准备数据
```python
from pymongo import MongoClient
import datetime
client=MongoClient('mongodb://localhost:27017')
table=client['db1']['emp']
l=[
('张飞','male',18,'20170301','',7300.33,401,1), #以下是教学部
('张云','male',78,'20150302','teacher',1000000.31,401,1),
('刘备','male',81,'20130305','teacher',8300,401,1),
('关羽','male',73,'20140701','teacher',3500,401,1),
('曹操','male',28,'20121101','teacher',2100,401,1),
('诸葛亮','female',18,'20110211','teacher',9000,401,1),
('周瑜','male',18,'19000301','teacher',30000,401,1),
('司马懿','male',48,'20101111','teacher',10000,401,1),
('袁绍','female',48,'20150311','sale',3000.13,402,2),#以下是销售部门
('张全蛋','female',38,'20101101','sale',2000.35,402,2),
('鹌鹑蛋','female',18,'20110312','sale',1000.37,402,2),
('王尼玛','female',18,'20160513','sale',3000.29,402,2),
('我尼玛','female',28,'20170127','sale',4000.33,402,2),
('杨过','male',28,'20160311','operation',10000.13,403,3), #以下是运营部门
('小龙女','male',18,'19970312','operation',20000,403,3),
('郭靖','female',18,'20130311','operation',19000,403,3),
('黄蓉','male',18,'20150411','operation',18000,403,3),
('梅超风','female',18,'20140512','operation',17000,403,3)
]
for n,item in enumerate(l):
d={
"_id":n,
'name':item[0],
'sex':item[1],
'age':item[2],
'hire_date':datetime.datetime.strptime(item[3],'%Y%m%d'),
'post':item[4],
'salary':item[5]
}
table.save(d)
# 准备数据
```
分组的概念与mysql相同,以某个字段作为依据进行归类,其目的是为了统计
## $match
```python
#match 用于对数据进行筛选
{"$match":{"字段":"条件"}},可以使用任何常用查询操作符$gt,$lt,$in等
#例1、select * from db1.emp where post='teacher';
db.emp.aggregate({"$match":{"post":"teacher"}})
#例2、select * from db1.emp where id > 3;
db.emp.aggregate(
{"$match":{"_id":{"$gt":3}}},
)
```
## $project
```python
# project翻译为投射 ,即将一个数据结果映射为另一个结果 过程中可以对某些数据进行修改 控制其最终显示的结果
{"$project":{"要保留的字段名":1,"要去掉的字段名":0,"新增的字段名":"表达式"}}
#1、select name,post,(age+1) as new_age from db1.emp;
db.emp.aggregate(
{"$project":{
"name":1,
"post":1
}})
#2、表达式之数学表达式
{"$add":[expr1,expr2,...,exprN]} #相加
{"$subtract":[expr1,expr2]} #第一个减第二个
{"$multiply":[expr1,expr2,...,exprN]} #相乘
{"$divide":[expr1,expr2]} #第一个表达式除以第二个表达式的商作为结果
{"$mod":[expr1,expr2]} #第一个表达式除以第二个表达式得到的余数作为结果
#例:所有人年龄加1显示
db.emp.aggregate(
{"$project":{
"name":1,
"post":1,
"new_age":{"$add":["$age",1]}
}})
# 错误示范: 原因:参加运算的字段不能被影藏
db.emp.aggregate(
{"$project":{
"name":1,
"salary":1,
"age":0,
"new_age":{"$add":["$age",1]}
}})
#3、表达式之日期表达式:$year,$month,$week,$dayOfMonth,$dayOfWeek,$dayOfYear,$hour,$minute,$second
#例如：select name,date_format("%Y") as hire_year from db1.emp
db.emp.aggregate(
{"$project":{"name":1,"hire_year":{"$year":"$hire_date"}}}
)
#例如查看每个员工的工作多长时间
db.emp.aggregate(
{"$project":{"name":1,"hire_period":{
"$subtract":[
{"$year":new Date()},
{"$year":"$hire_date"}
]
}}}
)
#4、字符串表达式
{"$substr":[字符串/$值为字符串的字段名,起始位置,截取几个字节]}
db.emp.aggregate({"$project":{"new_name":{"$substr":["$name",0,3]}}})
{"$concat":[expr1,expr2,...,exprN]} #指定的表达式或字符串连接在一起返回,只支持字符串拼接
db.emp.aggregate({"$project":{"new_name":{"$concat":["$name","$post"]}}})
{"$toLower":expr}
{"$toUpper":expr}
db.emp.aggregate({"$project":{"new_name":{"$toUpper":"$post"}}})
db.emp.aggregate( {"$project":{"NAME":{"$toUpper":"$name"}}})
#5、逻辑表达式
$and
$or
$not
其他见Mongodb权威指南
```
## $group
```python
# $group用于分组
# 分组后具体信息被影藏
db.emp.aggregate(
{"$match":{"_id":{"$gt":3}}},
{"$group":{"_id":"$post"}}
)
# 通常我们要对分组后的内容进行统计这就需要对应的几个聚合函数
# select id,avg(salary) from db1.emp where id > 3 group by post;
db.emp.aggregate(
{"$match":{"_id":{"$gt":3}}},
{"$group":{"_id":"$post",'avg_salary':{"$avg":"$salary"}}},
)
# math用于匹配 与mysql不同的是没有顺序限制 每一个操作像是一个管道接收上一个的数据进行处理再传给下一个
# select id,avg(salary) from db1.emp where id > 3 group by post having avg(salary) > 10000;
db.emp.aggregate(
{"$match":{"_id":{"$gt":3}}},
{"$group":{"_id":"$post",'avg_salary':{"$avg":"$salary"}}},
{"$match":{"avg_salary":{"$gt":10000}}}
)
# 对应的聚合函数 $sum、$avg、$max、$min、$first、$last
#1、将分组字段传给$group函数的_id字段即可
{"$group":{"_id":"$sex"}} #按照性别分组
{"$group":{"_id":"$post"}} #按照职位分组
{"$group":{"_id":{"state":"$state","city":"$city"}}} #按照多个字段分组，比如按照州市分组
#2、分组后聚合得结果,类似于sql中聚合函数的聚合操作符：$sum、$avg、$max、$min、$first、$last
#例1：select post,max(salary) from db1.emp group by post;
db.emp.aggregate({"$group":{"_id":"$post","max_salary":{"$max":"$salary"}}})
#例2：去每个部门最大薪资与最低薪资
db.emp.aggregate({"$group":{"_id":"$post","max_salary":{"$max":"$salary"},"min_salary":{"$min":"$salary"}}})
#例3：如果字段是排序后的，那么$first,$last会很有用,比用$max和$min效率高
db.emp.aggregate({"$group":{"_id":"$post","first_id":{"$first":"$_id"}}})
#例4：求每个部门的总工资
db.emp.aggregate({"$group":{"_id":"$post","count":{"$sum":"$salary"}}})
#例5：求每个部门的人数
db.emp.aggregate({"$group":{"_id":"$post","count":{"$sum":1}}})
#3、数组操作符
{"$addToSet":expr}：不重复
{"$push":expr}：重复
# 等同于group_concat
#例：查询岗位名以及各岗位内的员工姓名:select post,group_concat(name) from db1.emp group by post;
db.emp.aggregate({"$group":{"_id":"$post","names":{"$push":"$name"}}})
db.emp.aggregate({"$group":{"_id":"$post","names":{"$addToSet":"$name"}}})
```
## $sort ,limit,skip
```python
{"$sort":{"字段名":1,"字段名":-1}} #1升序，-1降序
{"$limit":n}
{"$skip":n} #跳过多少个文档
#例1、取平均工资最高的前两个部门
db.emp.aggregate(
{
"$group":{"_id":"$post","平均工资":{"$avg":"$salary"}}
},
{
"$sort":{"平均工资":-1}
},
{
"$limit":2
}
)
#例2、
db.emp.aggregate(
{
"$group":{"_id":"$post","平均工资":{"$avg":"$salary"}}
},
{
"$sort":{"平均工资":-1}
},
{
"$limit":2
},
{
"$skip":1
}
)
排序：$sort、限制：$limit、跳过：$skip
```
## $sample
```python
# 随机取出n条记录
#集合users包含的文档如下
{ "_id" : 1, "name" : "dave123", "q1" : true, "q2" : true }
{ "_id" : 2, "name" : "dave2", "q1" : false, "q2" : false }
{ "_id" : 3, "name" : "ahn", "q1" : true, "q2" : true }
{ "_id" : 4, "name" : "li", "q1" : true, "q2" : false }
{ "_id" : 5, "name" : "annT", "q1" : false, "q2" : true }
{ "_id" : 6, "name" : "li", "q1" : true, "q2" : true }
{ "_id" : 7, "name" : "ty", "q1" : false, "q2" : true }
#下述操作时从users集合中随机选取3个文档
db.users.aggregate({"$sample":{"size":3}})
随机选取n个：$sample
```
# 可视化工具
https://robomongo.org
from selenium.webdriver import Chrome
from urllib.parse import urlencode
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
import time
import mongo
driver=Chrome()
name='黄金'
encode_dict=urlencode({"keyword":name,"enc":"utf-8","wq":name})
url='https://search.jd.com/Search?'+encode_dict
driver.get(url)
def spider():
WebDriverWait(driver, 10).until(EC.element_to_be_clickable((By.CLASS_NAME, 'pn-next')))
height = driver.execute_script("return document.body.clientHeight")
driver.execute_script("""
window.scrollTo({
top: %s,
behavior: "smooth"
});""" % height)
items = driver.find_elements_by_css_selector(".gl-item")
if len(items)==60:
for i in items:
price=i.find_element_by_css_selector('.p-price i').text
img=i.find_element_by_css_selector('.p-scroll img').get_attribute('src')
url=i.find_element_by_css_selector('.p-img a').get_attribute('href')
print(price)
mongo_dict={'img':img,'price':price,'url':url}
mongo.insert(mongo_dict)
WebDriverWait(driver, 30).until(EC.element_to_be_clickable((By.CLASS_NAME, 'pn-next')))
next_page = driver.find_element_by_css_selector('.pn-next')
next_page.click()
time.sleep(3)
spider()
else:
return spider()
for i in range(5):
time.sleep(1)
spider()
time.sleep(6)
driver.close()
from pymongo import MongoClient
import datetime
c=MongoClient(host='127.0.0.1',port=27017)
db=c['admin']
db.authenticate('root','123')
db.c['db1']
def insert(data):
c['db1']['jingdong'].insert(data)
if __name__=='__main__':
insert({"url":"24qew"})