feat: 添加collect_pid_files函数，支持从进程输出文件中收集文件及处理结果

zhch/utils/__init__.py

@@ -18,7 +18,8 @@ from .file_utils import (
     get_image_files_from_list,
     get_image_files_from_csv,
     split_files,
-    create_temp_file_list
+    create_temp_file_list,
+    collect_pid_files,
 )
 
 __all__ = [
@@ -32,7 +33,8 @@ __all__ = [
     'get_image_files_from_list',
     'get_image_files_from_csv',
     'split_files',
-    'create_temp_file_list'
+    'create_temp_file_list',
+    'collect_pid_files'
 ]
 
 __version__ = "1.0.0"

zhch/utils/file_utils.py

@@ -1,6 +1,7 @@
 import tempfile
 from pathlib import Path
-from typing import List
+from typing import List, Tuple
+import json
 
 def split_files(file_list: List[str], num_splits: int) -> List[List[str]]:
     """
@@ -130,4 +131,47 @@ def get_image_files_from_csv(csv_file: str, status_filter: str = "fail") -> List
             if status.lower() == status_filter.lower():
                 image_files.append(image_file)
 
-    return image_files
+    return image_files
+
+
+def collect_pid_files(pid_output_file: str) -> List[Tuple[str, str]]:
+    """
+    从进程输出文件中收集文件
+
+    Args:
+        pid_output_file: 进程输出文件路径
+
+    Returns:
+        文件列表(文件路径，处理结果)
+    """
+
+    """
+    单进程结果统计文件格式
+    "results": [
+    {
+      "image_path": "docstructbench_dianzishu_zhongwenzaixian-o.O-61520612.pdf_140.jpg",
+      "processing_time": 2.0265579223632812e-06,
+      "success": true,
+      "device": "gpu:3",
+      "output_json": "/home/ubuntu/zhch/PaddleX/zhch/OmniDocBench_Results_Scheduler/process_3/docstructbench_dianzishu_zhongwenzaixian-o.O-61520612.pdf_140.json",
+      "output_md": "/home/ubuntu/zhch/PaddleX/zhch/OmniDocBench_Results_Scheduler/process_3/docstructbench_dianzishu_zhongwenzaixian-o.O-61520612.pdf_140.md"
+    },
+    ...
+    """
+    if not Path(pid_output_file).exists():
+        print(f"⚠️ Warning: PID output file not found: {pid_output_file}")
+        return []
+
+    with open(pid_output_file, 'r', encoding='utf-8') as f:
+        data = json.load(f)
+
+    if not isinstance(data, dict) or "results" not in data:
+        print(f"⚠️ Warning: Invalid PID output file format: {pid_output_file}")
+        return []
+    # 返回文件路径和处理状态, 如果“success”: True, 则状态为“success”, 否则为“fail”
+    file_list = []
+    for file_result in data.get("results", []):
+        image_path = file_result.get("image_path", "")
+        status = "success" if file_result.get("success", False) else "fail"
+        file_list.append((image_path, status))
+    return file_list