Python中处理文件移动时的Windows权限错误及fitz库的最佳实践

import os
import shutil
import re
import fitz # PyMuPDF

origin_directory = "path/to/origin"
holding_Directory = "path/to/holding"
invoice_no_search_terms = ["Invoice No", "INV#"]
invoice_numbers = []

for filename in os.listdir(origin_directory):
    file_path = os.path.join(origin_directory, filename)
    print(f"Processing file: {file_path}")

    found_Inv_no = False
    doc = None # Initialize doc outside try-with for broader scope if needed

    try:
        with fitz.open(file_path) as doc: # 文件句柄在此处打开
            print("Document opened.")
            for page in doc: # 迭代页面
                text = page.get_text("utf-8")
                for term in invoice_no_search_terms:
                    if term in text:
                        match = re.findall(r"\d+", text[text.find(term):], re.IGNORECASE)
                        invoice_numbers.extend(match)
                        found_Inv_no = True
                        # fitz.close(doc) # <--- 错误：不应在with块内手动关闭
                        break # 找到即跳出内部循环
                if found_Inv_no:
                    break # 找到即跳出页面循环
        # with 块在此处结束，fitz 会自动关闭 doc，释放文件句柄
        print("Document closed by 'with' statement.")

        if not found_Inv_no:
            moved_name = "moved " + filename
            new_file_path = os.path.join(holding_Directory, moved_name)
            print(f"Moving file from {file_path} to {new_file_path}")
            shutil.move(file_path, new_file_path) # <--- 尝试移动文件
            print(f"Successfully moved: {filename}")
        else:
            print(f"Invoice number found for {filename}. Keeping in origin directory.")

    except UnicodeDecodeError:
        print(f"Warning: Unable to decode content from {filename}.")
    except ValueError as e:
        # 原始问题中出现的 ValueError: document closed 错误通常是由于在with块内
        # 提前调用 fitz.close(doc) 导致，with块退出时再次尝试关闭已关闭的文档。
        print(f"Error processing {filename} (ValueError): {e}. This might be due to premature document closure.")
    except PermissionError as e:
        print(f"PermissionError moving {filename}: {e}. Ensure no other processes are holding the file.")
    except Exception as e:
        print(f"An unexpected error occurred with {filename}: {e}")

import os
import shutil
import re
import fitz # PyMuPDF
import time

origin_directory = "path/to/origin" # 替换为你的源目录
holding_Directory = "path/to/holding" # 替换为你的目标目录
invoice_no_search_terms = ["Invoice No", "INV#"]
invoice_numbers = []

for filename in os.listdir(origin_directory):
    file_path = os.path.join(origin_directory, filename)
    print(f"\n--- Processing file: {file_path} ---")

    found_Inv_no = False

    # 步骤1: 确保fitz文件操作在独立的with块中完成，并在with块结束后再进行移动操作
    try:
        # 使用with语句确保PDF文件句柄在处理完毕后自动关闭
        with fitz.open(file_path) as doc:
            print("Document opened for processing.")
            for page in doc: # 迭代PDF的每一页
                text = page.get_text("utf-8") # 获取页面文本
                for term in invoice_no_search_terms:
                    if term in text:
                        match = re.findall(r"\d+", text[text.find(term):], re.IGNORECASE)
                        invoice_numbers.extend(match)
                        found_Inv_no = True
                        print(f"Found invoice number for '{filename}'.")
                        break # 找到即跳出内部循环
                if found_Inv_no:
                    break # 找到即跳出页面循环
        # with 块在此处结束，doc 文件句柄已由 fitz 自动关闭和释放
        print("Document processing complete. File handle released by 'with' statement.")

        # 步骤2: 在确认文件句柄已释放后，再尝试移动文件
        if not found_Inv_no:
            moved_name = "moved " + filename
            new_file_path = os.path.join(holding_Directory, moved_name)

            # 尝试移动文件，并处理可能出现的权限错误
            max_retries = 3
            for attempt in range(max_retries):
                try:
                    print(f"Attempting to move file from {file_path} to {new_file_path} (Attempt {attempt + 1}/{max_retries})")
                    shutil.move(file_path, new_file_path)
                    print(f"Successfully moved: {filename}")
                    break # 移动成功，跳出重试循环
                except PermissionError as e:
                    print(f"PermissionError moving {filename}: {e}. Retrying in 1 second...")
                    time.sleep(1) # 等待1秒后重试
                except Exception as e:
                    print(f"An unexpected error occurred while moving {filename}: {e}")
                    break # 其他错误，不再重试
            else:
                print(f"Failed to move {filename} after {max_retries} attempts due to PermissionError.")
        else:
            print(f"Invoice number found for {filename}. Keeping in origin directory.")

    except fitz.fitz.EmptyFileError:
        print(f"Warning: {filename} is an empty or corrupt PDF file.")
    except Exception as e:
        print(f"An error occurred during processing of {filename}: {e}")