Java学习者

作者：微信小助手

<section style="box-sizing: border-box;"> <section class="V5" style="box-sizing: border-box;" powered-by="xiumi.us"> <section style="margin-top: 10px;margin-bottom: 10px;box-sizing: border-box;"> <section style="padding-top: 10px;padding-right: 10px;padding-left: 10px;box-sizing: border-box;background-color: rgb(239, 239, 239);"> <section style="box-sizing: border-box;"> “ </section> <section style="display: inline-block;vertical-align: top;float: right;width: 90%;line-height: 1.5;font-size: 15px;color: rgb(89, 89, 89);box-sizing: border-box;"> 何为锁？锁在现实中的意义为：封闭的器物，以钥匙或暗码开启。在计算机中的锁一般用来管理对共享资源的并发访问，比如我们 Java 同学熟悉的 Lock，synchronized 等都是我们常见的锁。 </section> <section style="clear: both;box-sizing: border-box;"></section> </section> </section> </section> </section> <h2 style="white-space: normal;line-height: 1.75em;"> </h2> <img class="" data-copyright="0" data-ratio="0.6197387518142236" data-s="300,640" src="/upload/e99b87f1d08fa7b6262f92d7949aa8cd.png" data-type="png" data-w="689" style=""> <h2 style="white-space: normal;text-align: justify;margin-left: 8px;margin-right: 8px;line-height: 1.75em;">当然在我们的数据库中也有锁用来控制资源的并发访问，这也是数据库和文件系统的区别之一。</h2> <section style="box-sizing: border-box;"> <section class="V5" style="box-sizing: border-box;" powered-by="xiumi.us"> <section style="border-bottom-width: 1px;border-bottom-style: solid;border-bottom-color: black;margin-top: 0.5em;margin-bottom: 0.5em;line-height: 1.2;box-sizing: border-box;"> <section style="display: inline-block;border-bottom-width: 6px;border-bottom-style: solid;border-color: rgb(89, 89, 89);margin-bottom: -1px;font-size: 20px;color: rgb(89, 89, 89);box-sizing: border-box;"> 为什么要懂数据库锁？ </section> </section> </section> </section> 通常来说对于一般的开发人员，在使用数据库的时候一般懂点 DQL(select)，DML(insert，update，delete)就够了。 小明是一个刚刚毕业在互联网公司工作的 Java 开发工程师，平常的工作就是完成 PM 的需求。 当然在完成需求的同时肯定逃脱不了 Spring，Spring MVC，Mybatis 的那一套框架，所以一般来说 SQL 还是自己手写，遇到比较复杂的 SQL 会从网上去百度一下。 对于一些比较重要操作，比如交易啊这些，小明会用 Spring 的事务来对数据库的事务进行管理，由于数据量比较小目前还涉及不到分布式事务。 前几个月小明过得都还风调雨顺，直到有一天，小明接了一个需求，商家有个配置项，叫优惠配置项，可以配置买一送一，买一送二等等规则。 当然这些配置是批量传输给后端的，这样就有个问题每个规则都得去匹配，他到底是删除还是添加还是修改，这样后端逻辑就比较麻烦。 聪明的小明想到了一个办法，直接删除这个商家的配置，然后全部添加进去。小明马上开发完毕，成功上线。 开始上线没什么毛病，但是日志经常会出现一些 mysql-insert-deadlock 异常。 由于小明经验比较浅，对于这类型的问题第一次遇见，于是去问了他们组的老司机大红。 大红一看见这个问题，然后看了他的代码之后，输出了几个命令看了几个日志，马上定位了问题，告诉了小明：这是因为 delete 的时候会加间隙锁。 但是间隙锁之间却可以兼容，但是插入新的数据的时候就会因为插入意向锁会被间隙锁阻塞，导致双方资源被互占，导致死锁。 小明听了之后似懂非懂，由于大红的事情比较多，不方便一直麻烦大红，所以决定自己下来自己想。 下班过后，小明回想大红说的话，什么是间隙锁，什么是插入意向锁，看来作为开发者对数据库不应该只会写 SQL 啊，不然遇到一些疑难杂症完全没法解决啊。想完，于是小明就踏上了学习 MySQL 锁这条不归之路。 <section style="box-sizing: border-box;"> <section class="V5" style="box-sizing: border-box;" powered-by="xiumi.us"> <section style="border-bottom-width: 1px;border-bottom-style: solid;border-bottom-color: black;margin-top: 0.5em;margin-bottom: 0.5em;line-height: 1.2;box-sizing: border-box;"> <section style="display: inline-block;border-bottom-width: 6px;border-bottom-style: solid;border-color: rgb(89, 89, 89);margin-bottom: -1px;font-size: 20px;color: rgb(89, 89, 89);box-sizing: border-box;"> 什么是 InnoDB？ </section> </section> </section> </section> <section style="box-sizing: border-box;"> <section class="V5" style="box-sizing: border-box;" powered-by="xiumi.us"> <section style="margin-top: 10px;margin-bottom: 10px;box-sizing: border-box;"> <section style="width: 0.6em;display: inline-block;vertical-align: middle;box-sizing: border-box;"> </section> <section style="display: inline-block;vertical-align: middle;font-size: 18px;padding-left: 5px;color: rgb(89, 89, 89);box-sizing: border-box;"> MySQL 体系架构 </section> </section> </section> </section> 小明没有着急去了解锁这方面的知识，他首先先了解了下 MySQL 体系架构： <img class="" data-ratio="0.7631103074141049" data-type="png" data-w="553" src="/upload/4289fb1925c4d24660ea00682bbf5b4f.png" style="box-sizing: border-box;border-width: 2px;border-style: solid;border-color: rgb(238, 238, 238);border-radius: 6px;word-wrap: break-word !important;width: 553px !important;visibility: visible !important;"> 可以发现 MySQL 由连接池组件、管理服务和工具组件、SQL 接口组件、查询分析器组件、优化器组件、缓冲组件、插件式存储引擎、物理文件组成。 小明发现在 MySQL 中存储引擎是以插件的方式提供的，在 MySQL 中有多种存储引擎，每个存储引擎都有自己的特点。 随后小明在命令行中打出了： <section class="output_wrapper" style="font-size: 16px;color: rgb(62, 62, 62);line-height: 1.6;letter-spacing: 0px;"> <pre style="font-size: inherit;color: inherit;line-height: inherit;margin-top: 0px;margin-bottom: 0px;padding: 0px;"><code class="hljs sql" style="margin-right: 2px;margin-left: 2px;line-height: 18px;font-size: 14px;letter-spacing: 0px;font-family: Consolas, Inconsolata, Courier, monospace;border-radius: 0px;color: rgb(169, 183, 198);padding: 0.5em;display: block !important;word-wrap: normal !important;word-break: normal !important;overflow: auto !important;background: rgb(40, 43, 46);">show engines \G; </code></pre> </section> 一看原来有这么多种引擎。又打出了下面的命令，查看当前数据库默认的引擎： <section class="output_wrapper" style="font-size: 16px;color: rgb(62, 62, 62);line-height: 1.6;letter-spacing: 0px;"> <pre style="font-size: inherit;color: inherit;line-height: inherit;margin-top: 0px;margin-bottom: 0px;padding: 0px;"><code class="hljs sql" style="margin-right: 2px;margin-left: 2px;line-height: 18px;font-size: 14px;letter-spacing: 0px;font-family: Consolas, Inconsolata, Courier, monospace;border-radius: 0px;color: rgb(169, 183, 198);padding: 0.5em;display: block !important;word-wrap: normal !important;word-break: normal !important;overflow: auto !important;background: rgb(40, 43, 46);">show variables like '%storage_engine%'; </code></pre> </section> <img class="" data-ratio="0.5019011406844106" data-type="png" data-w="789" src="/upload/324b043de0bed32eb5d751a5ff7117c0.png" style="box-sizing: border-box;border-width: 2px;border-style: solid;border-color: rgb(238, 238, 238);border-radius: 6px;word-wrap: break-word !important;width: 677px !important;visibility: visible !important;"> 小明恍然大悟：原来自己的数据库是使用的 InnoDB，依稀记得自己在上学的时候好像听说过有个引擎叫 MyIsAM，小明想这两个有啥不同呢？ 马上查找了一下资料： <img class="" data-copyright="0" data-ratio="0.40700808625336926" data-s="300,640" src="/upload/e82e61bb4bea9604445ffbdde808ecc9.png" data-type="png" data-w="742" style=""> 小明大概了解了一下 InnoDB 和 MyIsAM 的区别，由于使用的是 InnoDB，小明就没有过多的纠结这一块。 <section style="box-sizing: border-box;"> <section class="V5" style="box-sizing: border-box;" powered-by="xiumi.us"> <section style="margin-top: 10px;margin-bottom: 10px;box-sizing: border-box;"> <section style="width: 0.6em;display: inline-block;vertical-align: middle;box-sizing: border-box;"> </section> <section style="display: inline-block;vertical-align: middle;font-size: 18px;padding-left: 5px;color: rgb(89, 89, 89);box-sizing: border-box;"> 事务的隔离性 </section> </section> </section> </section> 小明在研究锁之前，又回想到之前上学的时候教过的数据库事务隔离性，其实锁在数据库中其功能之一也是用来实现事务隔离性。而事务的隔离性其实是用来解决脏读，不可重复读，幻读几类问题。 <h3 style="text-align: justify;margin-left: 8px;margin-right: 8px;line-height: 1.75em;">脏读</h3> 一个事务读取到另一个事务未提交的更新数据。什么意思呢？ <img class="" data-copyright="0" data-ratio="0.363754889178618" data-s="300,640" src="/upload/e1a0ced2b71ffc4cdfb0cb6b1c6ad35f.png" data-type="png" data-w="767" style=""> 在事务 A，B 中，事务 A 在时间点 2，4 分别对 user 表中 id = 1 的数据进行了查询。 但是事务 B 在时间点 3 进行了修改，导致了事务 A 在 4 中的查询出的结果其实是事务 B 修改后的。这样就破坏了数据库中的隔离性。 <h3 style="text-align: justify;margin-left: 8px;margin-right: 8px;line-height: 1.75em;">不可重复读</h3> 在同一个事务中，多次读取同一数据返回的结果不同，不可重复读和脏读不同的是这里读取的是已经提交过后的数据。 <img class="" data-copyright="0" data-ratio="0.4119170984455959" data-s="300,640" src="/upload/ac4b875b709399b42d6fa9e121a6d03a.png" data-type="png" data-w="772" style=""> 在事务 B 中提交的操作在事务 A 第二次查询之前，但是依然读到了事务 B 的更新结果，也破坏了事务的隔离性。 <h3 style="line-height: normal;"> </h3> <h3 style="text-align: justify;margin-left: 8px;margin-right: 8px;line-height: 1.75em;">幻读</h3> 一个事务读到另一个事务已提交的 insert 数据。 <img class="" data-copyright="0" data-ratio="0.4113573407202216" data-s="300,640" src="/upload/cae393aaab0271fb74d94f6bd8eafca7.png" data-type="png" data-w="722" style=""> 在事务 A 中查询了两次 id 大于 1 的，在第一次 id 大于 1 查询结果中没有数据，但是由于事务 B 插入了一条 id = 2 的数据，导致事务 A 第二次查询时能查到事务 B 中插入的数据。 事务中的隔离性： <img class="" data-copyright="0" data-ratio="0.3127690100430416" data-s="300,640" src="/upload/d73343e23099319f5236a86819df5ef8.png" data-type="png" data-w="697" style=""> 小明注意到在收集资料的过程中，有资料写到 InnoDB 和其他数据库有点不同，InnoDB 的可重复读其实就能解决幻读了，小明心想：这 InnoDB 还挺牛逼的，我得好好看看到底是怎么个原理。 <section style="box-sizing: border-box;"> <section class="V5" style="box-sizing: border-box;" powered-by="xiumi.us"> <section style="margin-top: 10px;margin-bottom: 10px;box-sizing: border-box;"> <section style="width: 0.6em;display: inline-block;vertical-align: middle;box-sizing: border-box;"> </section> <section style="display: inline-block;vertical-align: middle;font-size: 18px;padding-left: 5px;color: rgb(89, 89, 89);box-sizing: border-box;"> InnoDB 锁类型 </section> </section> </section> </section> 小明首先了解了 MySQL 中常见的锁类型有哪些： <h3 style="text-align: justify;margin-left: 8px;margin-right: 8px;line-height: 1.75em;">S or X</h3> 在 InnoDB 中实现了两个标准的行级锁，可以简单的看为两个读写锁： <ul class=" list-paddingleft-2" style="list-style-type: square;margin-left: 8px;margin-right: 8px;"> <li>S 共享锁：又叫读锁，其他事务可以继续加共享锁，但是不能继续加排他锁。</li> <li>X 排他锁：又叫写锁，一旦加了写锁之后，其他事务就不能加锁了。</li> </ul> <img class="" data-copyright="0" data-ratio="0.21084337349397592" data-s="300,640" src="/upload/74fb2a185207a645856512187ce0c1dc.png" data-type="png" data-w="664" style="line-height: 25.6px;text-align: center;white-space: normal;"> 兼容性：是指事务 A 获得一个某行某种锁之后，事务 B 同样的在这个行上尝试获取某种锁，如果能立即获取，则称锁兼容，反之叫冲突。 纵轴是代表已有的锁，横轴是代表尝试获取的锁。 <h3 style="text-align: justify;margin-left: 8px;margin-right: 8px;line-height: 1.75em;">意向锁</h3> 意向锁在 InnoDB 中是表级锁，和它的名字一样它是用来表达一个事务想要获取什么。 意向锁分为：</

sysbench 自定义表结构lua脚本

作者：じ☆ve宝贝

my_common.lua ``` -- Copyright (C) 2006-2017 Alexey Kopytov <akopytov@gmail.com> -- This program is free software; you can redistribute it and/or modify -- it under the terms of the GNU General Public License as published by -- the Free Software Foundation; either version 2 of the License, or -- (at your option) any later version. -- This program is distributed in the hope that it will be useful, -- but WITHOUT ANY WARRANTY; without even the implied warranty of -- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -- GNU General Public License for more details. -- You should have received a copy of the GNU General Public License -- along with this program; if not, write to the Free Software -- Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA -- ----------------------------------------------------------------------------- -- Common code for OLTP benchmarks. -- ----------------------------------------------------------------------------- function init() assert(event ~= nil, "this script is meant to be included by other OLTP scripts and " .. "should not be called directly.") end if sysbench.cmdline.command == nil then error("Command is required. Supported commands: prepare, prewarm, run, " .. "cleanup, help") end -- Command line options sysbench.cmdline.options = { table_size = {"Number of rows per table", 10000}, range_size = {"Range size for range SELECT queries", 100}, tables = {"Number of tables", 1}, point_selects = {"Number of point SELECT queries per transaction", 10}, simple_ranges = {"Number of simple range SELECT queries per transaction", 1}, sum_ranges = {"Number of SELECT SUM() queries per transaction", 1}, order_ranges = {"Number of SELECT ORDER BY queries per transaction", 1}, distinct_ranges = {"Number of SELECT DISTINCT queries per transaction", 1}, index_updates = {"Number of UPDATE index queries per transaction", 1}, non_index_updates = {"Number of UPDATE non-index queries per transaction", 1}, delete_inserts = {"Number of DELETE/INSERT combination per transaction", 1}, range_selects = {"Enable/disable all range SELECT queries", true}, auto_inc = {"Use AUTO_INCREMENT column as Primary Key (for MySQL), " .. "or its alternatives in other DBMS. When disabled, use " .. "client-generated IDs", true}, skip_trx = {"Don't start explicit transactions and execute all queries " .. "in the AUTOCOMMIT mode", false}, secondary = {"Use a secondary index in place of the PRIMARY KEY", false}, create_secondary = {"Create a secondary index in addition to the PRIMARY KEY", true}, mysql_storage_engine = {"Storage engine, if MySQL is used", "myisam"}, pgsql_variant = {"Use this PostgreSQL variant when running with the " .. "PostgreSQL driver. The only currently supported " .. "variant is 'redshift'. When enabled, " .. "create_secondary is automatically disabled, and " .. "delete_inserts is set to 0"} } -- Prepare the dataset. This command supports parallel execution, i.e. will -- benefit from executing with --threads > 1 as long as --tables > 1 function cmd_prepare() local drv = sysbench.sql.driver() local con = drv:connect() for i = sysbench.tid % sysbench.opt.threads + 1, sysbench.opt.tables, sysbench.opt.threads do create_table(drv, con, i) end end -- Preload the dataset into the server cache. This command supports parallel -- execution, i.e. will benefit from executing with --threads > 1 as long as -- --tables > 1 -- -- PS. Currently, this command is only meaningful for MySQL/InnoDB benchmarks function cmd_prewarm() local drv = sysbench.sql.driver() local con = drv:connect() assert(drv:name() == "mysql", "prewarm is currently MySQL only") -- Do not create on disk tables for subsequent queries con:query("SET tmp_table_size=2*1024*1024*1024") con:query("SET max_heap_table_size=2*1024*1024*1024") for i = sysbench.tid % sysbench.opt.threads + 1, sysbench.opt.tables, sysbench.opt.threads do local t = "sbtest" .. i print("Prewarming table " .. t) con:query("ANALYZE TABLE sbtest" .. i) con:query(string.format( "SELECT AVG(id) FROM " .. "(SELECT * FROM %s FORCE KEY (PRIMARY) " .. "LIMIT %u) t", t, sysbench.opt.table_size)) con:query(string.format( "SELECT COUNT(*) FROM " .. "(SELECT * FROM %s WHERE cn_name LIKE '%%0%%' LIMIT %u) t", t, sysbench.opt.table_size)) end end -- Implement parallel prepare and prewarm commands sysbench.cmdline.commands = { prepare = {cmd_prepare, sysbench.cmdline.PARALLEL_COMMAND}, prewarm = {cmd_prewarm, sysbench.cmdline.PARALLEL_COMMAND} } -- Template strings of random digits with 11-digit groups separated by dashes -- 10 groups, 119 characters local c_value_template = "###########-###########-###########-" .. "###########-###########-###########-" .. "###########-###########-###########-" .. "###########" -- 5 groups, 59 characters local pad_value_template = "###########-###########-###########-" .. "###########-###########" function get_c_value() return sysbench.rand.string(c_value_template) end function get_pad_value() return sysbench.rand.string(pad_value_template) end function create_table(drv, con, table_num) local id_index_def, id_def local engine_def = "" local extra_table_options = "" local query if sysbench.opt.secondary then id_index_def = "KEY xid" else id_index_def = "PRIMARY KEY" end if drv:name() == "mysql" or drv:name() == "attachsql" or drv:name() == "drizzle" then if sysbench.opt.auto_inc then id_def = "INTEGER NOT NULL AUTO_INCREMENT" else id_def = "INTEGER NOT NULL" end engine_def = "/*! ENGINE = " .. sysbench.opt.mysql_storage_engine .. " */" extra_table_options = mysql_table_options or "" elseif drv:name() == "pgsql" then if not sysbench.opt.auto_inc then id_def = "INTEGER NOT NULL" elseif pgsql_variant == 'redshift' then id_def = "INTEGER IDENTITY(1,1)" else id_def = "SERIAL" end else error("Unsupported database driver:" .. drv:name()) end print(string.format("Creating table 'sbtest%d'...", table_num)) query = string.format([[ CREATE TABLE sbtest%d( `id` char(100) DEFAULT NULL, `cn_name` varchar(200) DEFAULT NULL COMMENT '中文名', `cn_short` varchar(200) DEFAULT NULL COMMENT '中文简称', `en_name` varchar(200) DEFAULT NULL COMMENT '英文名', `en_short` varchar(200) DEFAULT NULL COMMENT '英文简称', `cn_once` varchar(200) DEFAULT NULL COMMENT '曾用名', `credit_code` varchar(200) DEFAULT NULL COMMENT '统一信用代码', `reg_institute` varchar(200) DEFAULT NULL COMMENT '登记机关', `scale` varchar(200) DEFAULT NULL COMMENT '公司规模', `organize_code` varchar(200) DEFAULT NULL COMMENT '企业注册码', `registration_number` varchar(200) DEFAULT NULL COMMENT '工商注册号', `legal_person` varchar(200) DEFAULT NULL COMMENT '法人', `company_type` varchar(200) DEFAULT NULL COMMENT '企业控股: 1：', `register_type` varchar(200) DEFAULT NULL COMMENT '企业注册类型 ', `estiblish_time` varchar(200) DEFAULT NULL COMMENT '营业期限开始', `to_time` varchar(200) DEFAULT NULL COMMENT '营业期限截止', `check_date` varchar(200) DEFAULT NULL COMMENT '发照日期', `area_type` varchar(200) DEFAULT NULL COMMENT '地区', `reg_location` varchar(200) DEFAULT NULL COMMENT '企业地址 ', `business_scope` varchar(200) DEFAULT NULL COMMENT '经营范围', `web_url` varchar(200) DEFAULT NULL COMMENT '网址', `company_logo` varchar(200) DEFAULT NULL COMMENT 'logo', `cn_desc` varchar(200) DEFAULT NULL COMMENT '中文简介', `en_desc` varchar(200) DEFAULT NULL COMMENT '中文简介', `begin_time` varchar(200) DEFAULT NULL COMMENT '成立时间', `stop_time` varchar(200) DEFAULT NULL COMMENT '停业时间', `registered_capital` varchar(200) DEFAULT NULL COMMENT '注册资本', `currency_type` varchar(200) DEFAULT NULL COMMENT '注册资本货币类型 ', `commerce_status` varchar(200) DEFAULT NULL COMMENT '商业状态', `remark` varchar(510) DEFAULT NULL, `creater_id` varchar(200) DEFAULT NULL COMMENT '修改人', `create_time` varchar(200) DEFAULT NULL COMMENT '创建时间', `status` varchar(200) DEFAULT NULL COMMENT '1：正常 2删除 3...', `modifier_id` varchar(200) DEFAULT NULL COMMENT '修改人', `modify_time` varchar(200) DEFAULT NULL COMMENT '修改时间', PRIMARY KEY (id) ) %s %s]], table_num, engine_def, extra_table_options) con:query(query) if (sysbench.opt.table_size > 0) then print(string.format("Inserting %d records into 'sbtest%d'", sysbench.opt.table_size, table_num)) end if sysbench.opt.auto_inc then query = "INSERT INTO sbtest" .. table_num .. "(id, cn_name, cn_short, en_name, en_short, cn_once, credit_code, reg_institute, scale, organize_code, registration_number, legal_person, company_type, register_type, estiblish_time, to_time, check_date, area_type, reg_location, business_scope, web_url, company_logo, cn_desc, en_desc, begin_time, stop_time, registered_capital, currency_type, commerce_status, remark, creater_id, create_time, status, modifier_id, modify_time) VALUES" else query = "INSERT INTO sbtest" .. table_num .. "(id, cn_name, cn_short, en_name, en_short, cn_once, credit_code, reg_institute, scale, organize_code, registration_number, legal_person, company_type, register_type, estiblish_time, to_time, check_date, area_type, reg_location, business_scope, web_url, company_logo, cn_desc, en_desc, begin_time, stop_time, registered_capital, currency_type, commerce_status, remark, creater_id, create_time, status, modifier_id, modify_time) VALUES" end con:bulk_insert_init(query) local c_val local pad_val for i = 1, sysbench.opt.table_size do c_val = get_c_value() pad_val = get_pad_value() if (sysbench.opt.auto_inc) then query = string.format("('%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s')", guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),get_pad_value(),get_pad_value(),getTimeStamp(), sb_rand(1, sysbench.opt.table_size), c_val,get_pad_value(),getTimeStamp(),get_pad_value(),getTimeStamp(),get_pad_value(),getTimeStamp(), pad_val,guid(),get_pad_value(),get_pad_value(),getTimeStamp(), sb_rand(1, sysbench.opt.table_size),guid(),get_pad_value(),get_pad_value(),getTimeStamp(), sb_rand(1, sysbench.opt.table_size),get_pad_value(),get_pad_value(),getTimeStamp()) else query = string.format("('%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s')", guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),guid(),get_pad_value(),get_pad_value(),getTimeStamp(), sb_rand(1, sysbench.opt.table_size), c_val,get_pad_value(),getTimeStamp(),get_pad_value(),getTimeStamp(),get_pad_value(),getTimeStamp(), pad_val,guid(),get_pad_value(),get_pad_value(),getTimeStamp(), sb_rand(1, sysbench.opt.table_size),guid(),get_pad_value(),get_pad_value(),getTimeStamp(), sb_rand(1, sysbench.opt.table_size),get_pad_value(),get_pad_value(),getTimeStamp()) end con:bulk_insert_next(query) end con:bulk_insert_done() if sysbench.opt.create_secondary then print(string.format("Creating a secondary index on 'sbtest%d'...", table_num)) con:query(string.format("CREATE INDEX k_%d ON sbtest%d(cn_name)", table_num, table_num)) end end local t = sysbench.sql.type local stmt_defs = { point_selects = { "SELECT cn_name FROM sbtest%u WHERE estiblish_time=?", {t.VARCHAR, 100}}, simple_ranges = { "SELECT cn_name FROM sbtest%u WHERE estiblish_time BETWEEN ? AND ?", {t.VARCHAR, 100}, {t.VARCHAR, 100}}, sum_ranges = { "SELECT SUM(cn_short) FROM sbtest%u WHERE estiblish_time BETWEEN ? AND ?", {t.VARCHAR, 100}, {t.VARCHAR, 100}}, order_ranges = { "SELECT cn_name FROM sbtest%u WHERE estiblish_time BETWEEN ? AND ? ORDER BY cn_name", {t.VARCHAR, 100}, {t.VARCHAR, 100}}, distinct_ranges = { "SELECT DISTINCT cn_name FROM sbtest%u WHERE estiblish_time BETWEEN ? AND ? ORDER BY cn_name", {t.VARCHAR, 100}, {t.VARCHAR, 100}}, index_updates = { "UPDATE sbtest%u SET en_name='test' WHERE estiblish_time=?", {t.VARCHAR, 100}}, non_index_updates = { "UPDATE sbtest%u SET cn_name='non_index_update' WHERE estiblish_time=?", {t.VARCHAR, 100}}, deletes = { "DELETE FROM sbtest%u WHERE estiblish_time=?", {t.VARCHAR, 100}}, inserts = { "INSERT INTO sbtest%u (id, cn_name, cn_short, en_name, estiblish_time) VALUES (?, ?, ?, ?, ?)", {t.VARCHAR, 100}, {t.VARCHAR, 200}, {t.VARCHAR, 200}, {t.VARCHAR, 200}, {t.VARCHAR, 200}}, } function prepare_begin() stmt.begin = con:prepare("BEGIN") end function prepare_commit() stmt.commit = con:prepare("COMMIT") end function prepare_for_each_table(key) for t = 1, sysbench.opt.tables do stmt[t][key] = con:prepare(string.format(stmt_defs[key][1], t)) -- print(string.format(stmt_defs[key][1], t)) local nparam = #stmt_defs[key] - 1 if nparam > 0 then param[t][key] = {} end -- print(string.format('----byte,typeby=%s,len=%s,key=%s',type(btype),len,key)) for p = 1, nparam do local btype = stmt_defs[key][p+1] local len if type(btype) == "table" then len = btype[2] btype = btype[1] end -- print(string.format('%s,%s---', btype,sysbench.sql.type.VARCHAR)) if btype == sysbench.sql.type.VARCHAR or btype == sysbench.sql.type.CHAR then -- print(string.format('----cahr,typeby=%s,key=%s,btype=%s,len=%s',type(btype),key,btype,len)) param[t][key][p] = stmt[t][key]:bind_create(btype, len) else -- print(string.format('----nochar,typeby=%s,key=%s,btype=%s',type(btype),key,btype)) param[t][key][p] = stmt[t][key]:bind_create(btype) end end if nparam > 0 then stmt[t][key]:bind_param(unpack(param[t][key])) end end end function prepare_point_selects() prepare_for_each_table("point_selects") end function prepare_simple_ranges() prepare_for_each_table("simple_ranges") end function prepare_sum_ranges() prepare_for_each_table("sum_ranges") end function prepare_order_ranges() prepare_for_each_table("order_ranges") end function prepare_distinct_ranges() prepare_for_each_table("distinct_ranges") end function prepare_index_updates() prepare_for_each_table("index_updates") end function prepare_non_index_updates() prepare_for_each_table("non_index_updates") end function prepare_delete_inserts() prepare_for_each_table("deletes") prepare_for_each_table("inserts") end function thread_init() drv = sysbench.sql.driver() con = drv:connect() -- Create global nested tables for prepared statements and their -- parameters. We need a statement and a parameter set for each combination -- of connection/table/query stmt = {} param = {} for t = 1, sysbench.opt.tables do stmt[t] = {} param[t] = {} end -- This function is a 'callback' defined by individual benchmark scripts prepare_statements() end function thread_done() -- Close prepared statements for t = 1, sysbench.opt.tables do for k, s in pairs(stmt[t]) do stmt[t][k]:close() end end if (stmt.begin ~= nil) then stmt.begin:close() end if (stmt.commit ~= nil) then stmt.commit:close() end con:disconnect() end function cleanup() local drv = sysbench.sql.driver() local con = drv:connect() for i = 1, sysbench.opt.tables do print(string.format("Dropping table 'sbtest%d'...", i)) con:query("DROP TABLE IF EXISTS sbtest" .. i ) end end local function get_table_num() return sysbench.rand.uniform(1, sysbench.opt.tables) end local function get_id() return string.format(sysbench.rand.default(1, sysbench.opt.table_size)) end function begin() for t = 1, sysbench.opt.tables do for k, s in pairs(stmt[t]) do -- print(k,s) end end stmt.begin:execute() end function commit() stmt.commit:execute() end function execute_point_selects() local tnum = get_table_num() local i for i = 1, sysbench.opt.point_selects do param[tnum].point_selects[1]:set(get_id()) stmt[tnum].point_selects:execute() end end local function execute_range(key) local tnum = get_table_num() for i = 1, sysbench.opt[key] do local id = get_id() param[tnum][key][1]:set(id) param[tnum][key][2]:set(guid()) stmt[tnum][key]:execute() end end function execute_simple_ranges() execute_range("simple_ranges") end function execute_sum_ranges() execute_range("sum_ranges") end function execute_order_ranges() execute_range("order_ranges") end function execute_distinct_ranges() execute_range("distinct_ranges") end function execute_index_updates() local tnum = get_table_num() for i = 1, sysbench.opt.index_updates do param[tnum].index_updates[1]:set(get_id()) stmt[tnum].index_updates:execute() end end function execute_non_index_updates() local tnum = get_table_num() for i = 1, sysbench.opt.non_index_updates do -- param[tnum].non_index_updates[1]:set_rand_str(c_value_template) -- param[tnum].non_index_updates[2]:set(guid()) param[tnum].non_index_updates[1]:set(get_id()) stmt[tnum].non_index_updates:execute() end end function execute_delete_inserts() local tnum = get_table_num() for i = 1, sysbench.opt.delete_inserts do -- print(i) local id = get_id() local k = get_id() local uid = get_id(); param[tnum].deletes[1]:set(uid) param[tnum].inserts[1]:set(uid) param[tnum].inserts[2]:set(k) param[tnum].inserts[3]:set_rand_str(c_value_template) param[tnum].inserts[4]:set_rand_str(pad_value_template) param[tnum].inserts[5]:set(uid) stmt[tnum].deletes:execute() stmt[tnum].inserts:execute() end end function guid() local seed = { '0','1','2','3','4','5','6','7','8','9', 'a','b','c','d','e','f','g','h','i','j', 'k','l','m','n','o','p','q','r','s','t', 'u','v','w','x','y','z' } local sid = "" for i=1, 32 do -- math.randomseed(tostring(os.time()):reverse():sub(1, 6)) sid = sid .. seed[random()] end return string.format('%s%s%s%s%s', string.sub(sid, 1, 8), string.sub(sid, 9, 12), string.sub(sid, 13, 16), string.sub(sid, 17, 20), string.sub(sid, 21, 32) ) end function random() local ret=0 math.randomseed(os.time()) for i=1,3 do n = math.random(1,36) ret=n end return ret end function getTimeStamp() local t = os.time(); return os.date("%Y-%m-%d %H:%M:%S",t/1000) end ``` ## my_read_write.lua ``` #!/usr/bin/env sysbench -- Copyright (C) 2006-2017 Alexey Kopytov <akopytov@gmail.com> -- This program is free software; you can redistribute it and/or modify -- it under the terms of the GNU General Public License as published by -- the Free Software Foundation; either version 2 of the License, or -- (at your option) any later version. -- This program is distributed in the hope that it will be useful, -- but WITHOUT ANY WARRANTY; without even the implied warranty of -- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -- GNU General Public License for more details. -- You should have received a copy of the GNU General Public License -- along with this program; if not, write to the Free Software -- Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA -- ---------------------------------------------------------------------- -- Read/Write OLTP benchmark -- ---------------------------------------------------------------------- require("my_common") function prepare_statements() if not sysbench.opt.skip_trx then prepare_begin() prepare_commit() end prepare_point_selects() if sysbench.opt.range_selects then prepare_simple_ranges() prepare_sum_ranges() prepare_order_ranges() prepare_distinct_ranges() end prepare_index_updates() prepare_non_index_updates() prepare_delete_inserts() end function event() if not sysbench.opt.skip_trx then begin() end execute_point_selects() if sysbench.opt.range_selects then execute_simple_ranges() execute_sum_ranges() execute_order_ranges() execute_distinct_ranges() end execute_index_updates() execute_non_index_updates() execute_delete_inserts() if not sysbench.opt.skip_trx then commit() end end ``` ## company_read_write.sh 执行脚本 ``` #!/bin/bash read -t 30 -p "请输入运行方式(prepare/run/cleanup)": status sysbench /data/lua/my_read_write.lua \ --mysql-host=127.0.0.1 \ --mysql-port=3306 \ --mysql-user=root \ --mysql-password='V5kNP47u28450Gpt' \ --mysql-db=sbtest \ --db-driver=mysql \ --tables=1 \ --table-size=1000000 \ --report-interval=10 \ --threads=128 \ --time=120 \ $status ```

揭开Java 泛型类型擦除神秘面纱

作者：微信小助手

<section class="xmteditor" style="display:none;" data-tools="新媒体管家" data-label="powered by xmt.cn"></section> <section data-role="outer" label="Powered by 135editor.com" data-mpa-powered-by="yiban.io"> <section class="_135editor" data-tools="135编辑器" data-id="91525"> <section> <section> 泛型，一个孤独的守门者。 </section> </section> </section> </section> 大家可能会有疑问，我为什么叫做泛型是一个守门者。这其实是我个人的看法而已，我的意思是说泛型没有其看起来那么深不可测，它并不神秘与神奇。泛型是 Java 中一个很小巧的概念，但同时也是一个很容易让人迷惑的知识点，它让人迷惑的地方在于它的许多表现有点违反直觉。 文章开始的地方，先给大家奉上一道经典的测试题。 <section data-mpa-preserve-tpl-color="t" data-mpa-template="t" class="mpa-template" mpa-preserve="t"> <pre style="margin:0;padding:0;border-radius:none;background:none;"><code style="border-radius: 4px;font-size: 0.85em;margin: 0px 0.15em;background: rgb(40, 44, 52);color: rgb(171, 178, 191);display: block;padding: 6px;overflow-x: auto;white-space: nowrap;" class="hljs-default">List<String> l1 = new ArrayList<String>(); List<Integer> l2 = new ArrayList<Integer>(); System.out.println(l1.getClass() == l2.getClass());</code></pre> </section> 请问，上面代码最终结果输出的是什么？不了解泛型的和很熟悉泛型的同学应该能够答出来，而对泛型有所了解，但是了解不深入的同学可能会答错。 正确答案是 true。 上面的代码中涉及到了泛型，而输出的结果缘由是类型擦除。先好好说说泛型。 <h1 style="outline: 0px;margin-top: 8px;margin-bottom: 16px;font-size: 28px;color: rgb(79, 79, 79);font-weight: 700;line-height: 36px;word-break: break-all;">泛型是什么？</h1> 泛型的英文是 generics，generic 的意思是通用,而翻译成中文，泛应该意为广泛，型是类型。所以泛型就是能广泛适用的类型。 但泛型还有一种较为准确的说法就是为了参数化类型，或者说可以将类型当作参数传递给一个类或者是方法。 那么，如何解释类型参数化呢？ <section data-mpa-preserve-tpl-color="t" data-mpa-template="t" class="mpa-template" mpa-preserve="t"> <pre style="margin:0;padding:0;border-radius:none;background:none;"><code style="border-radius: 4px;font-size: 0.85em;margin: 0px 0.15em;background: rgb(40, 44, 52);color: rgb(171, 178, 191);display: block;padding: 6px;overflow-x: auto;white-space: nowrap;" class="hljs-default">public class Cache {    Object value;    public Object getValue() {        return value;    }    public void setValue(Object value) {        this.value = value;    } }</code></pre> </section> 假设 Cache 能够存取任何类型的值，于是，我们可以这样使用它。 <section data-mpa-preserve-tpl-color="t" data-mpa-template="t" class="mpa-template" mpa-preserve="t"> <pre style="margin:0;padding:0;border-radius:none;background:none;"><code style="border-radius: 4px;font-size: 0.85em;margin: 0px 0.15em;background: rgb(40, 44, 52);color: rgb(171, 178, 191);display: block;padding: 6px;overflow-x: auto;white-space: nowrap;" class="hljs-default">Cache cache = new Cache(); cache.setValue(134); int value = (int) cache.getValue(); cache.setValue("hello"); String value1 = (String) cache.getValue();</code></pre> </section> 使用的方法也很简单，只要我们做正确的强制转换就好了。 但是，泛型却给我们带来了不一样的编程体验。 <section data-mpa-preserve-tpl-color="t" data-mpa-template="t" class="mpa-template" mpa-preserve="t"> <pre style="margin:0;padding:0;border-radius:none;background:none;"><code style="border-radius: 4px;font-size: 0.85em;margin: 0px 0.15em;background: rgb(40, 44, 52);color: rgb(171, 178, 191);display: block;padding: 6px;overflow-x: auto;white-space: nowrap;" class="hljs-default">public class Cache<T> {    T value;    public Object getValue() {        return value;    }    public void setValue(T value) {        this.value = value;    } }</code></pre> </section> 这就是泛型，它将 value 这个属性的类型也参数化了，这就是所谓的参数化类型。再看它的使用方法。 <section data-mpa-preserve-tpl-color="t" data-mpa-template="t" class="mpa-template" mpa-preserve="t"> <pre style="margin:0;padding:0;border-radius:none;background:none;"><code style="border-radius: 4px;font-size: 0.85em;margin: 0px 0.15em;background: rgb(40, 44, 52);color: rgb(171, 178, 191);display: block;padding: 6px;overflow-x: auto;white-space: nowrap;" class="hljs-default">Cache<String> cache1 = new Cache<String>(); cache1.setValue("123"); String value2 = cache1.getValue(); Cache<Integer> cache2 = new Cache<Integer>(); cache2.setValue(456); int value3 = cache2.getValue();</code></pre> </section> 最显而易见的好处就是它不再需要对取出来的结果进行强制转换了。但，还有另外一点不同。  <img class="" data-copyright="0" data-ratio="0.2890932982917214" data-s="300,640" data-type="png" data-w="761" src="/upload/f928dd3d710e77c19

一遍记住 8 种排序算法与 Java 代码实现

作者：微信小助手

<span class="" style="font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;letter-spacing: 0.544px;max-width: 100%;white-space: pre-wrap;font-size: 14px;color: rgb(255, 41, 65);line-height: 22.4px;box-sizing: border-box !important;overflow-wrap: break-word !important;">（给<span style="font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;letter-spacing: 0.544px;max-width: 100%;white-space: pre-wrap;font-size: 14px;line-height: 22.4px;color: rgb(0, 128, 255);box-sizing: border-box !important;overflow-wrap: break-word !important;">算法爱好者<span class="" style="font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;letter-spacing: 0.544px;max-width: 100%;white-space: pre-wrap;font-size: 14px;color: rgb(255, 41, 65);line-height: 22.4px;box-sizing: border-box !important;overflow-wrap: break-word !important;">加星标，修炼编程内功<span class="" style="font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;letter-spacing: 0.544px;max-width: 100%;white-space: pre-wrap;color: rgb(255, 41, 65);font-size: 14px;line-height: 22.4px;box-sizing: border-box !important;overflow-wrap: break-word !important;">） <blockquote style="white-space: normal;font-variant-ligatures: normal;orphans: 2;widows: 2;"> <p style="max-width: 100%;min-height: 1em;color: rgba(0, 0, 0, 0.5);font-size: 15px;white-space: normal;letter-spacing: 0.544px;background-color: rgb(255, 255, 255);text-align: left;box-sizing: border-box !important;overflow-wrap: break-word !important;">作者：KaelQ， <p style="max-width: 100%;min-height: 1em;color: rgba(0, 0, 0, 0.5);font-size: 15px;white-space: normal;letter-spacing: 0.544px;background-color: rgb(255, 255, 255);text-align: left;box-sizing: border-box !important;overflow-wrap: break-word !important;">www.jianshu.com/p/5e171281a387 </blockquote> 1.直接插入排序 <blockquote class="" data-type="2" data-url="" data-author-name="" data-content-utf8-length="31" data-source-title="" style="color: rgba(0, 0, 0, 0.5);max-width: 100%;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);box-sizing: border-box !important;overflow-wrap: break-word !important;"> <section class="" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"> 经常碰到这样一类排序问题：把新的数据插入到已经排好的数据列中。 </section> </blockquote> <ol class=" list-paddingleft-2" style=""> <li>将第一个数和第二个数排序，然后构成一个有序序列</li> <li>将第三个数插入进去，构成一个新的有序序列。</li> <li>对第四个数、第五个数……直到最后一个数，重复第二步。</li> </ol> <p style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);text-align: center;box-sizing: border-box !important;overflow-wrap: break-word !important;"><img class="rich_pages " data-ratio="1.033282904689864" data-s="300,640" data-type="jpeg" data-w="661" src="/upload/682a4375bf02696564c83e8ab49373a3.jpg" style="box-sizing: border-box !important;overflow-wrap: break-word !important;width: 661px !important;visibility: visible !important;"> 如何写写成代码： <ol class=" list-paddingleft-2" style=""> <li>首先设定插入次数，即循环次数，for(int i=1;i<length;i++)，1个数的那次不用插入。</li> <li>设定插入数和得到已经排好序列的最后一个数的位数。insertNum和j=i-1。</li> <li>从最后一个数开始向前循环，如果插入数小于当前数，就将当前数向后移动一位。</li> <li>将当前数放置到空着的位置，即j+1。</li> </ol> 代码实现如下： <pre style="max-width: 100%;letter-spacing: 0.544px;background-color: rgb(255, 255, 255);overflow-x: auto;box-sizing: border-box !important;overflow-wrap: break-word !important;"><code class="" style="margin-right: 0.15em;margin-left: 0.15em;padding: 0.5em;max-width: 100%;font-size: 0.85em;font-family: Consolas, Menlo, Courier, monospace;overflow: auto;color: rgb(171, 178, 191);text-size-adjust: none;min-width: 400px;background: none 0% 0% / auto repeat scroll padding-box border-box rgb(40, 44, 52);box-sizing: border-box !important;overflow-wrap: break-word !important;display: block !important;">public void insertSort(int[] a){ int length=a.length;//数组长度，将这个提取出来是为了提高速度。 int insertNum;//要插入的数 for(int i=1;i<length;i++){//插入的次数 insertNum=a[i];//要插入的数 int j=i-1;//已经排序好的序列元素个数 while(j>=0&&a[j]>insertNum){//序列从后到前循环，将大于insertNum的数向后移动一格 a[j+1]=a[j];//元素移动一格 j--; } a[j+1]=insertNum;//将需要插入的数放在要插入的位置。 } }</code></pre> 2.希尔排序 <blockquote class="" data-type="2" data-url="" data-author-name="" data-content-utf8-length="18" data-source-title="" style="color: rgba(0, 0, 0, 0.5);max-width: 100%;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);box-sizing: border-box !important;overflow-wrap: break-word !important;"> <section class="" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"> 对于直接插入排序问题，数据量巨大时。 </section> </blockquote> <ol class=" list-paddingleft-2" style=""> <li>将数的个数设为n，取奇数k=n/2，将下标差值为k的书分为一组，构成有序序列。</li> <li>再取k=k/2 ，将下标差值为k的书分为一组，构成有序序列。</li> <li>重复第二步，直到k=1执行简单插入排序。</li> </ol> <p style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);text-align: center;box-sizing: border-box !important;overflow-wrap: break-word !important;"><img class="rich_pages " data-ratio="1.332618025751073" data-s="300,640" data-type="jpeg" data-w="466" src="/upload/22c91e3f1857928356b0b6f700f48b3d.jpg" style="box-sizing: border-box !important;overflow-wrap: break-word !important;width: 466px !important;visibility: visible !important;"> 如何写成代码： <ol class=" list-paddingleft-2" style=""> <li>首先确定分的组数。</li> <li>然后对组中元素进行插入排序。</li> <li>然后将length/2，重复1,2步，直到length=0为止。</li> </ol> 代码实现如下： <pre style="max-width: 100%;letter-spacing: 0.544px;background-color: rgb(255, 255, 255);overflow-x: auto;box-sizing: border-box !important;overflow-wrap: break-word !important;"><code class="" style="margin-right: 0.15em;margin-left: 0.15em;padding: 0.5em;max-width: 100%;font-size: 0.85em;font-family: Consolas, Menlo, Courier, monospace;overflow: auto;color: rgb(171, 178, 191);text-size-adjust: none;min-width: 400px;background: none 0% 0% / auto repeat scroll padding-box border-box rgb(40, 44, 52);box-sizing: border-box !important;overflow-wrap: break-word !important;display: block !important;">public void sheelSort(int[] a){ int d = a.length; while (d!=0) { d=d/2; for (int x = 0; x < d; x++) {//分的组数 for (int i = x + d; i < a.length; i += d) {//组中的元素，从第二个数开始 int j = i - d;//j为有序序列最后一位的位数 int temp = a[i];//要插入的元素 for (; j >= 0 && temp < a[j]; j -= d) {//从后往前遍历。 a[j + d] = a[j];//向后移动d位 } a[j + d] = temp; } } } }</code></pre> 3.简单选择排序 <blockquote class="" data-type="2" data-url="" data-author-name="" data-content-utf8-length="17" data-source-title="" style="color: rgba(0, 0, 0, 0.5);max-width: 100%;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);box-sizing: border-box !important;overflow-wrap: break-word !important;"> <section class="" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"> 常用于取序列中最大最小的几个数时。 </section> </blockquote> (如果每次比较都交换，那么就是交换排序；如果每次比较完一个循环再交换，就是简单选择排序。) <ol class=" list-paddingleft-2" style=""> <li>遍历整个序列，将最小的数放在最前面。</li> <li>遍历剩下的序列，将最小的数放在最前面。</li> <li>重复第二步，直到只剩下一个数。</li> </ol> <p style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);text-align: center;box-sizing: border-box !important;overflow-wrap: break-word !important;"><img class="rich_pages " data-ratio="1.0056497175141244" data-s="300,640" data-type="jpeg" data-w="531" src="/upload/f2dc18a8e562698d53eba1ac06b6bb37.jpg" style="box-sizing: border-box !important;overflow-wrap: break-word !important;width: 531px !important;visibility: visible !important;"> 如何写成代码： <ol class=" list-paddingleft-2" style=""> <li>首先确定循环次数，并且记住当前数字和当前位置。</li> <li>将当前位置后面所有的数与当前数字进行对比，小数赋值给key，并记住小数的位置。</li> <li>比对完成后，将最小的值与第一个数的值交换。</li> <li>重复2、3步。</li> </ol> 代码实现如下： <pre style="max-width: 100%;letter-spacing: 0.544px;background-color: rgb(255, 255, 255);overflow-x: auto;box-sizing: border-box !important;overflow-wrap: break-word !important;"><code class="" style="margin-right: 0.15em;margin-left: 0.15em;padding: 0.5em;max-width: 100%;font-size: 0.85em;font-family: Consolas, Menlo, Courier, monospace;overflow: auto;color: rgb(171, 178, 191);text-size-adjust: none;min-width: 400px;background: none 0% 0% / auto repeat scroll padding-box border-box rgb(40, 44, 52);box-sizing: border-box !important;overflow-wrap: break-word !important;display: block !important;">public void selectSort(int[] a) { int length = a.length; for (int i = 0; i < length; i++) {//循环次数 int key = a[i]; int position=i; for (int j = i + 1; j < length; j++) {//选出最小的值和位置 if (a[j] < key) { key = a[j]; position = j; } } a[position]=a[i];//交换位置 a[i]=key; } }</code></pre> 4.堆排序 <blockquote class="" data-type="2" data-url="" data-author-name="" data-content-utf8-length="11" data-source-title="" style="color: rgba(0, 0, 0, 0.5);max-width: 100%;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);box-sizing: border-box !important;overflow-wrap: break-word !important;"> <section class="" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"> 对简单选择排序的优化。 </section> </blockquote> <ol class=" list-paddingleft-2" style=""> <li>将序列构建成大顶堆。</li> <li>将根节点与最后一个节点交换，然后断开最后一个节点。</li> <li>重复第一、二步，直到所有节点断开。</li> </ol> <p style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);text-align: center;box-sizing: border-box !important;overflow-wrap: break-word !important;"><img class="rich_pages " data-ratio="0.7598752598752598" data-s="300,640" data-type="jpeg" data-w="962" src="/upload/835d5978cf7becbdfcc32ca5f2e5c27.jpg" style="box-sizing: border-box !important;overflow-wrap: break-word !important;width: 677px !important;visibility: visible !important;"> 代码实现如下： <pre style="max-width: 100%;letter-spacing: 0.544px;background-color: rgb(255, 255, 255);overflow-x: auto;box-sizing: border-box !important;overflow-wrap: break-word !important;"><code class="" style="margin-right: 0.15em;margin-left: 0.15em;padding: 0.5em;max-width: 100%;font-size: 0.85em;font-family: Consolas, Menlo, Courier, monospace;overflow: auto;color: rgb(171, 178, 191);text-size-adjust: none;min-width: 400px;background: none 0% 0% / auto repeat scroll padding-box border-box rgb(40, 44, 52);box-sizing: border-box !important;overflow-wrap: break-word !important;display: block !important;">public void heapSort(int[] a){ System.out.println("开始排序"); int arrayLength=a.length; //循环建堆 for(int i=0;i<arrayLength-1;i++){ //建堆 buildMaxHeap(a,arrayLength-1-i); //交换堆顶和最后一个元素 swap(a,0,arrayLength-1-i); System.out.println(Arrays.toString(a)); } } private void swap(int[] data, int i, int j) { // TODO Auto-generated method stub int tmp=data[i]; data[i]=data[j]; data[j]=tmp; } //对data数组从0到lastIndex建大顶堆 private void buildMaxHeap(int[] data, int lastIndex) { // TODO Auto-generated method stub //从lastIndex处节点（最后一个节点）的父节点开始 for(int i=(lastIndex-1)/2;i>=0;i--){ //k保存正在判断的节点 int k=i; //如果当前k节点的子节点存在 while(k*2+1<=lastIndex){ //k节点的左子节点的索引 int biggerIndex=2*k+1; //如果biggerIndex小于lastIndex，即biggerIndex+1代表的k节点的右子节点存在 if(biggerIndex<lastIndex){ //若果右子节点的值较大 if(data[biggerIndex]<data[biggerIndex+1]){ //biggerIndex总是记录较大子节点的索引 biggerIndex++; } } //如果k节点的值小于其较大的子节点的值 if(data[k]<data[biggerIndex]){ //交换他们 swap(data,k,biggerIndex); //将biggerIndex赋予k，开始while循环的下一次循环，重新保证k节点的值大于其左右子节点的值 k=biggerIndex; }else{ break; } } } }</code></pre> <p style="max-width: 100%;min-height: 1em;letter-spacing: 0.544px;white-space: normal;background-color: rgb(255, 255, 255);box-sizing: border-box !important;overflow

文章列表