Python操作Mongodb2

Python操作Mongodb的高级操作

数据准备，向Mongodb中的test库下的students集合中插入4条数据：

from pymongo import MongoClient

client = MongoClient()
database = client.test
collection = database.students

students = [
    {"name":"张三","gender":"男","age":20,"score":89},
    {"name":"李四","gender":"男","age":21,"score":78},
    {"name":"小花","gender":"女","age":19,"score":95},
    {"name":"小白","gender":"女","age":21,"score":83}
]
collection.insert_many(students)

1、显式的OR查询：

语法：find({"$or":[{条件1},{条件2}...]}) ，例如：

students = collection.find({"$or":[{"age":{"$gt":20}},{"score":{"$gt":90}}]}) # 查询age>20或score>90的记录

2、显式的AND查询：

语法：find({"$and":[{条件1},{条件2}...]}) ，例如：

students = collection.find({"$and":[{"age":{"$gt":20}},{"score":{"$gt":80}}]}) # 查询age>20且score>80的记录

3、在AND中嵌套OR查询：

students = collection.find({"$and":[{"$or":[{"age":{"$gt":20}},{"gender":"女"}]},{"$or":[{"score":{"$gt":90}},{"name":"小白"}]}]})  # 查询（age>20或者gender为女）并且（score>90或name为小白）的记录

4、当记录中嵌套有字典和列表时的相关操作：

数据准备，向Mongodb中的test库下的students集合中插入5条数据：

from pymongo import MongoClient

client = MongoClient()
database = client.test
collection = database.students

students = [
     {"name":"张三","address":{"province":"广东省","city":"深圳"},"score":[80,87,79]},
     {"name":"李四","address":{"province":"广东省","city":"广州"},"score":[90,80,95]},
     {"name":"王五","address":{"province":"广东省","city":"深圳"},"score":[78,86,96]},
     {"name":"赵六","address":{"province":"广东省","city":"深圳"},"score":[90,92,86]},
     {"name":"刘七","address":{"province":"广东省","city":"东莞"},"score":[69,97,98]}
]

collection.insert_many(students)

① 使用点符号定位嵌套的字典中字段

datas = collection.find({"address.city":"深圳"}) # 查询city字段值为深圳的所有记录
datas = collection.find({"$or":[{"address.city":"深圳"},{"address.city":"广州"}]}) # 查询city字段值为深圳或广州的所有记录

② 记录中嵌套数组的操作

datas = collection.find({"score":80}) # 查询数组score中包含80的所有记录
datas = collection.find({"score.0":80}) # 查询数组score中的第一个元素为80的所有记录
datas = collection.find({"score":{"$ne":80}}) # 查询数组score中不包含80的所有记录
datas = collection.find({"score":{"$size":3}}) # 查询数组score长度为3的所有记录

5、聚合操作

语法格式：aggregate([{聚合条件1},{聚合条件2}...])

聚合条件可以有多个，前一个聚合条件的输出，作为下一个聚合条件的输入，当聚合条件为0个时，相当于find()

datas = collection.aggregate([]) # 查询所有数据相当于find()

聚合关键字$match：筛选符合条件的记录，用法跟find()的第一个参数类似

datas = collection.aggregate([{"$match":{"address.city":"深圳"}}]) # 筛选city字段值为深圳的记录

聚合关键字$project：

①可以指定返回部分字段值或指定不返回某些字段值，用法跟find()的第二个参数类似，返回则value为1，不返回则value为0：

datas = collection.aggregate([{"$project":{"_id":0,"name":1,"score":1}}]) # 只返回记录中的name和score字段值，且不返回_id字段值

②将已有的字段的值赋值给新的字段（新的字段可以不存在）

datas = collection.aggregate([{"$project":{"城市":"$address.city"}}]) # 将address下的city的值赋值给新字段“城市”

③特殊值处理，当值为0或1这样的情况，要使用$literal关键字，否则会报错

datas = collection.aggregate([{"$project":{"score":{"$literal":1}}}]) # 将所有记录的score赋值为1

聚合关键字$group：

插入新数据：

students = [
     {"name":"张三","score":90,"date":"2019-03-16"},
     {"name":"李四","score":87,"date":"2019-03-16"},
     {"name":"王五","score":89,"date":"2019-03-16"},
     {"name":"张三","score":88,"date":"2019-03-17"},
     {"name":"李四","score":98,"date":"2019-03-17"},
     {"name":"王五","score":86,"date":"2019-03-17"}
]
collection.insert_many(students)

①使用$group进行分组操作（会去重）

datas = collection.aggregate([{"$group":{"_id":"$name"}}]) # 按name字段进行分组

返回结果：

②分组并计算最大值、最小值、平均值、总和、记录条数等结果

datas = collection.aggregate([{"$group":{
                               "_id":"$name",                   # 根据name进行分组
                               "max_score":{"$max":"$score"},   # 获取最大值
                               "min_score":{"$min":"$score"},   # 获取最小值
                               "avg_score":{"$avg":"$score"},   # 获取平均值
                               "sum_score":{"$sum":"$score"},   # 获取总分数
                               "count":{"$sum":1}               # 统计每个name参与计算的记录数
                               }}])

返回结果：

③获取最早或最晚插入的记录

datas = collection.aggregate([{"$group":{"_id":"$name","date":{"$last":"$date"},"score":{"$last":"$score"}}}]) # 根据name字段分组，取最后插入的date和score字段值

返回结果：

datas = collection.aggregate([{"$group":{"_id":"$name","date":{"$first":"$date"},"score":{"$first":"$score"}}}]) # 根据name字段分组，取最早插入的date和score字段值

返回结果：

$last 和 $first 可以混合使用：

datas = collection.aggregate([{"$group":{"_id":"$name","date":{"$first":"$date"},"score":{"$last":"$score"}}}]) # # 根据name字段分组，取最早插入的date和最晚插入的score字段值

返回结果：

聚合关键字$unwind：