lowercase + sans accent + trim."""
+    if not s:
+        return ""
+    nfkd = unicodedata.normalize("NFKD", s.strip())
+    ascii_str = "".join(c for c in nfkd if not unicodedata.combining(c))
+    return ascii_str.lower()
+
+
 def get_or_create_domain(conn, name):
-    row = conn.execute(text("SELECT id, code FROM domains WHERE name=:n"),
-                       {"n": name}).fetchone()
-    if row:
-        return row.id
+    # Match case/accent-insensitive pour fusionner "Péage" et "peage"
+    key = norm_domain_key(name)
+    existing = conn.execute(text("SELECT id FROM domains")).fetchall()
+    for r in existing:
+        r_name = conn.execute(text("SELECT name FROM domains WHERE id=:i"),
+                              {"i": r.id}).fetchone().name
+        if norm_domain_key(r_name) == key:
+            return r.id
     code = slugify(name, 10)
-    # Eviter collision de code
     suffix = 0
     base_code = code
     while conn.execute(text("SELECT 1 FROM domains WHERE code=:c"),
@@ -102,6 +114,32 @@ def get_or_create_domain(conn, name):
                         {"n": name}).fetchone().id
 
 
+def merge_domain_duplicates(conn, dry_run=False):
+    """Fusionne les doublons 'Flux Libre'/'flux libre', 'Péage'/'peage' etc."""
+    rows = conn.execute(text("SELECT id, name FROM domains ORDER BY id")).fetchall()
+    groups = {}
+    for r in rows:
+        k = norm_domain_key(r.name)
+        groups.setdefault(k, []).append((r.id, r.name))
+    merged = 0
+    for k, items in groups.items():
+        if len(items) <= 1:
+            continue
+        # Garde le plus "propre" (avec accent/capitale en priorite)
+        items.sort(key=lambda x: (x[1] == x[1].lower(), x[0]))
+        keeper_id, keeper_name = items[0]
+        for dup_id, dup_name in items[1:]:
+            print(f"  [MERGE] domain {dup_name!r} (id={dup_id}) -> {keeper_name!r} (id={keeper_id})")
+            if not dry_run:
+                conn.execute(text(
+                    "UPDATE domain_environments SET domain_id=:k WHERE domain_id=:d"
+                ), {"k": keeper_id, "d": dup_id})
+                # Supprime le doublon (si plus reference)
+                conn.execute(text("DELETE FROM domains WHERE id=:d"), {"d": dup_id})
+            merged += 1
+    return merged
+
+
 def get_or_create_env(conn, name):
     row = conn.execute(text("SELECT id FROM environments WHERE name=:n"),
                        {"n": name}).fetchone()
@@ -148,6 +186,14 @@ def main():
 
     conn = engine.connect().execution_options(isolation_level="AUTOCOMMIT")
 
+    # 0. Fusion prealable des doublons de domaines (Flux Libre / flux libre, Peage / Péage)
+    print("\n[INFO] Fusion doublons domains (case/accent-insensitive)...")
+    merged = merge_domain_duplicates(conn, dry_run=args.dry_run)
+    if merged:
+        print(f"[INFO] {merged} doublons {'(DRY) ' if args.dry_run else ''}fusionnes")
+    else:
+        print("[INFO] Pas de doublon detecte")
+
     wb = openpyxl.load_workbook(args.xlsx_path, data_only=True)
     if args.sheet not in wb.sheetnames:
         print(f"[ERR] Sheet '{args.sheet}' introuvable. Sheets: {wb.sheetnames}")
@@ -212,13 +258,18 @@ def main():
         de_id = srv.domain_env_id
         if dom_name and env_name:
             if dom_name not in seen_dom:
-                existing = conn.execute(text("SELECT id FROM domains WHERE name=:n"),
-                                        {"n": dom_name}).fetchone()
-                if existing:
-                    seen_dom[dom_name] = existing.id
+                # Match case/accent-insensitive
+                key = norm_domain_key(dom_name)
+                existing_id = None
+                for r in conn.execute(text("SELECT id, name FROM domains")).fetchall():
+                    if norm_domain_key(r.name) == key:
+                        existing_id = r.id
+                        break
+                if existing_id:
+                    seen_dom[dom_name] = existing_id
                 elif args.dry_run:
                     would_create_dom.add(dom_name)
-                    seen_dom[dom_name] = -1  # sera cree au vrai run
+                    seen_dom[dom_name] = -1
                 else:
                     seen_dom[dom_name] = get_or_create_domain(conn, dom_name)
                     stats["dom_created"] += 1