fix(compat): map legacy encoding "text" to "none"

common/lib/vcon_compat.py

@@ -22,6 +22,9 @@
 - attachment / dialog ``mimetype`` → ``mediatype``
 - extension/critical: ``must_support`` (top-level or in entries) →
   ``critical``
+- dialog / analysis / attachment ``encoding: "text"`` → ``"none"``. Not a
+  spec value (``base64url``, ``json``, ``none``); legacy producers use it for
+  a plain-string body, which is exactly what ``none`` means.
 
 The function is conservative: if both the legacy and the spec field are
 present, the spec field wins and the legacy field is dropped.
@@ -53,6 +56,8 @@ def _normalize_entry(entry: Dict[str, Any]) -> None:
     _rename(entry, "schema_version", "schema")
     _rename(entry, "mimetype", "mediatype")
     _rename(entry, "must_support", "critical")
+    if entry.get("encoding") == "text":
+        entry["encoding"] = "none"
 
 
 def _normalize_attachment(att: Dict[str, Any]) -> None:

common/tests/test_vcon_compat.py

@@ -101,3 +101,19 @@ def test_empty_or_missing_fields_safe():
 def test_returns_same_dict():
     v = {"uuid": "u"}
     assert normalize_legacy_fields(v) is v
+
+
+def test_legacy_text_encoding_to_none():
+    v = {
+        "uuid": "u",
+        "attachments": [
+            {"type": "searchable_text", "encoding": "text", "body": "2024 Impreza"},
+            {"type": "tags", "encoding": "json", "body": "[]"},
+        ],
+        "analysis": [{"type": "summary", "encoding": "text", "body": "s"}],
+    }
+    normalize_legacy_fields(v)
+    assert v["attachments"][0]["encoding"] == "none"
+    assert v["attachments"][0]["body"] == "2024 Impreza"
+    assert v["attachments"][1]["encoding"] == "json"
+    assert v["analysis"][0]["encoding"] == "none"