百度网盘、夸克网盘、123网盘等备份云网盘，相册、视频文件备份恢复到手机后排序乱解决办法（python版）
·
楼主小米手机的五千多张纪念性照片疑似因系统bug丢失，因开启夸克网盘自动备份，遂清空本地相册，下载备份文件覆盖使用。但部分图片、视频无拍摄时间，导致相册乱序。
楼主结合EXIF信息恢复（无此信息则从文件命名提取时间，如时间戳、日期格式命名的文件），解决该问题。
相较于网上教程，本次脚本优化：采用递归方法，可自动逐级查找文件，无需手动更换路径；新增视频拍摄时间自动提取处理功能；失败文件将导出至桌面，方便二次兼容处理。
直接上教程。
废话不多说，直接上教程
照片存有EXIF信息
可以在这个网站查看是否存有exif
https://exif.tuchong.com/
准备工作
首先官网下载vscode，安装
https://code.visualstudio.com/
安装完成后按我的操作进行插件下载
随后打开CMD分别输入
pip install Pillow
pip3 install datetime
在Python中，datetime 模块提供了处理日期和时间的功能。
Pillow库来读取照片的exif信息，并使用os库来更改文件的修改日期。
安装完成后会提示成功
再重新打开vscode，新建一个python文件
文件→新建文件→python文件
配置文件 analyze_failures.py
import os
import datetime
import re
from PIL import Image
dir_path = r"D:\temp"
# 模拟当前的匹配逻辑
def get_date_from_file(file_path, file_name):
# ... (same logic as before) ...
dt = None
# ... (Previous rules omitted for brevity in thought block, but included in execution) ...
# Rule 1: YYYYMMDD[-_]HHMMSS
match_date = re.search(r'(\d{4})(\d{2})(\d{2})[-_](\d{2})(\d{2})(\d{2})', file_name)
if match_date:
try:
year, month, day, hour, minute, second = map(int, match_date.groups())
if 1970 <= year <= 2050:
return "Rule 1"
except ValueError:
pass
# Rule 2: YYYY-MM-DD-HH-MM-SS-mmm
match_date_long = re.search(r'(\d{4})-(\d{2})-(\d{2})-(\d{2})-(\d{2})-(\d{2})[-_](\d{3})', file_name)
if match_date_long:
try:
year, month, day, hour, minute, second, microsecond = map(int, match_date_long.groups())
if 1970 <= year <= 2050:
return "Rule 2"
except ValueError:
pass
# Rule 3: YYYYMMDDHHMMSS
match_date_pure = re.search(r'(\d{4})(\d{2})(\d{2})(\d{2})(\d{2})(\d{2})', file_name)
if match_date_pure:
try:
year, month, day, hour, minute, second = map(int, match_date_pure.groups())
if 1970 <= year <= 2050:
return "Rule 3"
except ValueError:
pass
# Rule 4: 13-digit timestamp
match = re.search(r'(\d{13})', file_name)
if match:
return "Rule 4 (13-digit)"
# Rule 5: 10-digit timestamp
match_10 = re.search(r'(?<!\d)(\d{10})(?!\d)', file_name)
if match_10:
try:
timestamp_s = int(match_10.group(1))
if 946684800 < timestamp_s < 2524608000:
return "Rule 5 (10-digit)"
except Exception:
pass
return None
unmatched_files = []
for root, dirs, files in os.walk(dir_path):
for file_name in files:
file_path = os.path.join(root, file_name)
if not file_name.lower().endswith(('.jpg', '.jpeg', '.png', '.heic', '.mp4', '.mov')):
continue
result = get_date_from_file(file_path, file_name)
if result is None:
unmatched_files.append(file_name)
# Print MORE files to see patterns better
print(f"Found {len(unmatched_files)} unmatched files.")
print("First 50 unmatched files:")
for f in unmatched_files[:50]:
print(f)
处理主程序-图片 import os.py
import os
import datetime
import re
from PIL import Image
from PIL.ExifTags import TAGS
import openpyxl
# 设置需要处理的根目录路径
# 代码会自动遍历该目录下的所有子文件夹
dir_path = r"D:\temp"
# 用于存储处理失败的文件信息
failed_files = []
# 使用 os.walk 递归遍历目录及其子目录
for root, dirs, files in os.walk(dir_path):
for file_name in files:
# 获取文件的完整路径
file_path = os.path.join(root, file_name)
# 标记是否成功处理时间
processed = False
dt = None
source = ""
# 1. 尝试从 EXIF 获取时间
try:
# 尝试打开图片文件
with Image.open(file_path) as img:
# 获取图片的EXIF信息
exif_data = img._getexif()
# 如果图片包含EXIF信息
if exif_data:
# 获取拍摄日期和时间信息 (36867 是 DateTimeOriginal 的标签 ID)
datetime_str = exif_data.get(36867)
if datetime_str:
# 将字符串解析为datetime对象
try:
dt = datetime.datetime.strptime(datetime_str, "%Y:%m:%d %H:%M:%S")
source = "EXIF"
except ValueError:
# EXIF 日期格式可能不标准
pass
except Exception:
# 不是图片或无法读取 EXIF，忽略错误，尝试下一步
pass
# 2. 如果 EXIF 没有找到时间，尝试从文件名提取时间戳
if dt is None:
# 尝试匹配 YYYYMMDD_HHMMSS 格式，例如 IMG_20241003_110253.jpg
# 兼容 Screenshot_20230726-220119_xxxx.jpg 这种格式 (YYYYMMDD-HHMMSS)
match_date = re.search(r'(\d{4})(\d{2})(\d{2})[-_](\d{2})(\d{2})(\d{2})', file_name)
if match_date:
try:
year, month, day, hour, minute, second = map(int, match_date.groups())
if 1970 <= year <= 2050:
dt = datetime.datetime(year, month, day, hour, minute, second)
source = "文件名日期(YYYYMMDD[-_]HHMMSS)"
except ValueError:
pass
# 尝试匹配 Screenshot_2023-11-03-22-56-14-707_lockscreen.jpg 这种格式
# (YYYY-MM-DD-HH-MM-SS-mmm)
if dt is None:
match_date_long = re.search(r'(\d{4})-(\d{2})-(\d{2})-(\d{2})-(\d{2})-(\d{2})[-_](\d{3})', file_name)
if match_date_long:
try:
year, month, day, hour, minute, second, microsecond = map(int, match_date_long.groups())
if 1970 <= year <= 2050:
# microsecond 是毫秒，需要乘以 1000 转换为微秒
dt = datetime.datetime(year, month, day, hour, minute, second, microsecond * 1000)
source = "文件名日期(YYYY-MM-DD-HH-MM-SS-mmm)"
except ValueError:
pass
# 尝试匹配 YYYYMMDDHHMMSS 格式，例如 beauty_20190608004810.jpg
if dt is None:
match_date_pure = re.search(r'(\d{4})(\d{2})(\d{2})(\d{2})(\d{2})(\d{2})', file_name)
if match_date_pure:
try:
year, month, day, hour, minute, second = map(int, match_date_pure.groups())
# 简单的年份校验，防止匹配到非日期的长数字串
if 1970 <= year <= 2050:
dt = datetime.datetime(year, month, day, hour, minute, second)
source = "文件名日期(YYYYMMDDHHMMSS)"
except ValueError:
pass
# 如果没有匹配到日期格式，再尝试匹配 13 位数字 (毫秒级时间戳)，例如 mmexport1612110000082.jpg
if dt is None:
# 或者匹配 10 位数字 (秒级时间戳) - 这里优先匹配 13 位
match = re.search(r'(\d{13})', file_name)
if match:
timestamp_str = match.group(1)
try:
timestamp_ms = int(timestamp_str)
# 转换为秒
timestamp_s = timestamp_ms / 1000.0
dt = datetime.datetime.fromtimestamp(timestamp_s)
source = "文件名时间戳(13位)"
except Exception:
pass
# 如果没有找到 13 位，也可以尝试找 10 位 (可选，视用户需求而定，用户主要提到 mmexport 格式通常是 13 位)
if dt is None:
match_10 = re.search(r'(?<!\d)(\d{10})(?!\d)', file_name)
if match_10:
try:
timestamp_s = int(match_10.group(1))
# 简单的合理性校验：时间戳应该在 1970 年到 2038 年之间 (0 - 2147483647)
# 这里放宽一点，排除一些明显不是时间戳的数字（比如单纯的序号）
# 2000年是 946684800， 2030年是 1893456000
if 946684800 < timestamp_s < 2524608000:
dt = datetime.datetime.fromtimestamp(timestamp_s)
source = "文件名时间戳(10位)"
except Exception:
pass
# 3. 应用修改
if dt:
try:
# os.utime 需要 (atime, mtime)
os.utime(file_path, (os.path.getatime(file_path), dt.timestamp()))
print(f"{file_name}: 拍摄日期设置为 {dt} (来源: {source}, 路径: {file_path})")
processed = True
except Exception as e:
print(f"{file_name}: 应用时间修改失败: {e}")
failed_files.append({"path": file_path, "name": file_name, "reason": f"应用修改失败: {e}"})
else:
# 只有当确实无法处理时才打印，或者打印出无法识别的文件以便用户处理
print(f"{file_name}: 未找到EXIF信息且无法从文件名提取有效时间戳，跳过 (路径: {file_path})")
failed_files.append({"path": file_path, "name": file_name, "reason": "未找到有效时间信息"})
# 处理完成后，如果有失败的文件，导出到 Excel
if failed_files:
try:
# 获取桌面路径
desktop_path = os.path.join(os.path.expanduser("~"), "Desktop")
excel_path = os.path.join(desktop_path, "处理失败照片清单.xlsx")
# 创建一个新的工作簿
wb = openpyxl.Workbook()
ws = wb.active
ws.title = "Failed Files"
# 添加表头
ws.append(["文件完整路径", "文件完整名", "失败原因"])
# 添加数据
for item in failed_files:
ws.append([item["path"], item["name"], item["reason"]])
# 调整列宽（可选，为了美观）
ws.column_dimensions['A'].width = 50
ws.column_dimensions['B'].width = 30
ws.column_dimensions['C'].width = 30
# 保存文件
wb.save(excel_path)
print(f"\n已将 {len(failed_files)} 个处理失败的文件信息导出到: {excel_path}")
except Exception as e:
print(f"\n导出 Excel 失败: {e}")
else:
print("\n所有文件均处理成功！")
处理主程序-视频 vedio.py
import os
import datetime
import re
import openpyxl
from hachoir.parser import createParser
from hachoir.metadata import extractMetadata
# 设置需要处理的根目录路径
# 代码会自动遍历该目录下的所有子文件夹
dir_path = r"D:\temp"
# 用于存储处理失败的文件信息
failed_files = []
# 支持的视频文件扩展名
VIDEO_EXTENSIONS = (
'.mp4', '.mov', '.avi', '.mkv', '.wmv', '.flv', '.3gp',
'.mpg', '.mpeg', '.m4v', '.webm', '.vob', '.mts', '.m2ts',
'.rm', '.rmvb', '.asf', '.ts', '.dat'
)
def get_video_creation_date(file_path):
"""尝试使用 hachoir 库提取视频元数据中的创建时间"""
parser = None
try:
parser = createParser(file_path)
if not parser:
return None
with parser:
metadata = extractMetadata(parser)
if not metadata:
return None
# 尝试获取创建时间
# 不同格式的元数据键值可能不同，常见的是 'creation_date'
if metadata.has('creation_date'):
return metadata.get('creation_date')
except Exception:
pass
return None
# 使用 os.walk 递归遍历目录及其子目录
for root, dirs, files in os.walk(dir_path):
for file_name in files:
# 检查是否为视频文件
if not file_name.lower().endswith(VIDEO_EXTENSIONS):
continue
# 获取文件的完整路径
file_path = os.path.join(root, file_name)
# 标记是否成功处理时间
dt = None
source = ""
# 1. 尝试从视频元数据获取时间
try:
dt_meta = get_video_creation_date(file_path)
if dt_meta:
# 增加年份校验，防止元数据中的时间异常
if 1970 <= dt_meta.year <= 2050:
dt = dt_meta
source = "视频元数据(Metadata)"
except Exception:
pass
# 2. 如果元数据没有找到时间，尝试从文件名提取时间戳 (复用图片处理的逻辑)
if dt is None:
# 尝试匹配 YYYYMMDD_HHMMSS 格式，例如 IMG_20241003_110253.mp4
# 兼容 Screenshot_20230726-220119_xxxx.mp4 这种格式 (YYYYMMDD-HHMMSS)
match_date = re.search(r'(\d{4})(\d{2})(\d{2})[-_](\d{2})(\d{2})(\d{2})', file_name)
if match_date:
try:
year, month, day, hour, minute, second = map(int, match_date.groups())
if 1970 <= year <= 2050:
dt = datetime.datetime(year, month, day, hour, minute, second)
source = "文件名日期(YYYYMMDD[-_]HHMMSS)"
except ValueError:
pass
# 尝试匹配 Screenshot_2023-11-03-22-56-14-707_lockscreen.mp4 这种格式
# (YYYY-MM-DD-HH-MM-SS-mmm)
if dt is None:
match_date_long = re.search(r'(\d{4})-(\d{2})-(\d{2})-(\d{2})-(\d{2})-(\d{2})[-_](\d{3})', file_name)
if match_date_long:
try:
year, month, day, hour, minute, second, microsecond = map(int, match_date_long.groups())
if 1970 <= year <= 2050:
# microsecond 是毫秒，需要乘以 1000 转换为微秒
dt = datetime.datetime(year, month, day, hour, minute, second, microsecond * 1000)
source = "文件名日期(YYYY-MM-DD-HH-MM-SS-mmm)"
except ValueError:
pass
# 尝试匹配 TG-2023-10-10-000149612.mp4 这种格式
# (YYYY-MM-DD-HHMMSSmmm)
if dt is None:
match_date_tg = re.search(r'(\d{4})-(\d{2})-(\d{2})-(\d{2})(\d{2})(\d{2})(\d{3})', file_name)
if match_date_tg:
try:
year, month, day, hour, minute, second, microsecond = map(int, match_date_tg.groups())
if 1970 <= year <= 2050:
dt = datetime.datetime(year, month, day, hour, minute, second, microsecond * 1000)
source = "文件名日期(YYYY-MM-DD-HHMMSSmmm)"
except ValueError:
pass
# 尝试匹配 YYYYMMDDHHMMSS 格式，例如 beauty_20190608004810.mp4
if dt is None:
match_date_pure = re.search(r'(\d{4})(\d{2})(\d{2})(\d{2})(\d{2})(\d{2})', file_name)
if match_date_pure:
try:
year, month, day, hour, minute, second = map(int, match_date_pure.groups())
# 简单的年份校验，防止匹配到非日期的长数字串
if 1970 <= year <= 2050:
dt = datetime.datetime(year, month, day, hour, minute, second)
source = "文件名日期(YYYYMMDDHHMMSS)"
except ValueError:
pass
# 如果没有匹配到日期格式，再尝试匹配 13 位数字 (毫秒级时间戳)，例如 mmexport1612110000082.mp4
if dt is None:
# 针对 mmexport 开头的文件，优先匹配其后的 13 位数字
# 例如 mmexport1771421845216.mp4
if 'mmexport' in file_name:
match_mm = re.search(r'mmexport(\d{13})', file_name)
if match_mm:
try:
timestamp_str = match_mm.group(1)
timestamp_ms = int(timestamp_str)
# 转换为秒
timestamp_s = timestamp_ms / 1000.0
dt_temp = datetime.datetime.fromtimestamp(timestamp_s)
# 增加年份校验
if 1970 <= dt_temp.year <= 2050:
dt = dt_temp
source = "文件名时间戳(mmexport+13位)"
except Exception:
pass
# 或者匹配 10 位数字 (秒级时间戳) - 这里优先匹配 13 位
if dt is None:
match_13 = re.search(r'(\d{13})', file_name)
if match_13:
timestamp_str = match_13.group(1)
try:
timestamp_ms = int(timestamp_str)
# 转换为秒
timestamp_s = timestamp_ms / 1000.0
# 处理可能存在的本地时区问题，这里直接使用 fromtimestamp 获取本地时间
# 如果需要 UTC 时间，可以使用 utcfromtimestamp
try:
dt_temp = datetime.datetime.fromtimestamp(timestamp_s)
except (OSError, OverflowError, ValueError):
# Windows 上时间戳如果是负数或者太大可能会报错
# 对于 1970 年以前的时间戳，或者超大数值，跳过
continue
# 增加年份校验，防止匹配到非时间戳的长数字串 (如哈希值)
# 1970年对应的时间戳是0，2050年对应的时间戳大约是25亿
if 1970 <= dt_temp.year <= 2050:
dt = dt_temp
source = "文件名时间戳(13位)"
except Exception:
pass
# 如果没有找到 13 位，也可以尝试找 10 位 (可选，视用户需求而定，用户主要提到 mmexport 格式通常是 13 位)
if dt is None:
match_10 = re.search(r'(?<!\d)(\d{10})(?!\d)', file_name)
if match_10:
try:
timestamp_s = int(match_10.group(1))
# 简单的合理性校验：时间戳应该在 1970 年到 2038 年之间 (0 - 2147483647)
# 这里放宽一点，排除一些明显不是时间戳的数字（比如单纯的序号）
# 2000年是 946684800， 2030年是 1893456000
if 946684800 < timestamp_s < 2524608000:
dt = datetime.datetime.fromtimestamp(timestamp_s)
source = "文件名时间戳(10位)"
except Exception:
pass
# 3. 应用修改
if dt:
try:
# 获取文件的当前修改时间作为备选 atime
# 使用 os.path.getatime 有时会返回 0 或无效值导致 utime 失败
# 对于整理照片来说，atime 不重要，可以设置为当前时间或与 mtime 一致
current_timestamp = dt.timestamp()
# 打印调试信息
# print(f"Preparing to set time for {file_name}: timestamp={current_timestamp} ({dt})")
# os.utime 需要 (atime, mtime)
# 这里我们将访问时间和修改时间都设置为拍摄时间，这是最安全的做法
os.utime(file_path, (current_timestamp, current_timestamp))
print(f"{file_name}: 拍摄日期设置为 {dt} (来源: {source}, 路径: {file_path})")
except Exception as e:
print(f"{file_name}: 应用时间修改失败: {e} (Attempted timestamp: {dt.timestamp() if dt else 'None'})")
failed_files.append({"path": file_path, "name": file_name, "reason": f"应用修改失败: {e}"})
else:
# 只有当确实无法处理时才打印，或者打印出无法识别的文件以便用户处理
print(f"{file_name}: 未找到元数据信息且无法从文件名提取有效时间戳，跳过 (路径: {file_path})")
failed_files.append({"path": file_path, "name": file_name, "reason": "未找到有效时间信息"})
# 处理完成后，如果有失败的文件，导出到 Excel
if failed_files:
try:
# 获取桌面路径
desktop_path = os.path.join(os.path.expanduser("~"), "Desktop")
excel_path = os.path.join(desktop_path, "视频处理失败清单.xlsx")
# 创建一个新的工作簿
wb = openpyxl.Workbook()
ws = wb.active
ws.title = "Failed Videos"
# 添加表头
ws.append(["文件完整路径", "文件完整名", "失败原因"])
# 添加数据
for item in failed_files:
ws.append([item["path"], item["name"], item["reason"]])
# 调整列宽（可选，为了美观）
ws.column_dimensions['A'].width = 50
ws.column_dimensions['B'].width = 30
ws.column_dimensions['C'].width = 30
# 保存文件
wb.save(excel_path)
print(f"\n已将 {len(failed_files)} 个处理失败的视频文件信息导出到: {excel_path}")
except Exception as e:
print(f"\n导出 Excel 失败: {e}")
else:
print("\n所有视频文件均处理成功！")
需要处理图片和视频依次点击对应的程序即可 vscode-运行-以非调试模式运行按钮
再点击python debugger即可
------------------------完毕 有疑问可添加企鹅 2466961646 有空会进行解答---------
Agent 垂直技术社区，欢迎活跃、内容共建。
更多推荐
6
0- 0
已为社区贡献1条内容
所有评论(0)