当前位置:网站首页>Overturn your cognition? The nature of get and post requests
Overturn your cognition? The nature of get and post requests
2022-07-06 04:09:00 【ZNineSun】
GET and POST yes HTTP Two basic methods of requesting , The difference between them , Maybe what we blurted out was :GET Include the parameter in URL in ,POST adopt request body Pass parameters .
Compare the official standard answer Like these :
- GET It is harmless when the browser is backout , and POST The request will be submitted again .
- GET Produced URL The address can be Bookmark, and POST Can not be .
- GET The request will be initiated by the browser cache, and POST Can't , Unless manually set .
- GET Request can only proceed url code , and POST Supports a variety of coding methods .
- GET The request parameters are fully preserved in the browser history , and POST The parameters in are not retained .
- GET Request in URL There is a length limit on the parameters passed in , and POST It has a .
- The data type for the parameter ,GET We only accept ASCII character , and POST There is no limit to .
- GET Than POST It's not safe , Because the parameters are directly exposed to URL On , So it can't be used to convey sensitive information .
- GET Parameters through URL Pass on ,POST Put it in Request body in .
Of course, it's good to answer these questions during the interview , In this article, do you think we just want to give you an interview answer ???no no no
If I told you GET and POST Is there no difference in essence between you and me ?
First of all, we need to understand GET and POST What is it? ?
HTTP Two ways to send requests in the protocol .
HTTP What is it? ?HTTP Is based on TCP/IP Protocol for how data is communicated on the world wide web .
HTTP The bottom is TCP/IP. therefore GET and POST The bottom layer is also TCP/IP, in other words ,GET/POST All are TCP link .GET and POST You can do the same thing . Do you want to GET add request body, to POST close url Parameters , Technically, it works perfectly .
that The problem is coming. , What are the standard answers above ?
TCP It's like a car , We use it TCP To transport data , It is very reliable , It never happens that fewer pieces are lost . But if we run as like as two peas, we'll run the same cars. , So the world looks like a mess , An urgent car may be blocked by a car full of goods ahead , The whole transportation system will be paralyzed .
To avoid that , traffic rules HTTP The birth of .HTTP There are several service categories for car transportation , Yes GET, POST, PUT, DELETE wait ,HTTP Regulations , When executed GET On request , To put a sticker on the car GET The label of ( Set up method by GET), And we need to put the transmitted data on the roof (url in ) To facilitate recording . If it is POST request , It's going to be pasted on the car POST The label of , And put the goods in the carriage .
Of course , You can also be in GET Sneak some goods into the car , But it was disgraceful ; It can also be in POST Put some data on the roof of the car , It feels silly .HTTP Just a Code of conduct , and TCP It's just GET and POST How to achieve the basic .
We only see HTTP Yes GET and POST Parameter transmission channel (url still requrest body) Put forward a request .“ The standard answer ” Where does the parameter size limit come from ?
In my world of the World Wide Web , There is another important role : Transportation company . Different browsers ( launch http request ) And the server ( Accept http request ) Just different shipping companies . Although theoretically , You can pile unlimited loads on the roof (url And an infinite number of arguments ).
But transportation companies are not stupid , Loading and unloading are also very costly , They will limit the number of shipments to limit the risk , Too much data is a burden on both the browser and the server . The unwritten rule in the industry is ,( majority ) Browsers often restrict url The length is in 2K Bytes , and ( majority ) Maximum server processing 64K The size of url. More than , We will not deal with it .
If you use GET service , stay request body Hiding the data , Different servers do different things , Some servers will unload for you , Read the data , Some servers just ignore it , therefore , although GET You can take request body, There is no guarantee that it will be received .
Okay , Now you know ,GET and POST It's essentially TCP link , There is no difference . But because of HTTP The rules and browsers / Server limitations , As a result, they show some differences in the application process .
GET and POST There is another important difference , To put it simply :
GET Produce a TCP Data packets ;POST Produce two TCP Data packets
- about GET Method request , The browser will http header and data Send along , Server response 200( Return the data );
- And for POST, Browser sends first header, Server response 100 continue, The browser sends data, Server response 200 ok( Return the data ).
in other words ,GET It only takes a car trip to get the goods to , and POST Have to run two , The first trip , Go and say hello to the server first “ hi , I have to deliver a shipment later , You open the door to meet me ”, Then go back and deliver the goods .
because POST Two steps are needed. , It takes a little more time , look GET Than POST More effective . therefore Yahoo The team has recommended GET Replace POST To optimize website performance . But this is a pit ! Jump in with caution . Why? ?
1.GET And POST Each has its own semantics , You can't mix them .
2. According to research , In a good network environment , The difference between the time to send a packet and the time to send two packets is negligible . And in the case of bad network environment , The two packages TCP On verifying packet integrity , It has great advantages .
3. Not all browsers will be there POST Send two packets in ,Firefox Just send it once .
So now look back at the standard answer above , Is there a deeper understanding !!!
边栏推荐
- Facebook等大廠超十億用戶數據遭泄露,早該關注DID了
- 【leetcode】1189. Maximum number of "balloons"
- MySQL about self growth
- Global and Chinese markets for fire resistant conveyor belts 2022-2028: Research Report on technology, participants, trends, market size and share
- Use js to complete an LRU cache
- 在 .NET 6 中使用 Startup.cs 更简洁的方法
- Mysql数据库慢sql抓取与分析
- MySQL learning record 13 database connection pool, pooling technology, DBCP, c3p0
- Benefits of automated testing
- Yyds dry goods inventory hcie security Day11: preliminary study of firewall dual machine hot standby and vgmp concepts
猜你喜欢
简易博客系统
In Net 6 CS more concise method
lora网关以太网传输
Cf464e the classic problem [shortest path, chairman tree]
After five years of testing in byte, I was ruthlessly dismissed in July, hoping to wake up my brother who was paddling
【可调延时网络】基于FPGA的可调延时网络系统verilog开发
记一次excel XXE漏洞
C form application of C (27)
[001] [stm32] how to download STM32 original factory data
关于进程、线程、协程、同步、异步、阻塞、非阻塞、并发、并行、串行的理解
随机推荐
Several important classes in unity
MySQL learning record 13 database connection pool, pooling technology, DBCP, c3p0
【leetcode】1189. Maximum number of "balloons"
Execution order of scripts bound to game objects
MySQL reads missing data from a table in a continuous period of time
51nod 1130 n factorial length V2 (Stirling approximation)
Path of class file generated by idea compiling JSP page
Python book learning notes - Chapter 09 section 01 create and use classes
Introduction to data types in MySQL
【按鍵消抖】基於FPGA的按鍵消抖模塊開發
How to modify field constraints (type, default, null, etc.) in a table
Maxay paper latex template description
Mysql数据库慢sql抓取与分析
Tips for using dm8huge table
10个 Istio 流量管理 最常用的例子,你知道几个?
Cf603e pastoral oddities [CDQ divide and conquer, revocable and search set]
Python book learning notes - Chapter 09 section 01 create and use classes
Use js to complete an LRU cache
软考 系统架构设计师 简明教程 | 总目录
80% of the diseases are caused by bad living habits. There are eight common bad habits, which are both physical and mental