Initial implementation of metadata merger

Author: furtib

src/BUILD

@@ -5,6 +5,12 @@ py_binary(
     visibility = ["//visibility:public"],
 )
 
+py_binary(
+    name = "metadata_merge",
+    srcs = ["metadata_merge.py"],
+    visibility = ["//visibility:public"],
+)
+
 # Build & Test script template
 exports_files(
     ["codechecker_script.py"],

src/code_checker.bzl

@@ -322,6 +322,18 @@ def _code_checker_impl(ctx):
                         sources_and_headers,
                     )
                     all_files += outputs
+    # merge metadata
+    metadata = [file for file in all_files if file.path.endswith("metadata.json")]
+    metadata_json = ctx.actions.declare_file(ctx.attr.name + "/data/metadata.json")
+    ctx.actions.run(
+        inputs = metadata,
+        outputs = [metadata_json],
+        executable = ctx.executable._metadata_merge_script,
+        arguments = [metadata_json.path] + [file.path for file in metadata],
+        mnemonic = "Metadata",
+        progress_message = "Merging metadata.json"
+    )
+    all_files.append(metadata_json)
     ctx.actions.write(
         output = ctx.outputs.test_script,
         is_executable = True,
@@ -360,6 +372,14 @@ code_checker_test = rule(
             ],
             doc = "List of default CodeChecker analyze options",
         ),
+        "_metadata_merge_script": attr.label(
+            default = ":metadata_merge",
+            executable = True,
+            cfg = 'exec',
+        ),
+        "_python_runtime": attr.label(
+            default = "@default_python_tools//:py3_runtime",
+        ),
         "targets": attr.label_list(
             aspects = [
                 compile_info_aspect,

src/metadata_merge.py

@@ -0,0 +1,86 @@
+import json
+import os
+import sys
+from typing import List, Dict, Any
+
+
+def merge_two_json(json1, json2):
+    if json1 == {}:
+        return json2
+    if json2 == {}:
+        return json1
+    json1_root = json1["tools"][0]
+    json2_root = json2["tools"][0]
+    # We append info from json2 to json1 from here on out
+    json1_root["result_source_files"].update(json2_root["result_source_files"])
+    json1_root["skipped"] = json1_root["skipped"] + json2_root["skipped"]
+    # Merge time
+    json1_root["timestamps"]["begin"] = min(
+        float(json1_root["timestamps"]["begin"]),
+        float(json2_root["timestamps"]["begin"]),
+    )
+    json1_root["timestamps"]["end"] = max(
+        float(json1_root["timestamps"]["end"]),
+        float(json2_root["timestamps"]["end"]),
+    )
+    # Merge analyzers
+    for key, value in json2_root["analyzers"].items():
+        json1_stat = json1_root["analyzers"][key]["analyzer_statistics"]
+        json2_stat = json2_root["analyzers"][key]["analyzer_statistics"]
+        json1_stat["failed"] = json1_stat["failed"] + json2_stat["failed"]
+        json1_stat["failed_sources"].extend(json2_stat["failed_sources"])
+        json1_stat["successful"] = (
+            json1_stat["successful"] + json2_stat["successful"]
+        )
+        json1_stat["successful_sources"].extend(
+            json2_stat["successful_sources"]
+        )
+    return json1
+
+
+def merge_json_files(file_paths: List[str]) -> Dict[str, Any]:
+    merged_data = {}
+    for file_path in file_paths:
+        if not os.path.exists(file_path):
+            print(
+                f"Error: File not found at '{file_path}'. Skipping.",
+                file=sys.stderr,
+            )
+            continue
+
+        try:
+            with open(file_path, "r", encoding="utf-8") as f:
+                data = json.load(f)
+                merged_data = merge_two_json(merged_data, data)
+        except json.JSONDecodeError:
+            print(
+                f"Error: Could not decode JSON from '{file_path}'. Skipping.",
+                file=sys.stderr,
+            )
+        except Exception as e:
+            print(
+                f"An unexpected error occurred while processing '{file_path}': {e}",
+                file=sys.stderr,
+            )
+
+    return merged_data
+
+
+def save_json_file(data: Dict[str, Any], output_path: str) -> None:
+    try:
+        with open(output_path, "w", encoding="utf-8") as f:
+            json.dump(data, f, indent=4)
+        print(f"\nSuccessfully saved merged data to '{output_path}'")
+    except Exception as e:
+        print(f"An error occurred while saving the file: {e}", file=sys.stderr)
+
+
+if __name__ == "__main__":
+    output_file = sys.argv[1]
+    input_files = sys.argv[2:]
+
+    merged_data = merge_json_files(input_files)
+    if merged_data:
+        save_json_file(merged_data, output_file)
+    else:
+        print("\nNo data was merged. Output file will not be created.")