fix: rebase error handling and registration

Author: ikemilian-lewis

src/orb/domain/template/factory.py

@@ -113,9 +113,9 @@ def create_template(
             except Exception as e:
                 if self._logger:
                     self._logger.error("Failed to create %s template: %s", provider_type, e)
-                # Fall back to core template
+                raise
 
-        # Fall back to core template
+        # Use the core template only when no provider-specific class is registered.
         try:
             template = Template(**template_data)
 

src/orb/domain/template/template_aggregate.py

@@ -74,6 +74,7 @@ class Template(BaseModel):
     # Tags and metadata
     tags: dict[str, Any] = Field(default_factory=dict)
     metadata: dict[str, Any] = Field(default_factory=dict)
+    provider_data: dict[str, Any] = Field(default_factory=dict)
 
     # Provider configuration (multi-provider support)
     provider_type: Optional[str] = None

src/orb/providers/azure/auth/azure_auth_strategy.py

@@ -7,14 +7,14 @@
 import asyncio
 from typing import Any, Optional
 
-from orb.domain.base.dependency_injection import injectable
 from orb.domain.base.ports import LoggingPort
 from orb.infrastructure.adapters.ports.auth import (
     AuthContext,
     AuthPort,
     AuthResult,
     AuthStatus,
 )
+from orb.infrastructure.di.injectable import injectable
 from orb.providers.azure.infrastructure.credential_factory import (
     AsyncAzureAccessTokenProviderProtocol,
     AsyncDefaultAzureAccessTokenProvider,

src/orb/providers/azure/exceptions/azure_exceptions.py

@@ -5,19 +5,37 @@
 
 from typing import Any, Optional
 
-from orb.domain.base.exceptions import InfrastructureError
+from orb.providers.base.exceptions import (
+    ProviderAuthError,
+    ProviderConfigError,
+    ProviderError,
+    ProviderPermanentError,
+    ProviderQuotaError,
+    ProviderTransientError,
+)
 
 
-class AzureError(InfrastructureError):
+class AzureError(ProviderError):
     """Base class for Azure-related errors."""
 
     def __init__(
         self,
         message: str,
         details: Optional[dict[str, Any]] = None,
         error_code: Optional[str] = None,
+        *,
+        provider_name: Optional[str] = None,
+        underlying_exception: Optional[BaseException] = None,
+        is_retryable: Optional[bool] = None,
     ) -> None:
-        super().__init__(message, error_code or self.__class__.__name__, details)
+        super().__init__(
+            message,
+            provider_type="azure",
+            provider_name=provider_name,
+            underlying_exception=underlying_exception,
+            details=details,
+            is_retryable=is_retryable,
+        )
         self.error_code = error_code or self.__class__.__name__
 
     def to_dict(self) -> dict[str, Any]:
@@ -27,13 +45,20 @@ def to_dict(self) -> dict[str, Any]:
             result["error_code"] = self.error_code
         return result
 
+    def safe_to_dict(self) -> dict[str, Any]:
+        """Serialize the exception without exposing its underlying exception."""
+        result = super().safe_to_dict()
+        if self.error_code and self.error_code != self.__class__.__name__:
+            result["error_code"] = self.error_code
+        return result
+
 
 # ---------------------------------------------------------------------------
 # Validation
 # ---------------------------------------------------------------------------
 
 
-class AzureValidationError(AzureError):
+class AzureValidationError(AzureError, ProviderPermanentError):
     """Raised when Azure resource validation fails."""
 
 
@@ -66,7 +91,7 @@ def __init__(self, message: str, image_ref: str, details: Optional[dict[str, Any
 # ---------------------------------------------------------------------------
 
 
-class AzureEntityNotFoundError(AzureError):
+class AzureEntityNotFoundError(AzureError, ProviderPermanentError):
     """Raised when an Azure resource is not found."""
 
 
@@ -91,7 +116,7 @@ def __init__(self, message: str, vmss_name: str, details: Optional[dict[str, Any
 # ---------------------------------------------------------------------------
 
 
-class QuotaExceededError(AzureError):
+class QuotaExceededError(AzureError, ProviderQuotaError):
     """Raised when Azure service quotas would be exceeded."""
 
 
@@ -128,11 +153,11 @@ def __init__(
 # ---------------------------------------------------------------------------
 
 
-class ResourceInUseError(AzureError):
+class ResourceInUseError(AzureError, ProviderTransientError):
     """Raised when an Azure resource is already in use."""
 
 
-class ResourceStateError(AzureError):
+class ResourceStateError(AzureError, ProviderTransientError):
     """Raised when a resource is in an invalid state for the operation."""
 
     def __init__(
@@ -162,11 +187,11 @@ def __init__(
 # ---------------------------------------------------------------------------
 
 
-class AuthenticationError(AzureError):
+class AuthenticationError(AzureError, ProviderAuthError):
     """Raised when Azure authentication fails."""
 
 
-class AuthorizationError(AzureError):
+class AuthorizationError(AzureError, ProviderAuthError):
     """Raised when there are insufficient permissions."""
 
 
@@ -193,11 +218,11 @@ def __init__(
 # ---------------------------------------------------------------------------
 
 
-class RateLimitError(AzureError):
+class RateLimitError(AzureError, ProviderQuotaError):
     """Raised when Azure API rate limits are exceeded."""
 
 
-class NetworkError(AzureError):
+class NetworkError(AzureError, ProviderTransientError):
     """Raised when there are network-related issues."""
 
 
@@ -206,11 +231,11 @@ class NetworkError(AzureError):
 # ---------------------------------------------------------------------------
 
 
-class AzureInfrastructureError(AzureError):
+class AzureInfrastructureError(AzureError, ProviderTransientError):
     """Raised for general Azure infrastructure errors."""
 
 
-class AzureConfigurationError(AzureError):
+class AzureConfigurationError(AzureError, ProviderConfigError):
     """Raised when Azure configuration is invalid."""
 
 
@@ -219,7 +244,7 @@ class AzureConfigurationError(AzureError):
 # ---------------------------------------------------------------------------
 
 
-class LaunchError(AzureError):
+class LaunchError(AzureError, ProviderTransientError):
     """Instance / VMSS launch failed."""
 
     def __init__(
@@ -265,7 +290,7 @@ def __init__(
         self.vmss_name = vmss_name
 
 
-class TerminationError(AzureError):
+class TerminationError(AzureError, ProviderTransientError):
     """Instance termination failed."""
 
     def __init__(
@@ -281,7 +306,7 @@ def __init__(
         self.resource_ids = resource_ids
 
 
-class ResourceCleanupError(AzureError):
+class ResourceCleanupError(AzureError, ProviderTransientError):
     """Failed to clean up Azure resources."""
 
     def __init__(
@@ -303,7 +328,7 @@ def __init__(
         self.resource_type = resource_type
 
 
-class TaggingError(AzureError):
+class TaggingError(AzureError, ProviderTransientError):
     """Failed to tag Azure resources."""
 
     def __init__(
@@ -330,7 +355,7 @@ class CycleCloudError(AzureError):
     """Base class for CycleCloud-related errors."""
 
 
-class CycleCloudConnectionError(CycleCloudError):
+class CycleCloudConnectionError(CycleCloudError, ProviderTransientError):
     """Failed to connect to the CycleCloud REST API."""
 
     def __init__(
@@ -346,7 +371,7 @@ def __init__(
         self.url = url
 
 
-class CycleCloudClusterNotFoundError(CycleCloudError):
+class CycleCloudClusterNotFoundError(CycleCloudError, ProviderPermanentError):
     """The specified CycleCloud cluster was not found."""
 
     def __init__(
@@ -362,7 +387,7 @@ def __init__(
         self.cluster_name = cluster_name
 
 
-class CycleCloudNodeError(CycleCloudError):
+class CycleCloudNodeError(CycleCloudError, ProviderTransientError):
     """Failed to add or manage CycleCloud nodes."""
 
     def __init__(

src/orb/providers/azure/infrastructure/adapters/azure_validation_adapter.py

@@ -2,9 +2,9 @@
 
 from typing import Any
 
-from orb.domain.base.dependency_injection import injectable
 from orb.domain.base.ports.logging_port import LoggingPort
 from orb.domain.base.ports.provider_validation_port import BaseProviderValidationAdapter
+from orb.infrastructure.di.injectable import injectable
 from orb.providers.azure.capabilities import get_supported_api_capabilities, get_supported_apis
 from orb.providers.azure.configuration.config import AzureProviderConfig
 from orb.providers.azure.configuration.validator import validate_azure_template

src/orb/providers/azure/infrastructure/adapters/template_adapter.py

@@ -6,8 +6,8 @@
 
 from typing import Any
 
-from orb.domain.base.dependency_injection import injectable
 from orb.domain.base.ports import LoggingPort
+from orb.infrastructure.di.injectable import injectable
 from orb.providers.azure.domain.template.azure_template_aggregate import AzureTemplate
 
 

src/orb/providers/azure/infrastructure/adapters/template_example_generator_adapter.py

@@ -0,0 +1,22 @@
+"""Azure adapter for provider-owned template example generation."""
+
+from typing import Any, Optional
+
+from orb.providers.azure.infrastructure.azure_handler_factory import (
+    generate_azure_example_templates,
+)
+
+
+class AzureTemplateExampleGeneratorAdapter:
+    """Expose Azure handler examples through the shared generator port."""
+
+    def generate_example_templates(
+        self,
+        provider_name: str,
+        provider_api: Optional[str] = None,
+    ) -> list[Any]:
+        """Return Azure examples, optionally restricted to one provider API."""
+        examples = generate_azure_example_templates()
+        if provider_api is None:
+            return examples
+        return [example for example in examples if example.get("provider_api") == provider_api]

src/orb/providers/azure/infrastructure/azure_client.py

@@ -4,7 +4,6 @@
 - Lazy initialization of Azure service clients
 - Explicit lifecycle cleanup for owned Azure SDK resources
 - Explicit runtime config assembly before client construction
-- Optional metrics instrumentation
 
 Azure SDK clients wrapped:
 - ComputeManagementClient (VMs, VMSS, Disks, Images, Galleries)
@@ -27,9 +26,8 @@
 from typing import TYPE_CHECKING, Any, Callable, Optional
 
 from orb.config import PerformanceConfig
-from orb.domain.base.dependency_injection import injectable
 from orb.domain.base.ports import LoggingPort
-from orb.monitoring.metrics import MetricsCollector
+from orb.infrastructure.di.injectable import injectable
 from orb.providers.azure.configuration.config import AzureProviderConfig
 from orb.providers.azure.exceptions.azure_exceptions import (
     AuthenticationError,
@@ -76,22 +74,18 @@ class AzureClient:
       and passed in as a typed infrastructure object.
     * Azure SDK management clients are created **lazily** on first access
       through ``@property`` accessors, keeping startup cost near zero.
-    * An optional :class:`MetricsCollector` can be injected for API-call
-      instrumentation (hooks are left as extension points for now).
     """
 
     def __init__(
         self,
         runtime_config: AzureClientRuntimeConfig,
         logger: LoggingPort,
-        metrics: Optional[MetricsCollector] = None,
     ) -> None:
         """Initialise the Azure client wrapper.
 
         Args:
             runtime_config: Fully resolved Azure client runtime settings.
             logger: Logger for diagnostic and operational messages.
-            metrics: Optional metrics collector for Azure API instrumentation.
         """
         self._runtime_config = runtime_config
         self._logger = logger
@@ -148,13 +142,6 @@ def __init__(
             network_lookup_error_types=self._network_lookup_error_types,
         )
 
-        # Metrics instrumentation (extension point)
-        self._metrics = metrics
-        if metrics:
-            logger.info("Azure API metrics collection enabled")
-        else:
-            logger.debug("Azure API metrics collection disabled - no MetricsCollector provided")
-
         self._logger.info(
             "Azure client initialised: region=%s, subscription=%s, resource_group=%s, "
             "retries=%d, connect_timeout=%ds, read_timeout=%ds",
@@ -711,16 +698,3 @@ async def resolve_network_identity_from_nic_refs_async(
     ) -> AzureNetworkIdentity:
         """Async variant of NIC-ref network identity resolution."""
         return await self._get_network_identity_resolver().resolve_from_nic_refs_async(nic_refs)
-
-    # ------------------------------------------------------------------
-    # Metrics / observability
-    # ------------------------------------------------------------------
-
-    def get_metrics_stats(self) -> dict[str, Any]:
-        """Return metrics collection statistics.
-
-        Returns:
-            Dictionary with metrics status; currently a placeholder for
-            future instrumentation parity with the AWS client.
-        """
-        return {"metrics_enabled": self._metrics is not None}