Mongo 实践之分组去重 aggregate group

from pymongo import MongoClient

# 连接到 MongoDB
client = MongoClient('mongodb://localhost:27017/')
db = client.your_database_name  # 更改为你的数据库名称
collection = db.datatagging

# 构建聚合管道
pipeline = [
    {
        "$sort": {"unique_path": 1, "modified": -1}  # 按照 unique_path 升序，modified 降序排序
    },
    {
        "$group": {
            "_id": "$unique_path",
            "latestModified": {"$first": "$modified"},  # 获取每个 unique_path 分组中 modified 最新的值
            "document": {"$first": "$$ROOT"}  # 获取整个文档对象
        }
    },
    {
        "$replaceRoot": {"newRoot": "$document"}  # 将保存的整个文档对象作为输出文档
    },
    {
        "$sort": {"modified": -1}  # 按照 modified 字段倒序排列
    }
]

# 执行聚合查询
result = list(collection.aggregate(pipeline))

# 输出结果
ids = []
for doc in result:
    print(doc["_id"])

delete_result = collection.delete_many({"_id": {"$nin": ids}})

# 输出删除的记录数
print("Deleted", delete_result.deleted_count, "documents.")