CT 1443 remove root path

.changes/unreleased/Under the Hood-20221028-104837.yaml

@@ -0,0 +1,7 @@
+kind: Under the Hood
+body: Remove the 'root_path' field from most nodes
+time: 2022-10-28T10:48:37.687886-04:00
+custom:
+  Author: gshank
+  Issue: "6171"
+  PR: "6172"

core/dbt/context/providers.py

@@ -803,6 +803,7 @@ def load_agate_table(self) -> agate.Table:
             raise_compiler_error(
                 "can only load_agate_table for seeds (got a {})".format(self.model.resource_type)
             )
+        assert self.model.root_path
         path = os.path.join(self.model.root_path, self.model.original_file_path)
         column_types = self.model.config.column_types
         try:

core/dbt/contracts/graph/compiled.py

@@ -97,6 +97,7 @@ class CompiledSeedNode(CompiledNode):
     # keep this in sync with ParsedSeedNode!
     resource_type: NodeType = field(metadata={"restrict": [NodeType.Seed]})
     config: SeedConfig = field(default_factory=SeedConfig)
+    root_path: Optional[str] = None
 
     @property
     def empty(self):

core/dbt/contracts/graph/manifest.py

@@ -1183,7 +1183,7 @@ def __init__(self, macros):
 
 
 @dataclass
-@schema_version("manifest", 7)
+@schema_version("manifest", 8)
 class WritableManifest(ArtifactMixin):
     nodes: Mapping[UniqueID, ManifestNode] = field(
         metadata=dict(description=("The nodes defined in the dbt project and its dependencies"))
@@ -1229,7 +1229,7 @@ class WritableManifest(ArtifactMixin):
 
     @classmethod
     def compatible_previous_versions(self):
-        return [("manifest", 4), ("manifest", 5), ("manifest", 6)]
+        return [("manifest", 4), ("manifest", 5), ("manifest", 6), ("manifest", 7)]
 
     def __post_serialize__(self, dct):
         for unique_id, node in dct["nodes"].items():

core/dbt/contracts/graph/parsed.py

@@ -412,6 +412,9 @@ class ParsedSeedNode(ParsedNode):
     # keep this in sync with CompiledSeedNode!
     resource_type: NodeType = field(metadata={"restrict": [NodeType.Seed]})
     config: SeedConfig = field(default_factory=SeedConfig)
+    # seeds need the root_path because the contents are not loaded initially
+    # and we need the root_path to load the seed later
+    root_path: Optional[str] = None
 
     @property
     def empty(self):

core/dbt/contracts/graph/unparsed.py

@@ -24,7 +24,6 @@
 @dataclass
 class UnparsedBaseNode(dbtClassMixin, Replaceable):
     package_name: str
-    root_path: str
     path: str
     original_file_path: str
 
@@ -364,7 +363,6 @@ def get_table_named(self, name: str) -> Optional[SourceTablePatch]:
 @dataclass
 class UnparsedDocumentation(dbtClassMixin, Replaceable):
     package_name: str
-    root_path: str
     path: str
     original_file_path: str
 

core/dbt/contracts/util.py

@@ -240,13 +240,32 @@ def rename_sql_attr(node_content: dict) -> dict:
 def upgrade_manifest_json(manifest: dict) -> dict:
     for node_content in manifest.get("nodes", {}).values():
         node_content = rename_sql_attr(node_content)
+        if node_content["resource_type"] != "seed" and "root_path" in node_content:
+            del node_content["root_path"]
     for disabled in manifest.get("disabled", {}).values():
         # There can be multiple disabled nodes for the same unique_id
         # so make sure all the nodes get the attr renamed
-        disabled = [rename_sql_attr(n) for n in disabled]
+        for node_content in disabled:
+            rename_sql_attr(node_content)
+            if node_content["resource_type"] != "seed" and "root_path" in node_content:
+                del node_content["root_path"]
     for metric_content in manifest.get("metrics", {}).values():
         # handle attr renames + value translation ("expression" -> "derived")
         metric_content = rename_metric_attr(metric_content)
+        if "root_path" in metric_content:
+            del metric_content["root_path"]
+    for exposure_content in manifest.get("exposures", {}).values():
+        if "root_path" in exposure_content:
+            del exposure_content["root_path"]
+    for source_content in manifest.get("sources", {}).values():
+        if "root_path" in exposure_content:
+            del source_content["root_path"]
+    for macro_content in manifest.get("macros", {}).values():
+        if "root_path" in macro_content:
+            del macro_content["root_path"]
+    for doc_content in manifest.get("docs", {}).values():
+        if "root_path" in doc_content:
+            del doc_content["root_path"]
     return manifest
 
 
@@ -291,7 +310,7 @@ def read_and_check_versions(cls, path: str):
                         expected=str(cls.dbt_schema_version),
                         found=previous_schema_version,
                     )
-        if get_manifest_schema_version(data) <= 6:
+        if get_manifest_schema_version(data) <= 7:
             data = upgrade_manifest_json(data)
         return cls.from_dict(data)  # type: ignore
 

core/dbt/graph/selector_methods.py

@@ -286,8 +286,6 @@ def search(self, included_nodes: Set[UniqueId], selector: str) -> Iterator[Uniqu
         root = Path.cwd()
         paths = set(p.relative_to(root) for p in root.glob(selector))
         for node, real_node in self.all_nodes(included_nodes):
-            if Path(real_node.root_path) != root:
-                continue
             ofp = Path(real_node.original_file_path)
             if ofp in paths:
                 yield node

core/dbt/parser/base.py

@@ -169,7 +169,6 @@ def _create_error_node(
             resource_type=self.resource_type,
             path=path,
             original_file_path=original_file_path,
-            root_path=self.project.project_root,
             package_name=self.project.project_name,
             raw_code=raw_code,
             language=language,
@@ -202,7 +201,6 @@ def _create_parsetime_node(
             "database": self.default_database,
             "fqn": fqn,
             "name": name,
-            "root_path": self.project.project_root,
             "resource_type": self.resource_type,
             "path": path,
             "original_file_path": block.path.original_file_path,

core/dbt/parser/docs.py

@@ -32,7 +32,6 @@ def parse_block(self, block: BlockContents) -> Iterable[ParsedDocumentation]:
         contents = get_rendered(block.contents, {}).strip()
 
         doc = ParsedDocumentation(
-            root_path=self.project.project_root,
             path=block.file.path.relative_path,
             original_file_path=block.path.original_file_path,
             package_name=self.project.project_name,

core/dbt/parser/generic_test.py

@@ -35,7 +35,6 @@ def parse_generic_test(
             macro_sql=block.full_block,
             original_file_path=base_node.original_file_path,
             package_name=base_node.package_name,
-            root_path=base_node.root_path,
             resource_type=base_node.resource_type,
             name=name,
             unique_id=unique_id,
@@ -96,7 +95,6 @@ def parse_file(self, block: FileBlock):
             original_file_path=original_file_path,
             package_name=self.project.project_name,
             raw_code=source_file.contents,
-            root_path=self.project.project_root,
             resource_type=NodeType.Macro,
             language="sql",
         )

core/dbt/parser/macros.py

@@ -41,7 +41,6 @@ def parse_macro(
             macro_sql=block.full_block,
             original_file_path=base_node.original_file_path,
             package_name=base_node.package_name,
-            root_path=base_node.root_path,
             resource_type=base_node.resource_type,
             name=name,
             unique_id=unique_id,
@@ -103,7 +102,6 @@ def parse_file(self, block: FileBlock):
             original_file_path=original_file_path,
             package_name=self.project.project_name,
             raw_code=source_file.contents,
-            root_path=self.project.project_root,
             resource_type=NodeType.Macro,
             language="sql",
         )

core/dbt/parser/schemas.py

@@ -244,7 +244,6 @@ def get_hashable_md(data: Union[str, int, float, List, Dict]) -> Union[str, List
             "database": self.default_database,
             "fqn": fqn,
             "name": name,
-            "root_path": self.project.project_root,
             "resource_type": self.resource_type,
             "tags": tags,
             "path": path,
@@ -728,7 +727,6 @@ def add_source_definitions(self, source: UnparsedSourceDefinition) -> None:
                 table=table,
                 path=original_file_path,
                 original_file_path=original_file_path,
-                root_path=self.project.project_root,
                 package_name=package_name,
                 unique_id=unique_id,
                 resource_type=NodeType.Source,
@@ -1031,7 +1029,6 @@ def parse_exposure(self, unparsed: UnparsedExposure):
 
         parsed = ParsedExposure(
             package_name=package_name,
-            root_path=self.project.project_root,
             path=path,
             original_file_path=self.yaml.path.original_file_path,
             unique_id=unique_id,
@@ -1135,7 +1132,6 @@ def parse_metric(self, unparsed: UnparsedMetric):
 
         parsed = ParsedMetric(
             package_name=package_name,
-            root_path=self.project.project_root,
             path=path,
             original_file_path=self.yaml.path.original_file_path,
             unique_id=unique_id,

core/dbt/parser/seeds.py

@@ -7,6 +7,8 @@
 
 class SeedParser(SimpleSQLParser[ParsedSeedNode]):
     def parse_from_dict(self, dct, validate=True) -> ParsedSeedNode:
+        # seeds need the root_path because the contents are not loaded
+        dct["root_path"] = self.project.project_root
         if validate:
             ParsedSeedNode.validate(dct)
         return ParsedSeedNode.from_dict(dct)

core/dbt/parser/sources.py

@@ -161,7 +161,6 @@ def parse_source(self, target: UnpatchedSourceDefinition) -> ParsedSourceDefinit
             database=(source.database or default_database),
             schema=(source.schema or source.name),
             identifier=(table.identifier or table.name),
-            root_path=target.root_path,
             path=target.path,
             original_file_path=target.original_file_path,
             columns=refs.column_info,

core/dbt/parser/sql.py

@@ -56,7 +56,6 @@ def parse_remote(self, contents) -> Iterable[ParsedMacro]:
             package_name=self.project.project_name,
             raw_code=contents,
             language="sql",
-            root_path=self.project.project_root,
             resource_type=NodeType.Macro,
         )
         for node in self.parse_unparsed_macros(base):