Fix: Hold Restrictions

src/aleph_client/commands/instance/__init__.py

@@ -335,6 +335,16 @@
             selector=True,
         ),
     )
+    if (tier.compute_units > 4 or confidential) and payment_type == PaymentType.hold:
+        console.print("VM with more than 4 Compute unit and/or confidential can't run using HOLD.", style="red")
+        if payment_chain in super_token_chains:
+            payment_type = PaymentType.superfluid
+            console.print("Switching payment type to PAY-As-You-Go (superfluid).", style="green")
+        else:
+            console.print("The current chain is not compatible with PAYG. Aborting instance creation.", style="red")
+            console.print(f"Compatible Chain : {super_token_chains}")
+            raise typer.Exit(code=1)
+
     name = name or validated_prompt("Instance name", lambda x: x and len(x) < 65)
     vcpus = tier.vcpus
     memory = tier.memory

src/aleph_client/commands/pricing.py

@@ -211,9 +211,13 @@ def display_table_for(
                 if "vram" in tier:
                     row.append(f"{tier['vram'] / 1024:.0f}")
                 if "holding" in price_unit:
-                    row.append(
-                        f"{displayable_amount(Decimal(price_unit['holding']) * current_units, decimals=3)} tokens"
-                    )
+                    # If the pricing entity is confidential, display "Not Available"
+                    if pricing_entity == PricingEntity.INSTANCE_CONFIDENTIAL:
+                        row.append("Not Available")
+                    else:
+                        row.append(
+                            f"{displayable_amount(Decimal(price_unit['holding']) * current_units, decimals=3)} tokens"
+                        )
                 if "payg" in price_unit and pricing_entity in PAYG_GROUP:
                     payg_hourly = Decimal(price_unit["payg"]) * current_units
                     row.append(

src/aleph_client/commands/program.py

@@ -182,6 +182,7 @@ async def upload(
                 verbose=verbose,
             ),
         )
+
         name = name or validated_prompt("Program name", lambda x: x and len(x) < 65)
         vcpus = tier.vcpus
         memory = tier.memory

tests/unit/test_instance.py

@@ -7,6 +7,7 @@
 from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
+import typer
 from aiohttp import InvalidURL
 from aleph_message.models import Chain, ItemHash
 from aleph_message.models.execution.base import Payment, PaymentType
@@ -380,7 +381,7 @@ def create_mock_vm_coco_client():
         "coco_superfluid_evm",
         "gpu_superfluid_evm",
     ],
-    argnames="args, expected",
+    argnames="args, expected, should_raise",
     argvalues=[
         (  # regular_hold_evm
             {
@@ -392,6 +393,7 @@ def create_mock_vm_coco_client():
                 "immutable_volume": [f"mount=/opt/packages,ref={FAKE_STORE_HASH}"],
             },
             (FAKE_VM_HASH, None, "ETH"),
+            False,
         ),
         (  # regular_superfluid_evm
             {
@@ -401,6 +403,7 @@ def create_mock_vm_coco_client():
                 "crn_url": FAKE_CRN_URL,
             },
             (FAKE_VM_HASH, FAKE_CRN_URL, "AVAX"),
+            False,
         ),
         (  # regular_hold_sol
             {
@@ -409,6 +412,7 @@ def create_mock_vm_coco_client():
                 "rootfs": "debian12",
             },
             (FAKE_VM_HASH, None, "SOL"),
+            False,
         ),
         (  # coco_hold_sol
             {
@@ -418,7 +422,8 @@ def create_mock_vm_coco_client():
                 "crn_url": FAKE_CRN_URL,
                 "confidential": True,
             },
-            (FAKE_VM_HASH, FAKE_CRN_URL, "SOL"),
+            None,
+            True,
         ),
         (  # coco_hold_evm
             {
@@ -428,7 +433,8 @@ def create_mock_vm_coco_client():
                 "crn_url": FAKE_CRN_URL,
                 "confidential": True,
             },
-            (FAKE_VM_HASH, FAKE_CRN_URL, "ETH"),
+            None,
+            True,
         ),
         (  # coco_superfluid_evm
             {
@@ -439,6 +445,7 @@ def create_mock_vm_coco_client():
                 "confidential": True,
             },
             (FAKE_VM_HASH, FAKE_CRN_URL, "BASE"),
+            False,
         ),
         (  # gpu_superfluid_evm
             {
@@ -449,11 +456,12 @@ def create_mock_vm_coco_client():
                 "gpu": True,
             },
             (FAKE_VM_HASH, FAKE_CRN_URL, "BASE"),
+            False,
         ),
     ],
 )
 @pytest.mark.asyncio
-async def test_create_instance(args, expected):
+async def test_create_instance(args, expected, should_raise):
     mock_validate_ssh_pubkey_file = create_mock_validate_ssh_pubkey_file()
     mock_load_account = create_mock_load_account()
     mock_account = mock_load_account.return_value
@@ -496,28 +504,36 @@ async def create_instance(instance_spec):
         all_args.update(instance_spec)
         return await create(**all_args)
 
-    returned = await create_instance(args)
-    # Basic assertions for all cases
-    mock_load_account.assert_called_once()
-    mock_validate_ssh_pubkey_file.return_value.read_text.assert_called_once()
-    mock_client.get_estimated_price.assert_called_once()
-    mock_auth_client.create_instance.assert_called_once()
-    # Payment type specific assertions
-    if args["payment_type"] == "hold":
-        mock_get_balance.assert_called_once()
-    elif args["payment_type"] == "superfluid":
-        assert mock_account.manage_flow.call_count == 2
-        assert mock_wait_for_confirmed_flow.call_count == 2
-    # CRN related assertions
-    if args["payment_type"] == "superfluid" or args.get("confidential") or args.get("gpu"):
-        mock_fetch_latest_crn_version.assert_called()
-        if not args.get("gpu"):
-            mock_fetch_crn_info.assert_called_once()
-        else:
-            mock_crn_table.return_value.run_async.assert_called_once()
-        mock_wait_for_processed_instance.assert_called_once()
-        mock_vm_client.start_instance.assert_called_once()
-    assert returned == expected
+    if should_raise:
+        with pytest.raises(typer.Exit) as exc_info:
+            returned = await create_instance(args)
+            mock_load_account.assert_called_once()
+            mock_validate_ssh_pubkey_file.return_value.read_text.assert_called_once()
+
+        assert exc_info.value.exit_code == 1
+    else:
+        returned = await create_instance(args)
+        # Basic assertions for all cases
+        mock_load_account.assert_called_once()
+        mock_validate_ssh_pubkey_file.return_value.read_text.assert_called_once()
+        mock_client.get_estimated_price.assert_called_once()
+        mock_auth_client.create_instance.assert_called_once()
+        # Payment type specific assertions
+        if args["payment_type"] == "hold":
+            mock_get_balance.assert_called_once()
+        elif args["payment_type"] == "superfluid":
+            assert mock_account.manage_flow.call_count == 2
+            assert mock_wait_for_confirmed_flow.call_count == 2
+        # CRN related assertions
+        if args["payment_type"] == "superfluid" or args.get("confidential") or args.get("gpu"):
+            mock_fetch_latest_crn_version.assert_called()
+            if not args.get("gpu"):
+                mock_fetch_crn_info.assert_called_once()
+            else:
+                mock_crn_table.return_value.run_async.assert_called_once()
+            mock_wait_for_processed_instance.assert_called_once()
+            mock_vm_client.start_instance.assert_called_once()
+        assert returned == expected
 
 
 @pytest.mark.asyncio