Adapt parser to monorepo

script/dependency-parser/src/enhanced_ninja_parser.py

@@ -163,11 +163,20 @@ def _build_file_to_executable_mapping(self):
         """Build the final mapping from files to executables."""
         print("Building file-to-executable mapping...")
 
+        # For monorepo, truncate the path before and including projects/<project_name>
+        self.project = None
+        rl_regex = rf"rocm-libraries[\\/]+projects[\\/]+([^\\/]+)[\\/]+(.*)"
         for exe, object_files in self.executable_to_objects.items():
             for obj_file in object_files:
                 # Add all dependencies of this object file
                 if obj_file in self.object_to_all_deps:
                     for dep_file in self.object_to_all_deps[obj_file]:
+                        match = re.search(rl_regex, dep_file, re.IGNORECASE)
+                        if match:
+                            dep_file = match.group(2)
+                            if not self.project:
+                                print(f"Found rocm-libraries project: '{match.group(1)}'")
+                                self.project = match.group(1)
                         # Filter out system files and focus on project files
                         if self._is_project_file(dep_file):
                             self.file_to_executables[dep_file].add(exe)
@@ -244,6 +253,10 @@ def export_to_json(self, output_file):
                 exe_to_files[exe].add(file_path)
 
         mapping_data = {
+            "repo": {
+                "type": "monorepo" if self.project else "component",
+                "project": self.project
+            },
             "file_to_executables": {
                 file_path: list(exes)
                 for file_path, exes in self.file_to_executables.items()

script/dependency-parser/src/selective_test_filter.py

@@ -31,7 +31,7 @@
 import os
 
 
-def get_changed_files(ref1, ref2):
+def get_changed_files(ref1, ref2, project: str = None):
     """Return a set of files changed between two git refs."""
     try:
         result = subprocess.run(
@@ -40,7 +40,21 @@ def get_changed_files(ref1, ref2):
             text=True,
             check=True,
         )
-        files = set(line.strip() for line in result.stdout.splitlines() if line.strip())
+
+        raw_files = set(line.strip() for line in result.stdout.splitlines() if line.strip())
+
+        if project is None:
+            files = raw_files
+            print(f"Identified {len(files)} modified files")
+        else:
+            root = f"projects/{project}/"
+            root_len = len(root)
+            files = set()
+            for f in raw_files:
+                if f.startswith(root):
+                    files.add(f[root_len:])
+            print(f"Identified {len(files)} files modified in project '{project}'")
+
         return files
     except subprocess.CalledProcessError as e:
         print(f"Error running git diff: {e}")
@@ -52,9 +66,12 @@ def load_depmap(depmap_json):
     with open(depmap_json, "r") as f:
         data = json.load(f)
     # Support both old and new formats
+    json_project = None
+    if "repo" in data and data["repo"]["type"] == "monorepo":
+        json_project = data["repo"]["project"]
     if "file_to_executables" in data:
-        return data["file_to_executables"]
-    return data
+        return data["file_to_executables"], json_project
+    return data, json_project
 
 
 def select_tests(file_to_executables, changed_files, filter_mode):
@@ -132,12 +149,12 @@ def main():
         print(f"Dependency map JSON not found: {depmap_json}")
         sys.exit(1)
 
-    changed_files = get_changed_files(ref1, ref2)
+    file_to_executables, json_project = load_depmap(depmap_json)
+    changed_files = get_changed_files(ref1, ref2, json_project)
     if not changed_files:
         print("No changed files detected.")
         tests = []
     else:
-        file_to_executables = load_depmap(depmap_json)
         tests = select_tests(file_to_executables, changed_files, filter_mode)
 
     with open(output_json, "w") as f: