原文地址: https://ethereum.stackexchange.com/questions/19341/address-send-vs-address-transfer-best-practice-usage

address.transfer()

  • throws on failure
  • forwards 2,300 gas stipend (not adjustable), safe against reentrancy
  • should be used in most cases as it's the safest way to send ether

address.send()

  • returns false on failure
  • forwards 2,300 gas stipend (not adjustable), safe against reentrancy
  • should be used in rare cases when you want to handle failure in the contract

address.call.value().gas()()

  • returns false on failure
  • forwards all available gas (adjustable), not safe against reentrancy
  • should be used when you need to control how much gas to forward when sending ether or to call a function of another contract

Detailed version below:

The relative tradeoffs between the use of someAddress.send()someAddress.transfer(), and someAddress.call.value()():

  • someAddress.send()and someAddress.transfer() are considered safe against reentrancy. While these methods still trigger code execution, the called contract is only given a stipend of 2,300 gas which is currently only enough to log an event.
  • x.transfer(y) is equivalent to require(x.send(y)), it will automatically revert if the send fails.
  • someAddress.call.value(y)() will send the provided ether and trigger code execution. The executed code is given all available gas for execution making this type of value transfer unsafe against reentrancy.

Using send() or transfer() will prevent reentrancy but it does so at the cost of being incompatible with any contract whose fallback function requires more than 2,300 gas. It is also possible to use someAddress.call.value(ethAmount).gas(gasAmount)() to forward a custom amount of gas.

One pattern that attempts to balance this trade-off is to implement both a push and pull mechanism, using send() or transfer() for the push component and call.value()() for the pull component.

It is worth pointing out that exclusive use of send() or transfer() for value transfers does not itself make a contract safe against reentrancy but only makes those specific value transfers safe against reentrancy.

More details are here https://consensys.github.io/smart-contract-best-practices/recommendations/#be-aware-of-the-tradeoffs-between-send-transfer-and-callvalue

Reasons for adding transfer()https://github.com/ethereum/solidity/issues/610


call() can also be used to issue a low-level CALL opcode to make a message call to another contract:

if (!contractAddress.call(bytes4(keccak256("someFunc(bool, uint256)")), true, 3)) {
revert;
}

The forwarded value and gas can be customized:

contractAddress.call.gas(5000)
.value(1000)(bytes4(keccak256("someFunc(bool, uint256)")), true, 3);

This is equivalent to using a function call on a contract:

SomeContract(contractAddress).someFunc.gas(5000)
.value(1000)(true, 3);

Beware of the right padding of the input data in call()https://github.com/ethereum/solidity/issues/2884


transfer()send() and call() functions are translated by the Solidity compiler into the CALLopcode.

As explained on the Subtleties page in Ethereum's wiki:

CALL has a multi-part gas cost:

  • 700 base
  • 9000 additional if the value is nonzero
  • 25000 additional if the destination account does not yet exist (note: there is a difference between zero-balance and nonexistent!)

Solidity transfer vs send 区别的更多相关文章

  1. read、write 与recv、send区别 gethostname

    recv相对于read有什么区别呢? 其实它跟read函数功能一样,都可以从套接口缓冲区sockfd中取数据到buf,但是recv仅仅只能够用于套接口IO,并不能用于文件IO以及其它的IO,而read ...

  2. 页面跳转Transfer与Redirect的区别你知道吗?

    一 前言 关于页面跳转的方式常用的应该就是,链接跳转,js跳转,Server.Tranfser和Response.Redirect 这几种,可是在Tranfser与Redirect之间用哪种更好(本文 ...

  3. Response.Redirect()、Server.Execute和Server.Transfer的区别

    1.Response.Redirect(): Response.Redirect方法导致浏览器链接到一个指定的URL. 当Response.Redirect()方法被调用时,它会创建一个应答,应答头中 ...

  4. 【转】页面跳转Transfer与Redirect的区别你知道吗?

    一 前言 关于页面跳转的方式常用的应该就是,链接跳转,js跳转,Server.Tranfser和Response.Redirect 这几种,可是在Tranfser与Redirect之间用哪种更好(本文 ...

  5. 页面跳转 Server.Transfer和 Response.Redirect的区别

    1.Server.Transfer 用于把处理的控制权从一个页面转移到另一个页面,在转移的工程中没有离开服务器内部控件(如request,session等)保存的信息不变.因此你能从a页面跳转到b页面 ...

  6. Solidity

    起因是Xenc师傅给我截了张图,我日 居然看不懂 ,一搜才知道,之前学的版本有些老了.. 这次学下新一点的记录下 HelloWorld pragma solidity ^0.6.0; // versi ...

  7. Solidity的三种转账方式与比较

    转账的3种方式 123 address.transfer()address.send()address.call.value().gas()() 转账transfer 12345678910 func ...

  8. java并发:阻塞队列

    第一节 阻塞队列 1.1 初识阻塞队列 队列以一种先进先出的方式管理数据,阻塞队列(BlockingQueue)是一个支持两个附加操作的队列,这两个附加的操作是:在队列为空时,获取元素的线程会等待队列 ...

  9. 10分钟 5步 发布以太坊 ERC20 代币

    1.安装 METAMASK Brings Ethereum to your browser 一个可以浏览器上进行操作的以太坊钱包,推荐 Chrome. Chrome 插件安装地址: https://c ...

随机推荐

  1. ubuntn 配置webpy nginx

    webpy环境搭建 在开始webpy搭建之前,有必要熟悉一下什么是fastcgi,因为搭建环境时都是使用这个模式去运行webpy程序的,具体的fastcgi描述可以参考各种百科:fastcgi协议官网 ...

  2. 如何在本地浏览器访问nginx

    1.打开vmware"编辑虚拟机"设置,点击“网络适配器”选择“桥联模式”: 2.开启该虚拟机,输入用户名root及密码登陆服务器: 3.以管理员身份打开cmd,在命令窗口输入ip ...

  3. SpringMVC的启动过程

    前言 下面是一个SpringMVC应用的配置文件,需要注意两个地方,一个是ContextLoaderListener,一个是dispatcherServlet.web容器正是通过这两个配置才和spri ...

  4. [原]zeromq框架测试报告

    一.环境: 服务器:linux 4核 16G 虚拟机 1台 客户端:linux 4核 16G 2000台(模拟) 数据包大小:1036字节 二.参数设置: ulimit -n 65536 服务端处理线 ...

  5. fn project 生产环境使用

    此为官方的参考说明   Running Fn in Production The QuickStart guide is intended to quickly get started and kic ...

  6. 洛谷 2680 (NOIp2015) 运输计划

    题目:https://www.luogu.org/problemnew/show/P2680 因为是最长的时间最短,所以二分! 离线LCA可以知道路径长度.每次只看超过二分值的路径. 原本的想法是遍历 ...

  7. linux下PS1命令提示符设置

    linux下PS1命令提示符设置 在此文件最后一行添加:vim /etc/profileexport PS1='[\u@\h \W]\$ '   #这里必须用单引号.     \d :代表日期,格式为 ...

  8. Git&Repo 命令大全 ***

    首先理解几个基本概念: origin:默认远程版本库: master:默认开发分支: 查看本地更新状态: git status jiangzhaowei@ubuntu$ git status # On ...

  9. (转)pipe row的用法, Oracle split 函数写法.

    本文转载自:http://www.cnblogs.com/newsea/archive/2010/12/14/1905482.html 关于 pipe row的用法2009/12/30 14:53 = ...

  10. 分布式缓存系统 Memcached 半同步/半异步模式

    在前面工作线程初始化的分析中讲到Memcached采用典型的Master_Worker模式,也即半同步/半异步的高效网络并发模式.其中主线程(异步线程)负责接收客户端连接,然后分发给工作线程,具体由工 ...