data_model_gen: treat clusters with codegen implementation as non-SCI/non-migrated

tools/data_model_gen/artifacts/migrated_clusters.json

@@ -0,0 +1,41 @@
+{
+    "migrated_cluster": [
+        "access_control",
+        "administrator_commissioning",
+        "basic_information",
+        "binding",
+        "boolean_state_configuration",
+        "chime",
+        "descriptor",
+        "device_energy_management",
+        "diagnostic_logs",
+        "electrical_energy_measurement",
+        "electrical_power_measurement",
+        "ethernet_network_diagnostics",
+        "fixed_label",
+        "general_commissioning",
+        "general_diagnostics",
+        "group_key_management",
+        "groupcast",
+        "icd_management",
+        "identify",
+        "localization_configuration",
+        "network_commissioning",
+        "occupancy_sensor",
+        "operational_credentials",
+        "ota_software_update_provider",
+        "power_topology",
+        "push_av_stream_transport",
+        "resource_monitoring",
+        "scenes",
+        "software_diagnostics",
+        "soil_measurement",
+        "time_format_localization",
+        "time_synchronization",
+        "user_label",
+        "webrtc_transport_provider",
+        "webrtc_transport_requestor",
+        "wifi_network_diagnostics"
+    ],
+    "migrated_cluster_with_codegen_impl": []
+}

tools/data_model_gen/chip_source_deps/cluster_mapping.py

@@ -92,9 +92,15 @@ def find_delegate_server_files(root_dir):
     """
     delegate_server_files = set()
 
-    for dirpath, _, filenames in os.walk(root_dir):
+    for dirpath, dirnames, filenames in os.walk(root_dir):
+        if "codegen" in dirnames:
+            scan_dir = os.path.join(dirpath, "codegen")
+            filenames = os.listdir(scan_dir)
+            dirnames[:] = []
+        else:
+            scan_dir = dirpath
         for filename in filenames:
-            full_path = os.path.join(dirpath, filename)
+            full_path = os.path.join(scan_dir, filename)
 
             # Check for delegate callback in server files
             if "-delegate.h" in filename.lower():

tools/data_model_gen/chip_source_deps/parser.py

@@ -13,7 +13,8 @@
 # limitations under the License.
 import os
 import logging
-from utils.config import FileNames
+import subprocess
+from utils.config import ARTIFACTS_DIR, FileNames
 from utils.helper import write_to_file
 from utils.exceptions import ConfigurationError, CodeGenerationError
 from chip_source_deps.cluster_mapping import normalize_cluster_name
@@ -29,49 +30,70 @@ from chip_source_deps.internally_managed_attributes import (
 logger = logging.getLogger(__name__)
 
 
-def generate_migrated_clusters(
-    root_dir, migrated_clusters_json_file_path
-) -> tuple[bool, str]:
-    """Find all clusters that have a CodegenIntegration.cpp file
+def _find_migrated_cluster_dirs(root_dir: str) -> set[str]:
+    """Return the set of top-level cluster directories under ``root_dir`` that
+    contain a header declaring a class inheriting ``DefaultServerCluster``.
+    """
+    result = subprocess.run(
+        ["grep", "-rl", "--include=*.h", "public DefaultServerCluster", root_dir],
+        check=False,
+        capture_output=True,
+        text=True,
+    )
+    if result.returncode >= 2:
+        raise RuntimeError(f"grep failed ({result.returncode}): {result.stderr}")
+
+    abs_root = os.path.abspath(root_dir)
+    cluster_dirs: set[str] = set()
+    for header_path in result.stdout.splitlines():
+        rel = os.path.relpath(header_path, abs_root)
+        top = rel.split(os.sep, 1)[0]
+        if top and top != "..":
+            cluster_dirs.add(os.path.join(abs_root, top))
+    return cluster_dirs
+
+
+def generate_migrated_clusters(root_dir) -> tuple[bool, str]:
+    """Find all migrated clusters and write them to ``Artifacts/migrated_clusters.json``.
+
+    The JSON holds two lists:
+      - ``migrated_cluster``: clusters using ``DefaultServerCluster`` at the
+        cluster top level; these feed data_model_gen.
+      - ``migrated_cluster_with_codegen_impl``: clusters that own a ``codegen/``
+        subdir; they have legacy implementation as it is and are skipped by data_model_gen.
 
     Args:
-        root_dir: The root directory to search in
-        migrated_clusters_json_file_path: Path to save the migrated clusters list
+        root_dir: The root directory to search in.
     Returns:
         True if successful, False otherwise
     """
-    migrated_clusters = []
+    migrated_clusters = set()
+    migrated_clusters_with_codegen_impl = set()
 
     try:
-        for dirpath, _, filenames in os.walk(root_dir):
-            for filename in filenames:
-                if not filename.endswith(".cpp") and not filename.endswith(".h"):
-                    continue
-                if (
-                    filename.lower() == "codegenintegration.cpp"
-                    or filename.lower() == "codegeninstance.cpp"
-                ):
-                    cluster_name = os.path.basename(dirpath)
-                    migrated_clusters.append(normalize_cluster_name(cluster_name))
-                else:
-                    with open(
-                        os.path.join(dirpath, filename), "r", encoding="utf-8"
-                    ) as file:
-                        if "DefaultServerCluster" in file.read():
-                            cluster_name = os.path.basename(dirpath)
-                            migrated_clusters.append(
-                                normalize_cluster_name(cluster_name)
-                            )
+        for cluster_dir in _find_migrated_cluster_dirs(root_dir):
+            cluster_name = normalize_cluster_name(os.path.basename(cluster_dir))
+            if not cluster_name:
+                continue
+            if os.path.isdir(os.path.join(cluster_dir, "codegen")):
+                migrated_clusters_with_codegen_impl.add(cluster_name)
+            else:
+                migrated_clusters.add(cluster_name)
 
-        migrated_clusters.sort()
+        payload = {
+            "migrated_cluster": sorted(migrated_clusters),
+            "migrated_cluster_with_codegen_impl": sorted(
+                migrated_clusters_with_codegen_impl
+            ),
+        }
 
-        if write_to_file(migrated_clusters_json_file_path, migrated_clusters, "json"):
-            logger.info(
-                f"Successfully written Migrated Clusters to {migrated_clusters_json_file_path}"
-            )
+        os.makedirs(ARTIFACTS_DIR, exist_ok=True)
+        artifact_path = os.path.join(ARTIFACTS_DIR, FileNames.MIGRATED_CLUSTERS.value)
+
+        if write_to_file(artifact_path, payload, "json"):
+            logger.info(f"Successfully written Migrated Clusters to {artifact_path}")
             return True, None
-        else:
-            return False, f"Error writing to {migrated_clusters_json_file_path}"
+        return False, f"Error writing to {artifact_path}"
     except Exception as e:
         return False, f"Error generating migrated clusters: {str(e)}"
 
@@ -133,9 +155,6 @@ def generate_requirements(esp_matter_path, output_dir):
         FileNames.ZAP_FILTER_LIST: os.path.join(
             output_dir, FileNames.ZAP_FILTER_LIST.value
         ),
-        FileNames.MIGRATED_CLUSTERS: os.path.join(
-            output_dir, FileNames.MIGRATED_CLUSTERS.value
-        ),
     }
 
     logger.debug(
@@ -199,13 +218,11 @@ def generate_requirements(esp_matter_path, output_dir):
         )
 
     logger.debug("Finding migrated clusters...")
-    is_generated, error_message = generate_migrated_clusters(
-        root_cluster_server_dir, file_paths[FileNames.MIGRATED_CLUSTERS]
-    )
+    is_generated, error_message = generate_migrated_clusters(root_cluster_server_dir)
     if not is_generated:
         raise CodeGenerationError(
             error_message,
-            file_path=file_paths[FileNames.MIGRATED_CLUSTERS],
+            file_path=os.path.join(ARTIFACTS_DIR, FileNames.MIGRATED_CLUSTERS.value),
             context="generate_requirements",
             suggestion=f"Check {root_cluster_server_dir} and write permissions.",
         )

tools/data_model_gen/utils/config.py

@@ -36,6 +36,10 @@ COLORED_FORMATTER = ColoredFormatter(
 DEFAULT_OUTPUT_DIR = "out"
 DEFAULT_CHIP_VERSION = "1.5"
 
+ARTIFACTS_DIR = os.path.join(
+    os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "artifacts"
+)
+
 SPECIFICATION_VERSIONS = ["1.1", "1.2", "1.3", "1.4", "1.4.2", "1.5", "1.6"]
 
 ALLOW_PROVISIONAL = False

tools/data_model_gen/xml_processing/parse_context.py

@@ -19,22 +19,24 @@ Loads JSON artifacts once per run to avoid repeated file I/O.
 import json
 import os
 
-from utils.config import FileNames
+from utils.config import ARTIFACTS_DIR, FileNames
 from utils.conversion_utils import hex_to_int
 
 
 def load_cluster_parse_context(output_dir: str):
     """Load all cluster-related metadata from output_dir into a single context."""
 
-    def _load_json(filename: str):
-        path = os.path.join(output_dir, filename)
+    def _load_json(filename: str, base_dir: str = output_dir):
+        path = os.path.join(base_dir, filename)
         with open(path, "r") as f:
             return json.load(f)
 
     return ClusterParseContext(
         delegate_clusters=_load_json(FileNames.DELEGATE_CLUSTERS.value),
         plugin_init_cb_clusters=_load_json(FileNames.PLUGIN_INIT_CB_CLUSTERS.value),
-        migrated_clusters=_load_json(FileNames.MIGRATED_CLUSTERS.value),
+        migrated_clusters=_load_json(
+            FileNames.MIGRATED_CLUSTERS.value, base_dir=ARTIFACTS_DIR
+        )["migrated_cluster"],
         zap_filter_list=_load_json(FileNames.ZAP_FILTER_LIST.value),
         internally_managed_attributes=_load_json(
             FileNames.INTERNALLY_MANAGED_ATTRIBUTES.value