Merge pull request #141 from strongdm/schema-guided-entity-parsing

Schema-guided entity parsing

Author: patjakdev

.gitignore

@@ -3,6 +3,7 @@ tmp/
 .DS_Store
 coverage.out
 test/cedar-validation-tool/target/
+test/cedar-entity-parsing-tool/target/
 corpus-tests/
 corpus-tests-json-schemas/
 corpus-tests-validation/

Makefile

@@ -1,4 +1,4 @@
-.PHONY: test linters corpus-update check-upstream-corpus testdata-validation
+.PHONY: test linters corpus-update check-upstream-corpus testdata-validation testdata-entity-parsing
 
 # Run all tests
 test:
@@ -77,5 +77,22 @@ testdata-validation: test/cedar-validation-tool/target/release/cedar-validation-
 	done
 	@echo "Done! Regenerated testdata validation files."
 
+# Build cedar-entity-parsing-tool and generate entity parsing results
+test/cedar-entity-parsing-tool/target/release/cedar-entity-parsing-tool: test/cedar-entity-parsing-tool/src/main.rs test/cedar-entity-parsing-tool/Cargo.toml
+	@echo "Building cedar-entity-parsing-tool..."
+	@cd test/cedar-entity-parsing-tool && cargo build --release
+
+# Regenerate entity parsing data for x/exp/types/testdata
+testdata-entity-parsing: test/cedar-entity-parsing-tool/target/release/cedar-entity-parsing-tool
+	@echo "Regenerating entity parsing test files..."
+	@for testjson in x/exp/types/testdata/*.json; do \
+		case "$$testjson" in *.entities.json|*.parsing.json) continue ;; esac; \
+		basename=$$(basename $$testjson .json); \
+		echo "  Parsing $$basename..."; \
+		test/cedar-entity-parsing-tool/target/release/cedar-entity-parsing-tool \
+			"$$testjson" "x/exp/types/testdata/$${basename}.parsing.json"; \
+	done
+	@echo "Done! Regenerated entity parsing test files."
+
 # Download, convert, and validate
-corpus-update: corpus-tests-json-schemas.tar.gz corpus-tests-validation.tar.gz testdata-validation
+corpus-update: corpus-tests-json-schemas.tar.gz corpus-tests-validation.tar.gz testdata-validation testdata-entity-parsing

corpus_test.go

@@ -18,6 +18,7 @@ import (
 	"github.com/cedar-policy/cedar-go/types"
 	"github.com/cedar-policy/cedar-go/x/exp/batch"
 	"github.com/cedar-policy/cedar-go/x/exp/schema"
+	exptypes "github.com/cedar-policy/cedar-go/x/exp/types"
 )
 
 // jsonEntity is not part of entityValue as I can find
@@ -161,16 +162,6 @@ func TestCorpus(t *testing.T) {
 				t.Fatal("error unmarshalling test", err)
 			}
 
-			entitiesContent, err := fdm.GetFileData(tt.Entities)
-			if err != nil {
-				t.Fatal("error reading entities content", err)
-			}
-
-			var entities cedar.EntityMap
-			if err := json.Unmarshal(entitiesContent, &entities); err != nil {
-				t.Fatal("error unmarshalling test", err)
-			}
-
 			schemaContent, err := fdm.GetFileData(tt.Schema)
 			if err != nil {
 				t.Fatal("error reading schema content", err)
@@ -261,6 +252,18 @@ func TestCorpus(t *testing.T) {
 			rs, err := s.Resolve()
 			testutil.OK(t, err)
 
+			// Load and parse entities using schema-guided parsing
+			entitiesContent, err := fdm.GetFileData(tt.Entities)
+			if err != nil {
+				t.Fatal("error reading entities content", err)
+			}
+
+			var expEntities exptypes.EntityMap
+			if err := expEntities.UnmarshalJSONWithSchema(entitiesContent, rs); err != nil {
+				t.Fatal("error unmarshalling entities", err)
+			}
+			entities := types.EntityMap(expEntities)
+
 			// Build requests for validation checks
 			var requests []cedar.Request
 			for _, r := range tt.Requests {