feat(BA-6083): add ResolveDomainIDByName and ResolveResourceGroupIDByName actions (#11660)

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: fregataa

changes/11660.feature.md

@@ -0,0 +1 @@
+Add `ResolveDomainIDByName` and `ResolveResourceGroupIDByName` actions to resolve entity names to their canonical UUID identifiers.

src/ai/backend/manager/repositories/domain/db_source/db_source.py

@@ -4,6 +4,7 @@
 
 import sqlalchemy as sa
 
+from ai.backend.common.identifier.domain import DomainID, DomainName
 from ai.backend.logging.utils import BraceStyleAdapter
 from ai.backend.manager.data.domain.types import DomainData
 from ai.backend.manager.errors.resource import DomainNotFound
@@ -41,6 +42,14 @@ async def get_domain(self, domain_name: str) -> DomainData:
                 raise DomainNotFound(f"Domain '{domain_name}' not found")
             return row.to_data()
 
+    async def get_domain_id_by_name(self, name: DomainName) -> DomainID:
+        async with self._db.begin_readonly_session_read_committed() as db_sess:
+            query = sa.select(DomainRow.id).where(DomainRow.name == name)
+            domain_id = await db_sess.scalar(query)
+            if domain_id is None:
+                raise DomainNotFound(f"Domain '{name}' not found")
+            return domain_id
+
     async def search_domains(self, querier: BatchQuerier) -> DomainSearchResult:
         """Search all domains with pagination and filters.
 

src/ai/backend/manager/repositories/domain/repository.py

@@ -7,6 +7,7 @@
 from sqlalchemy.ext.asyncio import AsyncSession as SASession
 
 from ai.backend.common.exception import BackendAIError, DomainNotFound, InvalidAPIParameters
+from ai.backend.common.identifier.domain import DomainID, DomainName
 from ai.backend.common.metrics.metric import DomainType, LayerType
 from ai.backend.common.resilience.policies.metrics import MetricArgs, MetricPolicy
 from ai.backend.common.resilience.policies.retry import BackoffStrategy, RetryArgs, RetryPolicy
@@ -380,6 +381,10 @@ async def _ensure_sgroup_permission(
 
     # ==================== V2 Repository Methods ====================
 
+    @domain_repository_resilience.apply()
+    async def get_domain_id_by_name(self, name: DomainName) -> DomainID:
+        return await self._db_source.get_domain_id_by_name(name)
+
     @domain_repository_resilience.apply()
     async def get_domain(self, domain_name: str) -> DomainData:
         """Get a single domain by name.

src/ai/backend/manager/repositories/scaling_group/db_source/db_source.py

@@ -9,7 +9,7 @@
 from sqlalchemy.dialects.postgresql import insert as pg_insert
 
 from ai.backend.common.data.permission.types import RBACElementType, RelationType
-from ai.backend.common.identifier.resource_group import ResourceGroupName
+from ai.backend.common.identifier.resource_group import ResourceGroupID, ResourceGroupName
 from ai.backend.common.types import SlotQuantity
 from ai.backend.manager.data.agent.types import AgentStatus
 from ai.backend.manager.data.deployment.types import DeploymentOptions
@@ -119,6 +119,14 @@ async def search_scaling_groups(
                 has_previous_page=result.has_previous_page,
             )
 
+    async def get_resource_group_id_by_name(self, name: ResourceGroupName) -> ResourceGroupID:
+        async with self._db.begin_readonly_session_read_committed() as db_sess:
+            query = sa.select(ScalingGroupRow.id).where(ScalingGroupRow.name == name)
+            resource_group_id = await db_sess.scalar(query)
+            if resource_group_id is None:
+                raise ScalingGroupNotFound(name)
+            return resource_group_id
+
     async def get_scaling_group_by_name(
         self,
         name: str,

src/ai/backend/manager/repositories/scaling_group/repository.py

@@ -3,7 +3,7 @@
 from typing import TYPE_CHECKING
 from uuid import UUID
 
-from ai.backend.common.identifier.resource_group import ResourceGroupName
+from ai.backend.common.identifier.resource_group import ResourceGroupID, ResourceGroupName
 from ai.backend.common.metrics.metric import DomainType, LayerType
 from ai.backend.common.resilience import (
     MetricArgs,
@@ -89,6 +89,10 @@ async def search_scaling_groups(
         """Searches scaling groups with total count."""
         return await self._db_source.search_scaling_groups(querier=querier)
 
+    @scaling_group_repository_resilience.apply()
+    async def get_resource_group_id_by_name(self, name: ResourceGroupName) -> ResourceGroupID:
+        return await self._db_source.get_resource_group_id_by_name(name)
+
     @scaling_group_repository_resilience.apply()
     async def get_scaling_group_by_name(
         self,

src/ai/backend/manager/services/domain/actions/resolve_domain_id_by_name.py

@@ -0,0 +1,32 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import override
+
+from ai.backend.common.identifier.domain import DomainID, DomainName
+from ai.backend.manager.actions.action import BaseActionResult
+from ai.backend.manager.actions.types import ActionOperationType
+from ai.backend.manager.services.domain.actions.base import DomainAction
+
+
+@dataclass
+class ResolveDomainIDByNameAction(DomainAction):
+    name: DomainName
+
+    @override
+    def entity_id(self) -> str | None:
+        return str(self.name)
+
+    @override
+    @classmethod
+    def operation_type(cls) -> ActionOperationType:
+        return ActionOperationType.GET
+
+
+@dataclass
+class ResolveDomainIDByNameActionResult(BaseActionResult):
+    domain_id: DomainID
+
+    @override
+    def entity_id(self) -> str | None:
+        return str(self.domain_id)

src/ai/backend/manager/services/domain/processors.py

@@ -32,6 +32,10 @@
     PurgeDomainAction,
     PurgeDomainActionResult,
 )
+from ai.backend.manager.services.domain.actions.resolve_domain_id_by_name import (
+    ResolveDomainIDByNameAction,
+    ResolveDomainIDByNameActionResult,
+)
 from ai.backend.manager.services.domain.actions.search_domains import (
     SearchDomainsAction,
     SearchDomainsActionResult,
@@ -54,6 +58,9 @@ class DomainProcessors(AbstractProcessorPackage):
     get_domain: ActionProcessor[GetDomainAction, GetDomainActionResult]
     search_domains: ActionProcessor[SearchDomainsAction, SearchDomainsActionResult]
     search_rg_domains: ActionProcessor[SearchRGDomainsAction, SearchRGDomainsActionResult]
+    resolve_domain_id_by_name: ActionProcessor[
+        ResolveDomainIDByNameAction, ResolveDomainIDByNameActionResult
+    ]
 
     def __init__(
         self,
@@ -70,6 +77,9 @@ def __init__(
         self.get_domain = ActionProcessor(service.get_domain, action_monitors)
         self.search_domains = ActionProcessor(service.search_domains, action_monitors)
         self.search_rg_domains = ActionProcessor(service.search_rg_domains, action_monitors)
+        self.resolve_domain_id_by_name = ActionProcessor(
+            service.resolve_domain_id_by_name, action_monitors
+        )
 
     @override
     def supported_actions(self) -> list[ActionSpec]:
@@ -83,4 +93,5 @@ def supported_actions(self) -> list[ActionSpec]:
             GetDomainAction.spec(),
             SearchDomainsAction.spec(),
             SearchRGDomainsAction.spec(),
+            ResolveDomainIDByNameAction.spec(),
         ]

src/ai/backend/manager/services/domain/service.py

@@ -35,6 +35,10 @@
     PurgeDomainAction,
     PurgeDomainActionResult,
 )
+from ai.backend.manager.services.domain.actions.resolve_domain_id_by_name import (
+    ResolveDomainIDByNameAction,
+    ResolveDomainIDByNameActionResult,
+)
 from ai.backend.manager.services.domain.actions.search_domains import (
     SearchDomainsAction,
     SearchDomainsActionResult,
@@ -126,6 +130,12 @@ async def modify_domain_node(
             domain_data=domain_data,
         )
 
+    async def resolve_domain_id_by_name(
+        self, action: ResolveDomainIDByNameAction
+    ) -> ResolveDomainIDByNameActionResult:
+        domain_id = await self._repository.get_domain_id_by_name(action.name)
+        return ResolveDomainIDByNameActionResult(domain_id=domain_id)
+
     async def get_domain(self, action: GetDomainAction) -> GetDomainActionResult:
         """Get a single domain by name.
 

src/ai/backend/manager/services/scaling_group/actions/resolve_resource_group_id_by_name.py

@@ -0,0 +1,32 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import override
+
+from ai.backend.common.identifier.resource_group import ResourceGroupID, ResourceGroupName
+from ai.backend.manager.actions.action import BaseActionResult
+from ai.backend.manager.actions.types import ActionOperationType
+from ai.backend.manager.services.scaling_group.actions.base import ScalingGroupAction
+
+
+@dataclass(frozen=True)
+class ResolveResourceGroupIDByNameAction(ScalingGroupAction):
+    name: ResourceGroupName
+
+    @override
+    def entity_id(self) -> str | None:
+        return str(self.name)
+
+    @override
+    @classmethod
+    def operation_type(cls) -> ActionOperationType:
+        return ActionOperationType.GET
+
+
+@dataclass(frozen=True)
+class ResolveResourceGroupIDByNameActionResult(BaseActionResult):
+    resource_group_id: ResourceGroupID
+
+    @override
+    def entity_id(self) -> str | None:
+        return str(self.resource_group_id)

src/ai/backend/manager/services/scaling_group/processors.py

@@ -80,6 +80,10 @@
     ReplaceDefaultSessionOptionsAction,
     ReplaceDefaultSessionOptionsActionResult,
 )
+from ai.backend.manager.services.scaling_group.actions.resolve_resource_group_id_by_name import (
+    ResolveResourceGroupIDByNameAction,
+    ResolveResourceGroupIDByNameActionResult,
+)
 from ai.backend.manager.services.scaling_group.actions.update_allowed_domains_for_rg import (
     UpdateAllowedDomainsForResourceGroupAction,
     UpdateAllowedDomainsForResourceGroupActionResult,
@@ -177,6 +181,10 @@ class ScalingGroupProcessors(AbstractProcessorPackage):
         GetAllowedProjectsForResourceGroupAction,
         GetAllowedProjectsForResourceGroupActionResult,
     ]
+    resolve_resource_group_id_by_name: ActionProcessor[
+        ResolveResourceGroupIDByNameAction,
+        ResolveResourceGroupIDByNameActionResult,
+    ]
 
     def __init__(
         self,
@@ -242,6 +250,9 @@ def __init__(
         self.get_allowed_projects_for_rg = ActionProcessor(
             service.get_allowed_projects_for_resource_group, action_monitors
         )
+        self.resolve_resource_group_id_by_name = ActionProcessor(
+            service.resolve_resource_group_id_by_name, action_monitors
+        )
 
     @override
     def supported_actions(self) -> list[ActionSpec]:
@@ -270,4 +281,5 @@ def supported_actions(self) -> list[ActionSpec]:
             GetAllowedResourceGroupsForProjectAction.spec(),
             GetAllowedDomainsForResourceGroupAction.spec(),
             GetAllowedProjectsForResourceGroupAction.spec(),
+            ResolveResourceGroupIDByNameAction.spec(),
         ]