CT 1990 add access property to parsed nodes (#7007)

Update manifest schema to version 9

.changes/unreleased/Features-20230218-092816.yaml

@@ -0,0 +1,6 @@
+kind: Features
+body: Add access attribute to parsed nodes
+time: 2023-02-18T09:28:16.448175-05:00
+custom:
+  Author: gshank
+  Issue: "6824"

core/dbt/contracts/graph/manifest.py

@@ -1166,7 +1166,7 @@ def __init__(self, macros):
 
 
 @dataclass
-@schema_version("manifest", 8)
+@schema_version("manifest", 9)
 class WritableManifest(ArtifactMixin):
     nodes: Mapping[UniqueID, ManifestNode] = field(
         metadata=dict(description=("The nodes defined in the dbt project and its dependencies"))
@@ -1220,7 +1220,13 @@ class WritableManifest(ArtifactMixin):
 
     @classmethod
     def compatible_previous_versions(self):
-        return [("manifest", 4), ("manifest", 5), ("manifest", 6), ("manifest", 7)]
+        return [
+            ("manifest", 4),
+            ("manifest", 5),
+            ("manifest", 6),
+            ("manifest", 7),
+            ("manifest", 8),
+        ]
 
     def __post_serialize__(self, dct):
         for unique_id, node in dct["nodes"].items():

core/dbt/contracts/graph/nodes.py

@@ -37,16 +37,17 @@
 from dbt.contracts.util import Replaceable, AdditionalPropertiesMixin
 from dbt.events.proto_types import NodeInfo
 from dbt.events.functions import warn_or_error
-from dbt.exceptions import ParsingError
+from dbt.exceptions import ParsingError, InvalidAccessTypeError
 from dbt.events.types import (
     SeedIncreased,
     SeedExceedsLimitSamePath,
     SeedExceedsLimitAndPathChanged,
     SeedExceedsLimitChecksumChanged,
+    ValidationWarning,
 )
 from dbt.events.contextvars import set_contextvars
 from dbt.flags import get_flags
-from dbt.node_types import ModelLanguage, NodeType
+from dbt.node_types import ModelLanguage, NodeType, AccessType
 from dbt.utils import cast_dict_to_dict_of_strings
 
 
@@ -365,6 +366,26 @@ def patch(self, patch: "ParsedNodePatch"):
         self.created_at = time.time()
         self.description = patch.description
         self.columns = patch.columns
+        # This might not be the ideal place to validate the "access" field,
+        # but at this point we have the information we need to properly
+        # validate and we don't before this.
+        if patch.access:
+            if self.resource_type == NodeType.Model:
+                if AccessType.is_valid(patch.access):
+                    self.access = AccessType(patch.access)
+                else:
+                    raise InvalidAccessTypeError(
+                        unique_id=self.unique_id,
+                        field_value=patch.access,
+                    )
+            else:
+                warn_or_error(
+                    ValidationWarning(
+                        field_name="access",
+                        resource_type=self.resource_type.value,
+                        node_name=patch.name,
+                    )
+                )
 
     def same_contents(self, old) -> bool:
         if old is None:
@@ -465,6 +486,7 @@ class HookNode(CompiledNode):
 @dataclass
 class ModelNode(CompiledNode):
     resource_type: NodeType = field(metadata={"restrict": [NodeType.Model]})
+    access: AccessType = AccessType.Protected
 
 
 # TODO: rm?
@@ -1140,6 +1162,7 @@ class ParsedPatch(HasYamlMetadata, Replaceable):
 @dataclass
 class ParsedNodePatch(ParsedPatch):
     columns: Dict[str, ColumnInfo]
+    access: Optional[str]
 
 
 @dataclass

core/dbt/contracts/graph/unparsed.py

@@ -96,6 +96,7 @@ class HasDocs(AdditionalPropertiesMixin, ExtensibleDbtClassMixin, Replaceable):
     constraints: Optional[List[str]] = None
     constraints_check: Optional[str] = None
     docs: Docs = field(default_factory=Docs)
+    access: Optional[str] = None
     _extra: Dict[str, Any] = field(default_factory=dict)
 
 

core/dbt/contracts/util.py

@@ -342,7 +342,7 @@ def read_and_check_versions(cls, path: str):
                         expected=str(cls.dbt_schema_version),
                         found=previous_schema_version,
                     )
-        if get_manifest_schema_version(data) <= 7:
+        if get_manifest_schema_version(data) <= 8:
             data = upgrade_manifest_json(data)
         return cls.from_dict(data)  # type: ignore
 

core/dbt/events/types.proto

@@ -762,7 +762,30 @@ message FinishedRunningStatsMsg {
 
 // I - Project parsing
 
-// Skipping I001, I002, I003, I004, I005, I006, I007, I008, I009
+// Skipping I001, I002, I003, I004, I005, I006, I007
+
+// I008
+message InvalidValueForField {
+    string field_name = 1;
+    string field_value = 2;
+}
+
+message InvalidValueForFieldMsg {
+    EventInfo info = 1;
+    InvalidValueForField data = 2;
+}
+
+// I009
+message ValidationWarning {
+    string resource_type = 1;
+    string field_name = 2;
+    string node_name = 3;
+}
+
+message ValidationWarningMsg {
+    EventInfo info = 1;
+    ValidationWarning data = 2;
+}
 
 // I010
 message ParsePerfInfoPath {

core/dbt/events/types.py

@@ -768,7 +768,25 @@ def message(self) -> str:
 # =======================================================
 
 
-# Skipping I001, I002, I003, I004, I005, I006, I007, I008, I009
+# Skipping I001, I002, I003, I004, I005, I006, I007
+
+
+@dataclass
+class InvalidValueForField(WarnLevel, pt.InvalidValueForField):
+    def code(self):
+        return "I008"
+
+    def message(self) -> str:
+        return f"Invalid value ({self.field_value}) for field {self.field_name}"
+
+
+@dataclass
+class ValidationWarning(WarnLevel, pt.ValidationWarning):
+    def code(self):
+        return "I009"
+
+    def message(self) -> str:
+        return f"Field {self.field_name} is not valid for {self.resource_type} ({self.node_name})"
 
 
 @dataclass

core/dbt/exceptions.py

@@ -1114,6 +1114,16 @@ def __init__(self, exc: ValidationError, node):
         super().__init__(msg=self.msg)
 
 
+class InvalidAccessTypeError(ParsingError):
+    def __init__(self, unique_id: str, field_value: str):
+        self.unique_id = unique_id
+        self.field_value = field_value
+        msg = (
+            f"Node {self.unique_id} has an invalid value ({self.field_value}) for the access field"
+        )
+        super().__init__(msg=msg)
+
+
 class SameKeyNestedError(CompilationError):
     def __init__(self):
         msg = "Test cannot have the same key at the top-level and in config"

core/dbt/node_types.py

@@ -3,6 +3,20 @@
 from dbt.dataclass_schema import StrEnum
 
 
+class AccessType(StrEnum):
+    Protected = "protected"
+    Private = "private"
+    Public = "public"
+
+    @classmethod
+    def is_valid(cls, item):
+        try:
+            cls(item)
+        except ValueError:
+            return False
+        return True
+
+
 class NodeType(StrEnum):
     Model = "model"
     Analysis = "analysis"

core/dbt/parser/schemas.py

@@ -878,6 +878,7 @@ def parse_patch(self, block: TargetBlock[NodeTarget], refs: ParserRef) -> None:
             meta=block.target.meta,
             docs=block.target.docs,
             config=block.target.config,
+            access=block.target.access,
         )
         assert isinstance(self.yaml.file, SchemaSourceFile)
         source_file: SchemaSourceFile = self.yaml.file