Why does GridDB Cloud Python client throw GSException: invalid RowKey when inserting rows with explicit UUID keys?

I’m testing **GridDB Cloud with the Python client** (griddb_python) and want to use UUIDs as row keys instead of integers or strings. **Schema:**

import griddb_python as griddb
import uuid, datetime

factory = griddb.StoreFactory.get_instance()
store = factory.get_store(
    notification_member="...",
    cluster_name="...",
    user="...",
    password="..."
)

cols = [
    ("id", griddb.Type.STRING),  # intended for UUID string
    ("ts", griddb.Type.TIMESTAMP),
    ("value", griddb.Type.FLOAT)
]

container = store.put_container("uuid_test", cols, True)  # true = rowkey

**Insert attempt:**

row = (str(uuid.uuid4()), datetime.datetime.utcnow(), 42.5)
container.put(row)

**Error**:

[GSException] (errorCode=0x7d010001, message=RowKey value is invalid)

If I switch the row key to a plain string like "sensor_1", it works fine. **Question:** Are there restrictions on **row key formats in GridDB Cloud** (e.g., length, character set, UUID format)? If UUIDs aren’t allowed directly, what’s the recommended workaround store them as a secondary column with an index, or encode them differently before inserting? **What I’ve checked:** - GridDB docs state row keys can be STRING, INTEGER, or TIMESTAMP, but don’t clarify format restrictions on string keys. - Tried both Python uuid.uuid4() and hex string version (uuid.uuid4().hex) — same error. - Java client test with the same schema also rejects UUIDs as row keys.