MongoDB数据库文件的读与写

首先，在ubuntu16.04上使用Docker启动一个MongoDB 4.0.13的容器，基本操作如下：

hpl@hepengli:~$ sudo -i
[sudo] hpl 的密码： 
root@hepengli:~# systemctl start docker
root@hepengli:~# docker images
REPOSITORY   TAG       IMAGE ID   CREATED   SIZE
root@hepengli:~# docker pull mongo:4.0.13
4.0.13: Pulling from library/mongo
976a760c94fc: Pull complete 
c58992f3c37b: Pull complete 
0ca0e5e7f12e: Pull complete 
f2a274cc00ca: Pull complete 
7f6568107a70: Pull complete 
08957b2477b2: Pull complete 
a66dbd57d2a3: Pull complete 
135a9132a862: Pull complete 
c9b7c17ba1df: Pull complete 
7b595c9e65de: Pull complete 
5e74b440cab6: Pull complete 
9c04ed8f6ca0: Pull complete 
3dbe824121e4: Pull complete 
Digest: sha256:1aea2377c5b17dcd5e67d4c33f8f7ee09da62a4dd27ae988958b70efdf3a0a6a
Status: Downloaded newer image for mongo:4.0.13
docker.io/library/mongo:4.0.13
root@hepengli:~# docker images
REPOSITORY   TAG       IMAGE ID       CREATED       SIZE
mongo        4.0.13    0712bd00d695   2 years ago   416MB
root@hepengli:~# docker run -d -p 27017:27017 --name mongo4.0 0712bd00d695
1e81e661666a5aa01df516663058de8b768a7b09fafa5c53b832c14a6122a97a
root@hepengli:~# docker ps
CONTAINER ID   IMAGE          COMMAND                  CREATED         STATUS         PORTS                                           NAMES
1e81e661666a   0712bd00d695   "docker-entrypoint.s…"   5 seconds ago   Up 3 seconds   0.0.0.0:27017->27017/tcp, :::27017->27017/tcp   mongo4.0

查看linux的IP地址为：192.168.23.129

 在宿主机上可以使用Navicat来测试连接：

 对于16M以内的小文件，可以使用MongoDB直接存储：

import os
import datetime
from pymongo import MongoClient


class DBDataSave(object):
    client = MongoClient('mongodb://192.168.23.129:27017')
    db = client.file_data  # 使用的数据库
    collection = db.datas  # 使用数据库中的集合名称

    def upload_little_file(self):
        upload_dir = r'G:/project/study1/others/mongo_test/upload_files'
        file_list = os.listdir(upload_dir)
        for file in file_list:
            file_path = upload_dir + '/' + file
            file_division = file.split('.')
            file_size = os.stat(file_path).st_size
            f = open(file_path, 'rb')
            data = f.read()
            doc = [{
                    "file_name": file,
                    "filet_type": file_division[-1],
                    'content': data,
                    "size": file_size,
                    'add_time': datetime.datetime.now()
                    }]
            f.close()
            self.collection.insert_many(doc)

    def download_little_file(self):
        if not os.path.exists('./download_files'):
            os.mkdir('./download_files')
        files = self.collection.find()
        for file in files:
            with open('./download_files/' + f"{file.get('file_name')}", 'wb') as f:
                f.write(file.get('content'))


if __name__ == '__main__':
    dds = DBDataSave()
    # dds.upload_little_file()  # 存储文件
    dds.download_little_file()  # 导出文件

 MongoDB内置一套文件系统名为GridFS，可以用来存储大于16M以上的文件：

GridFS具有分布式管理文件的能力，可以突破一般文件系统对file的限制，分段存储，不像普通文件系统是整个存储的。这样读取大型文件时就不会占用大量的内存。

from pymongo import MongoClient
from gridfs import *
import os


class DBDataSaveGF(object):
    client = MongoClient('mongodb://192.168.23.129:27017')  # 连接本地的MongoDB数据库
    db = client.file_data
    file_put = GridFS(db)

    def db_write_file(self):
        upload_path = r'G:/project/study1/others/mongo_test/upload_files'
        files = os.listdir(upload_path)  # 列出文件夹下的所有文件
        for file in files:
            file_path = upload_path + '\\' + file
            with open(file_path, 'rb') as f1:
                self.file_put.put(f1, content_type=file.split('.')[-1], filename=file)

    def db_read_file(self):
        if not os.path.exists('./download_files'):
            os.mkdir('./download_files')
        files = self.file_put.find()
        # files = file_put.find().sort("uploadDate", -1).limit(1)  # 返回最近上传的1个文件
        for file in files:
            with open('./download_files/' + file.filename, 'wb') as f1:
                f1.write(file.read())


if __name__ == '__main__':
    db = DBDataSaveGF()
    # db.db_write_file()  # 存储文件
    db.db_read_file()  # 读取文件