backend/app/modules/scanner/application/services.py

@@ -0,0 +1,127 @@
+"""ScanService — orchestrates parsers, file status detection, and entity collection."""
+
+from __future__ import annotations
+
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+from app.modules.design.domain.services import DesignValidationService
+from app.modules.design.domain.value_objects import FileStatus
+from app.modules.project.domain.entities import Project
+from app.modules.scanner.domain.entities import (
+    FileStatusEntry,
+    ScanResult,
+    ScanSummary,
+)
+from app.modules.scanner.infrastructure.parsers.csv_parser import CsvParser
+from app.modules.scanner.infrastructure.parsers.md_parser import MdParser
+from app.modules.scanner.infrastructure.parsers.openapi_parser import OpenapiParser
+
+
+class ScanService:
+    """Scan a project's design directory and produce a ScanResult."""
+
+    def __init__(self) -> None:
+        self._csv_parser = CsvParser()
+        self._md_parser = MdParser()
+        self._openapi_parser = OpenapiParser()
+        self._cache: dict[str, ScanResult] = {}
+
+    def scan(self, project: Project) -> ScanResult:
+        design_dir = Path(project.design_dir)
+        file_statuses: list[FileStatusEntry] = []
+        all_entities: dict[str, list[Any]] = {}
+
+        # Walk design directory recursively
+        for file_path in sorted(design_dir.rglob("*")):
+            if not file_path.is_file():
+                continue
+
+            # Determine file status
+            try:
+                content = file_path.read_text(encoding="utf-8")
+            except Exception:
+                content = ""
+
+            status = DesignValidationService.determine_file_status(
+                content, str(file_path)
+            )
+            lines = len(content.splitlines()) if content else 0
+            rel_path = str(file_path.relative_to(design_dir))
+
+            file_statuses.append(FileStatusEntry(
+                path=rel_path,
+                status=status,
+                content_lines=lines,
+            ))
+
+            # Dispatch to appropriate parser
+            parsed: dict[str, list[Any]] = {}
+            suffix = file_path.suffix.lower()
+            fname = file_path.name.lower()
+
+            if suffix == ".csv":
+                parsed = self._csv_parser.parse(file_path)
+            elif suffix == ".md":
+                parsed = self._md_parser.parse(file_path)
+            elif suffix == ".yaml" or suffix == ".yml":
+                if "openapi" in fname or "api-contracts" in fname:
+                    parsed = self._openapi_parser.parse(file_path)
+
+            # Merge parsed entities
+            for key, entities in parsed.items():
+                if key not in all_entities:
+                    all_entities[key] = []
+                all_entities[key].extend(entities)
+
+        # Build summary
+        summary = self._build_summary(file_statuses)
+
+        # Assemble ScanResult
+        # Singleton fields (take first item from list or None)
+        singleton_keys = {
+            "scope_and_goals", "system_context", "solution_layer",
+            "module_boundary_rule", "runtime_topology",
+            "operational_baseline", "release_plan",
+        }
+
+        kwargs: dict[str, Any] = {
+            "project_id": project.id,
+            "scanned_at": datetime.now(timezone.utc),
+            "file_statuses": file_statuses,
+            "summary": summary,
+        }
+
+        for key, entities in all_entities.items():
+            if key in singleton_keys:
+                kwargs[key] = entities[0] if entities else None
+            else:
+                kwargs[key] = entities
+
+        result = ScanResult(**kwargs)
+        self._cache[project.id] = result
+        return result
+
+    def get_latest_scan(self, project_id: str) -> ScanResult | None:
+        return self._cache.get(project_id)
+
+    @staticmethod
+    def _build_summary(file_statuses: list[FileStatusEntry]) -> ScanSummary:
+        ok = sum(1 for fs in file_statuses if fs.status == FileStatus.OK)
+        sparse = sum(1 for fs in file_statuses if fs.status == FileStatus.SPARSE)
+        missing = sum(1 for fs in file_statuses if fs.status == FileStatus.MISSING)
+        placeholder_heavy = sum(
+            1 for fs in file_statuses if fs.status == FileStatus.PLACEHOLDER_HEAVY
+        )
+        template_residue = sum(
+            1 for fs in file_statuses if fs.status == FileStatus.TEMPLATE_RESIDUE
+        )
+        return ScanSummary(
+            total_files=len(file_statuses),
+            ok=ok,
+            sparse=sparse,
+            missing=missing,
+            placeholder_heavy=placeholder_heavy,
+            template_residue=template_residue,
+        )

backend/tests/test_scanner_service.py

@@ -0,0 +1,110 @@
+"""Tests for ScanService — integration with real design directory."""
+
+import pytest
+from datetime import datetime
+from pathlib import Path
+
+from app.modules.project.domain.entities import Project
+from app.modules.scanner.application.services import ScanService
+
+
+@pytest.fixture
+def scan_service():
+    return ScanService()
+
+
+@pytest.fixture
+def test_project():
+    return Project(
+        id="test-proj",
+        name="test",
+        design_dir="/workspace/arch-design-agent-skill-dashboard/design",
+        code_dir=None,
+        created_at=datetime(2026, 1, 1),
+    )
+
+
+def test_scan_produces_result(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert result.project_id == "test-proj"
+    assert result.scanned_at is not None
+    assert len(result.file_statuses) > 0
+    assert result.summary.total_files > 0
+
+
+def test_scan_has_capabilities(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.capabilities) > 0
+    assert result.capabilities[0].capability_id.startswith("CAP-")
+
+
+def test_scan_has_modules(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.modules) > 0
+
+
+def test_scan_has_traceability_links(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.traceability_links) > 0
+    # entity_ids should be a list (space-split)
+    assert isinstance(result.traceability_links[0].entity_ids, list)
+
+
+def test_scan_has_design_documents(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.design_documents) > 0
+
+
+def test_scan_has_api_contracts(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.api_contracts) > 0
+    assert result.api_contracts[0].path.startswith("/")
+
+
+def test_scan_has_value_flows(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.value_flows) > 0
+
+
+def test_scan_has_integrations(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.integrations) > 0
+
+
+def test_scan_has_external_systems(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.external_systems) > 0
+
+
+def test_scan_has_entities(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    assert len(result.entities) > 0
+
+
+def test_scan_summary_counts_match(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    s = result.summary
+    assert s.total_files == len(result.file_statuses)
+    assert s.total_files == s.ok + s.sparse + s.missing + s.placeholder_heavy + s.template_residue
+
+
+def test_get_latest_scan_none_before_scan(scan_service):
+    assert scan_service.get_latest_scan("nonexistent") is None
+
+
+def test_get_latest_scan_cached(scan_service, test_project):
+    scan_service.scan(test_project)
+    cached = scan_service.get_latest_scan("test-proj")
+    assert cached is not None
+    assert cached.project_id == "test-proj"
+
+
+def test_scan_has_singleton_fields(scan_service, test_project):
+    result = scan_service.scan(test_project)
+    # These MD files have frontmatter and should produce singleton entities
+    assert result.system_context is not None
+    assert result.solution_layer is not None
+    assert result.module_boundary_rule is not None
+    assert result.runtime_topology is not None
+    assert result.operational_baseline is not None
+    assert result.release_plan is not None