向东博客 专注WEB应用 构架之美 --- 构架之美，在于尽态极妍 | 应用之美，在于药到病除|赢在IT，Playin' with IT,Focus on Killer Application,Marketing Meets Technology.

执行命令：sudo su -
source ~/.zshrc
compinit:503: no such file or directory: /usr/local/share/zsh/site-functions/_brew_cask

出现：compinit:503: no such file or directory: /usr/local/share/zsh/site-functions/_brew_cask
解决办法：
vim /usr/local/share/zsh/site-functions/_brew_cask

#compdef brew-cask #autoload # Zsh Autocompletion script for Homebrew Cask #   https://github.com/homebrew/brew # Authors: #   Patrick Stadler (https://github.com/pstadler) #   Josh McKinney (https://github.com/joshka) # only display the main commands (but enable completing aliases like 'ls') zstyle -T ':completion:*:*:*:brew-cask:*' tag-order && \   zstyle ':completion:*:*:*:brew-cask:*' tag-order 'commands' __brew_all_casks() {   local -a list   local expl   local comp_cachename=brew_casks   if ! _retrieve_cache $comp_cachename; then     list=( $(brew search --cask) )     _store_cache $comp_cachename list   fi   _wanted list expl 'all casks' compadd -a list } __brew_installed_casks() {   local -a list   local expl   list=( $(brew list --cask) )   _wanted list expl 'installed casks' compadd -a list } __brew_cask_commands() {   local -a commands   commands=(     'audit:verifies installability of Casks'     'cat:dump raw source of the given Cask to the standard output'     'create:creates the given Cask and opens it in an editor'     'doctor:checks for configuration issues'     'edit:edits the given Cask'     'fetch:downloads remote application files to local cache'     'home:opens the homepage of the given Cask'     'info:displays information about the given Cask'     'install:installs the given Cask'     'list:with no args, lists installed Casks; given installed Casks, lists staged files'     'outdated:list the outdated installed Casks'     'reinstall:reinstalls the given Cask'     'style:checks Cask style using RuboCop'     'uninstall:uninstalls the given Cask'     'upgrade:upgrade installed Casks with newer versions'     'zap:zaps all files associated with the given Cask'   )   _describe -t commands "brew cask command" commands } __brew_cask_aliases() {   local -a aliases   aliases=(     'dr'     'homepage'     'abv'     'ls'     '-S'     'rm'     'remove'   )   _describe -t commands "brew cask command aliases" aliases } __brew_cask_command() {   local command="$1"   local completion_func="_brew_cask_${command//-/_}"   declare -f "$completion_func" >/dev/null && "$completion_func" && return } _brew_cask_abv() {   _brew_cask_info } _brew_cask_audit() {   __brew_all_casks } _brew_cask_cat() {   __brew_all_casks } _brew_cask_create() {   _arguments '*::token:' } _brew_cask_edit() {   __brew_all_casks } _brew_cask_fetch() {   _arguments : \     '--force:force re-download even if the files are already cached' \     '*::token:__brew_all_casks' } _brew_cask_home() {   __brew_all_casks } _brew_cask_homepage() {   __brew_cask_home } _brew_cask_info() {   __brew_all_casks } _brew_cask_install() {   _arguments : \     '--force:re-install even if the Cask appears to be already present' \     '--skip-cask-deps:skip any Cask dependencies' \     '--require-sha:abort installation if the Cask does not have a checksum defined' \     '*::token:__brew_all_casks' } _brew_cask_list() {   _arguments : \     '-1[format output in a single column]' \     '-l[format as detailed list]' \     '*::token:__brew_installed_casks' } _brew_cask_outdated() {   _arguments : \     '--greedy:also list Casks with auto_updates or version \:latest' \     '*::token:__brew_installed_casks' } _brew_cask_remove() {   _brew_cask_uninstall } _brew_cask_rm() {   _brew_cask_uninstall } _brew_cask_style() {   _arguments : \     '--fix:auto-correct any style errors if possible' \     '*::token:__brew_all_casks' } _brew_cask_uninstall() {   _arguments : \     '--force:uninstall even if the Cask does not appear to be present' \     '*::token:__brew_installed_casks' } _brew_cask_upgrade() {   _arguments : \     '--force:upgrade even if Cask is not present, and --force the install' \     '--greedy:also upgrade Casks with auto_updates or version \:latest' \     '*::token:__brew_installed_casks' } _brew_cask_zap() {   __brew_all_casks } _brew_cask() {   local curcontext="$curcontext" state state_descr line   typeset -A opt_args   _arguments -C : \     '--verbose:Give additional feedback during installation.' \     '--appdir=-:Target location for Applications. The default value is /Applications:' \     '--colorpickerdir=-:Target location for Color Pickers. The default value is ~/Library/ColorPickers.' \     '--prefpanedir=-:Target location for Preference Panes. The default value is ~/Library/PreferencePanes.' \     '--qlplugindir=-:Target location for QuickLook Plugins. The default value is ~/Library/QuickLook.' \     '--dictionarydir=-:Target location for Dictionaries. The default value is ~/Library/Dictionaries.' \     '--fontdir=-:Target location for Fonts. The default value is ~/Library/Fonts.' \     '--servicedir=-:Target location for Services. The default value is ~/Library/Services.' \     '--input_methoddir=-:Target location for Input Methods. The default value is ~/Library/Input Methods.' \     '--internet_plugindir=-:Target location for Internet Plugins. The default value is ~/Library/Internet Plug-Ins.' \     '--audio_unit_plugindir=-:Target location for Audio Unit Plugins. The default value is ~/Library/Audio/Plug-Ins/Components.' \     '--vst_plugindir=-:Target location for VST Plugins. The default value is ~/Library/Audio/Plug-Ins/VST.' \     '--vst3_plugindir=-:Target location for VST3 Plugins. The default value is ~/Library/Audio/Plug-Ins/VST3.' \     '--screen_saverdir=-:Target location for Screen Savers. The default value is ~/Library/Screen Savers.' \     '--no-binaries:Do not link "helper" executables to /usr/local/bin.' \     '--debug:Output debugging information of use to Cask authors and developers.' \     ':command:->command' \     '*::options:->options'   case "$state" in   (command)     _alternative -C 'brew-cask' \       'aliases:alias:__brew_cask_aliases' \       'commands:command:__brew_cask_commands' ;;   (options)     __brew_cask_command "$line[1]" ;;   esac } _brew_cask "$@"

来自：https://blog.csdn.net/shentian885/article/details/115902464

在 Linux 系统中，你可以通过访问 /proc/<PID>/cmdline 文件来查看进程的启动命令和参数，其中 <PID> 是进程的 ID。这个文件包含了进程启动时使用的完整命令行，参数之间用 null 字符（ASCII码值为0）分隔。

例如，你可以使用以下命令来查看特定进程的命令行参数：
cat /proc/<PID>/cmdline

这将输出进程的启动命令及参数列表，以 null 字符分隔。

问：
的确，cat /proc/45063/cmdline
./multiupd-d
这个实际参数-d前面有空格，如何分离出来
答：
在 `/proc/<PID>/cmdline` 中，命令行参数以 null 字符（ASCII码值为0）分隔。你可以通过读取该文件并以 null 字符分割内容来获得参数列表。在某些编程语言中，你可以按照 null 字符（ASCII码值为0）来分割字符串。

例如，在 Bash 中，你可以使用 `tr` 命令来将 null 字符替换成其他分隔符，然后查看参数列表：

```bash
cat /proc/45063/cmdline | tr '\0' '\n'
```

这将把 null 字符替换成换行符，使得参数列表更易于阅读。
ps -ef|grep multi
root     45055     1  0 11:41 ?        00:00:00 ./multiupd -d
root     45063 45055  0 11:41 ?        00:00:00 ./multiupd -d

cat /proc/45055/cmdline | tr '\0' '\n'
./multiupd
-d
cat /proc/45063/cmdline | tr '\0' '\n'
./multiupd
-d





ps只能看到部分：
root      3195  2205  0 Jun19 ?        00:00:13 containerd-shim -namespace moby -workdir /var/lib/containerd/io.containerd.runtime.v1.linux/moby/0a8be205faf5aa1500e55e83e36a3027c626ae

cat查看：
#cat /proc/3195/cmdline
containerd-shim-namespacemoby-workdir/var/lib/containerd/io.containerd.runtime.v1.linux/moby/0a8be205faf5aa1500e55e83e36a3027c626aea234747a51c949ffb1a9f95c83-address/run/containerd/containerd.sock-containerd-binary/usr/bin/containerd-runtime-root/var/run/docker/runtime-runc-systemd-cgroup



二、执行程序的软链接：
#ls -lart /proc/3195/exe
lrwxrwxrwx 1 root root 0 Jun 25 10:06 /proc/3195/exe -> /usr/bin/containerd-shim



三、环境变量：
#cat /proc/3195/environ
LANG=en_US.UTF-8PATH=/opt/containerd/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/binLD_LIBRARY_PATH=:/opt/containerd/libGOMAXPROCS=2



四、对应的fd编号：
#ls /proc/3195/fd
0  1  10  11  12  13  14  15  16  17  18  19  2  20  3  4  5  6  7  8  9

zsh completions and functions have been installed to:
  /usr/local/share/zsh/site-functions


mac下面sudo su - 后出现：
sudo su
Password:
[Oh My Zsh] Would you like to update? [Y/n]: y
Updating Oh My Zsh
error: 不能变基式拉取：您有未暂存的变更。
error: 请提交或贮藏它们。
There was an error updating. Try again later?

最后我的解决办法是：
cd  ~/.oh-my-zsh/

git status
位于分支 master
您的分支与上游分支 'origin/master' 一致。

尚未暂存以备提交的变更：
  （使用 "git add <文件>..." 更新要提交的内容）
  （使用 "git restore <文件>..." 丢弃工作区的改动）
  类型变更： plugins/fedora/fedora.plugin.zsh
  类型变更： plugins/go/go.plugin.zsh
  类型变更： plugins/gradle/_gradle
  类型变更： plugins/gradle/_gradlew
  类型变更： plugins/per-directory-history/per-directory-history.plugin.zsh

git reset --hard origin/master #强制让本地的代码和远端一致
HEAD 现在位于 08da19e3 emacs: revert cab1ac6e682038b2ceca9d16f8c458c0e79149ca (#7765)

git status
位于分支 master
您的分支与上游分支 'origin/master' 一致。

git pull
remote: Enumerating objects: 9644, done.

root@bogon# upgrade_oh_my_zsh
Note: `upgrade_oh_my_zsh` is deprecated. Use `omz update` instead.
Updating Oh My Zsh
From https://github.com/ohmyzsh/ohmyzsh
* branch              master     -> FETCH_HEAD
Current branch master is up to date.
         __                                     __
  ____  / /_     ____ ___  __  __   ____  _____/ /_
/ __ \/ __ \   / __ `__ \/ / / /  /_  / / ___/ __ \
/ /_/ / / / /  / / / / / / /_/ /    / /_(__  ) / / /
\____/_/ /_/  /_/ /_/ /_/\__, /    /___/____/_/ /_/
                        /____/

Oh My Zsh is already at the latest version.
To keep up with the latest news and updates, follow us on Twitter: https://twitter.com/ohmyzsh
Want to get involved in the community? Join our Discord: https://discord.gg/ohmyzsh
Get your Oh My Zsh swag at: https://shop.planetargon.com/collections/oh-my-zsh


参考：https://www.chendalei.com/articles/2018/10/09/1539052284615.html
提示：

Updating Oh My Zsh
error: cannot pull with rebase: You have unstaged changes.
error: please commit or stash them.
There was an error updating. Try again later?

这种情况很可能发生，因为您已经在.oh-my-zsh目录中编辑了一些内容但尚未提交。

导航到该目录（.oh-my-zsh/目录）。
运行git status。 它应该显示已更改的文件列表。
使用git add暂存这些文件
使用git commit -m提交这些更改
你现在应该可以升级了。

可选：在步骤2之后，您可以运行git commit -a -m，它应该暂存所有未暂存的文件并提交这些更改（组合步骤3和4）。
然后通过upgrade_oh_my_zsh命令升级zsh。

参考来源：https://github.com/robbyrussell/oh-my-zsh/issues/1984

解决办法：
下载VCRUNTIME140.dll文件；微软官网：下载地址：https://www.microsoft.com/en-us/download/details.aspx?id=52685

下载真实地址及名称：vc_redist.x64.exe
https://download.microsoft.com/download/0/6/4/064F84EA-D1DB-4EAA-9A5C-CC2F0FF6A638/vc_redist.x64.exe

此次学习来源：https://mp.weixin.qq.com/s/YqZI4XUlsyPpaGIjy0pG3A
#cat p.c
#include <stdio.h> void (*fun)(int a); void function(int value) {   printf("value= %d\r\n",value); } int main(void) { fun = function;//把function赋值给fun fun(520);//fun就等同于function      while(1) { } }

#./p
value= 520

cat pp.c
#include <stdio.h> int res; //void (*fun)(int a); int (*fun)(int a); int function(int value) {   return value; } int main(void) { fun = function;//把function赋值给fun res = fun(520);//fun就等同于function    printf("res = %d",res);   while(1) { } }
#./pp


cat ppp.c
#include <stdio.h> char temp[3]={1,2,3}; char *p; int main(void) {   p=temp;//将数组名赋值给指针变量p，p就指向数组temp的首地址      printf("value0 = %d\r\n",*p);    //p就代表数组的第一个数据的地址   printf("value1 = %d\r\n",*(p+1));//p+1就代表数组的第二个数据的地址   printf("value2 = %d\r\n",*(p+2));//p+2就代表数组的第三个数据的地址      printf("temp[0] = %d\r\n",p[0]);//p[0]等同于temp[0]   printf("temp[1] = %d\r\n",p[1]);//p[1]等同于temp[1]   printf("temp[2] = %d\r\n",p[2]);//p[2]等同于temp[2]   while(1) { } }
#./ppp
value0 = 1
value1 = 2
value2 = 3
temp[0] = 1
temp[1] = 2
temp[2] = 3

以前的学习：
https://jackxiang.com/post/7485/

https://www.oracle.com/java/technologies/javase/javase-jdk8-downloads.html

真实下载地址：https://download.oracle.com/otn/java/jdk/8u291-b10/d7fc238d0cbf4b0dac67be84580cfb4b/jdk-8u291-macosx-x64.dmg

源码包下载地址：https://download.oracle.com/otn-pub/java/jdk/16.0.1+9/7147401fd7354114ac51ef3e1328291f/jdk-16.0.1_osx-x64_bin.tar.gz?AuthParam=1620358646_aa79aa4a7788c49057533e4b661b7f35
下载前的用户名密码：
https://jackxiang.com/post/6461/
来自：https://zhuanlan.zhihu.com/p/34915988


start.sh
#!/bin/bash JAVA_HOME=/Users/jackXiang/app/jdk-16.0.1.jdk/Contents/Home JAVA=$JAVA_HOME/bin/java $JAVA -jar cntvFtpClientV3.jar -Djava.ext.dirs=$JAVA_HOME/lib

在Centos8环境中配置PHP出现如上报错：
 
原因是没有安装配置oniguruma导致的 
解决方法如下：
1、获取源码，2.并解压

wget https://github.com/kkos/oniguruma/archive/v6.9.4.tar.gz -O oniguruma-6.9.4.tar.gz
tar -zxf oniguruma-6.9.4.tar.gz
cd oniguruma-6.9.4
备用下载： http://down.24kplus.com/linux/oniguruma/oniguruma-6.9.4.tar.gz

3、编译安装

./autogen.sh && ./configure --prefix=/usr
make && make install
如果是Centos7环境中的话：
 解决方法如下：

yum -y install http://mirror.centos.org/centos-7/7.7.1908/cloud/x86_64/openstack-queens/oniguruma-6.7.0-1.el7.x86_64.rpm
yum -y install http://mirror.centos.org/centos-7/7.7.1908/cloud/x86_64/openstack-queens/oniguruma-devel-6.7.0-1.el7.x86_64.rpm
备用下载：

oniguruma：http://down.24kplus.com/linux/oniguruma/oniguruma-6.7.0-1.el7.x86_64.rpm
oniguruma-devel：http://down.24kplus.com/linux/oniguruma/oniguruma-devel-6.7.0-1.el7.x86_64.rpm
————————————————
版权声明：本文为CSDN博主「薄凉小伙」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/Kangshuo2471781030/article/details/107241779

#coding=utf-8 import sys import os import xlrd  # 导入库 for i in range(len(sys.argv)):     print ("sys.argv[%d] = %s" % (i, sys.argv[i]))

python excelsearchKeyList.py k.txt
sys.argv[0] = excelsearchKeyList.py
sys.argv[1] = k.txt

二）判断字符为空：
#coding=utf-8 import sys import os import xlrd  # 导入库 # 打开文件 if len(sys.argv) >= 2:   file = open(sys.argv[1])   # file = open("/Users/jackXiang/pylearning/k.txt") else:   file = open("/Users/jackXiang/pylearning/cctvhyjykeywordslist.txt") for line in file:       line = line.replace('\r','').replace('\n','').replace('\t','')       if line.strip()=='':         print ('line is null')       else:         print(line)

python excelsearchKeyList.py  k.txt
原创视频
line is null
line is null



三）判断是否传入新的参数：
# 打开文件 if len(sys.argv) >= 2:   print(sys.argv[1])   file = open(sys.argv[1]) else:   file = open("/Users/jackXiang/pylearning/cctvhyjykeywordslist.txt") for line in file:       print(line) os._exit(-1) # os._exit(-1) sys.exit()

来自：https://blog.csdn.net/index20001/article/details/74294945#:~:text=Python%E7%A8%8B%E5%BA%8F%E6%9C%89%E4%B8%A4%E7%A7%8D,%E8%A7%A3%E9%87%8A%E5%99%A8%E5%B0%86%E4%BC%9A%E9%80%80%E5%87%BA%E3%80%82

问题：
#ssh -l xiangdong 10.10.0.109 xiangdong@10.10.0.109's password: sudo su - Activate the web console with: systemctl enable --now cockpit.socket Last login: Wed Feb  3 04:50:21 2021 from 10.10.2.1 -bash: warning: setlocale: LC_ALL: cannot change locale (zh_CN.UTF-8) sudo su - /bin/sh: warning: setlocale: LC_ALL: cannot change locale (zh_CN.UTF-8) /bin/sh: warning: setlocale: LC_ALL: cannot change locale (zh_CN.UTF-8) /bin/sh: warning: setlocale: LC_ALL: cannot change locale (zh_CN.UTF-8) -bash: warning: setlocale: LC_ALL: cannot change locale (zh_CN.utf8): No such file or directory -bash: warning: setlocale: LC_ALL: cannot change locale (zh_CN.utf8): No such file or directory -bash: warning: setlocale: LC_ALL: cannot change locale (zh_CN.utf8): No such file or directory [xiangdong@localhost ~]$ sudo su - -bash: warning: setlocale: LC_ALL: cannot change locale (zh_CN.utf8): No such file or directory -bash: warning: setlocale: LC_ALL: cannot change locale (zh_CN.utf8): No such file or directory

rpm -qa *langpack*

cat /etc/locale.conf
LC_ALL="zh_CN.utf8"
LC_CTYPE="zh_CN.utf8"
LANG="zh_CN.utf8"

解决办法：
yum install glibc-langpack-zh -y

一、现象


Linux关机或重启时提示A stop job is running for ..


导致关机慢。


二、解决方法


编辑/etc/systemd/system.conf


修改下面两个变量为：


DefaultTimeoutStartSec=10s
DefaultTimeoutStopSec=10s
执行：systemctl daemon-reload



来自：https://blog.csdn.net/penge199761/article/details/79136826

质数定义：只有1和它本身两个约数，除此之外再没有其它约数。

示例1：
#include <stdio.h> #include <stdlib.h> int main() {    int a;    a=5;    if(a%2!=0 && a%3 !=0 && a%4!=0)        printf("质数");    else       printf("合数"); }
make zs
cc     zs.c   -o zs

#./zs
质数

示例2：
#include <stdio.h> #include <stdlib.h> int main() {    int a,count;    a=5;       if(a%2==0)       count++;    if(a%3==0)       count++;    if(a%4==0)       count++;    if(count == 0)        printf("质数");    else       printf("合数"); }

for循环介入：
#cat zs.c #include <stdio.h> #include <stdlib.h> int main() {    int a,count,i;    a=5;    for(i=2;i<=4;i++){       if(a%i==0)         count++;    }      if(count == 0)        printf("质数");    else       printf("合数"); }

引入scanf：
#include <stdio.h> #include <stdlib.h> int main() {    int a,count,i;    scanf("%d",&a);    for(i=2;i<=a-1;i++){       if(a%i==0)         count++;    }      if(count == 0)        printf("质数");    else       printf("合数"); }


#./zs
5
质数
#./zs
6
合数

打印出约数：
#include <stdio.h> #include <stdlib.h> int main() {    int a,count,i;    scanf("%d",&a);    for(i=2;i<=a-1;i++){       if(a%i==0){         count++;         printf("%d ",i);       }    }      if(count == 0)        printf("质数");    else       printf("合数"); }


#./zs
5
质数
#./zs
6
2 3 合数

更快一点，引入break：
#include <stdio.h> #include <stdlib.h> int main() {    int a,count,i;    scanf("%d",&a);    for(i=2;i<=a-1;i++){       if(a%i==0){         count++;         break;       }    }      if(count == 0)        printf("质数");    else       printf("合数"); }


#./zs
6
合数
#./zs
1000
合数



任一大于2的偶数都可以写成两个质数之和，采用部分穷举法，可以从1-100，1-10000啥的，但无法穷举所有：
#include <stdio.h> #include <stdlib.h> int main() {    int n,k,a,b,i,count1,count2;    scanf("%d",&n);//4到n,如：4-n内的偶数都可以写成两个质数之和    for(k=4;k<=n;k=k+2){       for(a=2;a<=k/2;a++){         //判断a是否为质数         count1=0;         for(i=2;i<=a-1;i++){            if(a%i==0){              count1++;              break;            }         }         if(count1==0){ //如果a为质数           b=k-a;           //判断b是否为质数           count2=0;           for(i=2;i<=b-1;i++){              if(b%i==0){                count2++;                break;              }           }           if(count2==0){ //如果b也是质数              printf("%d=%d+%d\n",k,a,b);                        break;           }         }        }    } }

#./zs
100
4=2+2
6=3+3
8=3+5
10=3+7
12=5+7
14=3+11
16=3+13
18=5+13
20=3+17
22=3+19
24=5+19
26=3+23
28=5+23
30=7+23
32=3+29
34=3+31
36=5+31
...

背景：有一篇朋友圈说是一个k8s里面Server的Pod销毁上报Zookeeper出现IP地址连接还是原来的，用到了zk的watcher，理论上不应该这样，于是查来查去，其大体意思就是说用了Go的多进程，有一个进程去读取配置了，因为配置还是老的，而用到了Go的多进程，于是文末那代码，我看了下，简单，但不明白，于是顺藤摸瓜学习一下这个rsync的原理。


——————————
#tree  -L 2 .|grep go
├── loop.go
├── pipe.go
├── waitgrouppointer.go
└── wg.go


#cat loop.go
package main

import (
    "fmt"
    //"time"
)

func main(){
    for i := 0; i < 100 ; i++{
        go fmt.Println(i)
    }
    //time.Sleep(time.Second)
}

这个说是运行太快，子进程没输出主进程就退了，啥也不输出。



#cat pipe.go
package main

import (
    "fmt"
    //"time"
)
func main() {
    c := make(chan bool, 100)
    for i := 0; i < 100; i++ {
        go func(i int) {
            fmt.Println(i)
            c <- true
        }(i)
    }

    for i := 0; i < 100; i++ {
        <-c
    }
}

这个说是管道要是上万，会耗尽系统管道资源。


#cat waitgroup.go      
package main

import (
    "fmt"
    "sync"
    //"time"
)
func main() {
    wg := sync.WaitGroup{}
    wg.Add(100)
    for i := 0; i < 100; i++ {
        go func(i int) {
            fmt.Println(i)
            wg.Done()
        }(i)
    }
    wg.Wait()
}


这个说是挺好的，因为它用到了原子加减法，这里好像C语言也有类似的函数，



再就是说它是用来阻塞方进程的，WaitGroup 对象内部有一个计数器，最初从0开始，它有三个方法：Add(), Done(), Wait() 用来控制计数器的数量。Add(n) 把计数器设置为n ，Done() 每次把计数器-1 ，wait() 会阻塞代码的运行，直到计数器地值减为0。

这里首先把wg 计数设置为100， 每个for循环运行完毕都把计数器减一，主函数中使用Wait() 一直阻塞，直到wg为零——也就是所有的100个for循环都运行完毕。相对于使用管道来说，WaitGroup 轻巧了许多。


最后，如果要搞裂成函数，得用引用传入，指针变量：
WaitGroup对象不是一个引用类型
WaitGroup对象不是一个引用类型，在通过函数传值的时候需要使用地址：

#cat waitgrouppointer.go
package main

import (
    "fmt"
    "sync"
    //"time"
)
func main() {
    wg := sync.WaitGroup{}
    wg.Add(100)
    for i := 0; i < 100; i++ {
        go f(i, &wg)
    }
    wg.Wait()
}

// 一定要通过指针传值，不然进程会进入死锁状态
func f(i int, wg *sync.WaitGroup) {
    fmt.Println(i)
    wg.Done()
}


以上来自：https://blog.csdn.net/u013474436/article/details/88749749

溯源来自：https://mp.weixin.qq.com/s/s9G6yDqfeFepbKT6DJASng



#cat wg.go
package main


import (
  "fmt"
  "sync"
)


func main() {
  ok := true
  for i := 0; i <1000; i ++ {
    var arr []int
    wg := sync.WaitGroup{}


    for j := 0; j <2; j ++ {
      wg.Add(1)


      go func() {
        defer wg.Done()
        arr = append(arr, i)
      }()
    }
    wg.Wait()


    if len(arr) < 2 {
      fmt.Printf("error:%d \n", i)
      ok = false
      break
    }
  }


  if ok {
    fmt.Println("ok")
  }
}

阅读全文

Server：echoServer.go
/** * File: echoServer.go * Author: 蛇叔 * 公众号: 蛇叔编程心法(snakeCodeHeart) */ package main import (   "fmt"   "net"   "os"   "syscall" ) const (   PORT = 8888   ADDR = "127.0.0.1"   SIZE = 100 ) func main() {   // 1. create socketfd   socketFd, err := syscall.Socket(syscall.AF_INET, syscall.SOCK_STREAM, 0)   if err != nil || socketFd < 0 {     fmt.Println("socket create err: ", err)     os.Exit(-1)   }   // 2. bind addr and port   ip4 := net.ParseIP(ADDR).To4()   if ip4 == nil {     fmt.Println("net.ParseIP err")     os.Exit(-1)   }   sa := &syscall.SockaddrInet4{Port: PORT}   copy(sa.Addr[:], ip4)   err = syscall.Bind(socketFd, sa)   if err != nil {     fmt.Println("socket bind err: ", err)     os.Exit(-1)   }   // 3. listen   err = syscall.Listen(socketFd, 128)   if err != nil {     fmt.Println("socket listen err: ", err)     os.Exit(-1)   }   // 4. accept   acceptFd, _, err := syscall.Accept(socketFd)   if err != nil {     fmt.Println("socket accept err: ", err)     os.Exit(-1)   }   var (     buf = make([]byte, SIZE)     readn int     err2 error   )   for {     // 5. read data     readn, err2 = syscall.Read(acceptFd, buf)     if err2 != nil {       break     }     if readn > 0 {       // 6. write data       writen, _ := syscall.Write(acceptFd, buf[:readn])       if writen < 0 {         fmt.Println("socket write err: ", err)         break       }     } else if readn == 0 {       fmt.Println("客户端已关闭")       break     } else if readn < 0 && err2 != nil {       fmt.Println("socket read err: ", err2.Error())       break     }   }   // 7. close   _ = syscall.Close(acceptFd)   _ = syscall.Close(socketFd) }

Client：echoClient.go
/** * File: echoClient.go * Author: 蛇叔 * 公众号: 蛇叔编程心法 */ package main import (     "bufio"     "fmt"     "net"     "os"     "syscall" ) const (     PORT = 8888     ADDR = "127.0.0.1"     SIZE = 100 ) func main() {     // 1. 建立socket     socketFd, err := syscall.Socket(syscall.AF_INET, syscall.SOCK_STREAM, 0)     if err != nil || socketFd < 0 {         fmt.Println("socket create err: ", err)         os.Exit(-1)     }     ip4 := net.ParseIP(ADDR).To4()     if ip4 == nil {         fmt.Println("net.ParseIP err")         os.Exit(-1)     }     sa := &syscall.SockaddrInet4{Port:PORT}     copy(sa.Addr[:], ip4)     // 2. 发起主动连接     err = syscall.Connect(socketFd, sa)     if err != nil {         fmt.Println("socket connect err: ", err)         os.Exit(-1)     }     var (         bufReader = bufio.NewReader(os.Stdin)         buf = make([]byte, SIZE)         writen int         readn int         err2 error     )     for {         // 3. 从标准输入读取数据         line, _, err := bufReader.ReadLine()         if err != nil {             fmt.Println("bufReader.ReadLine err: ", err)             break         }         buf = line[:]         // 4. 向socket对端写入数据         writen, err2 = syscall.Write(socketFd, buf)         if writen > 0 {             readn, err2 =syscall.Read(socketFd, buf)             if readn > 0 {                 fmt.Println("read from socket: ")                 fmt.Println(string(buf[:readn]))             } else {                 break             }         } else if writen <= 0 && err2 != nil {             fmt.Printf("socket write; writen:%d,  err: %s\n", writen, err2)             break         }     }     // 5. close socketFd     _ = syscall.Close(socketFd) }

jackXiang@localhost  ~/golearning  ./echoClient
jack
read from socket:
jack
xiang
read from socket:
xiang

Q:果然Ok，两个问题探讨下：
1）        ip4 := net.ParseIP(ADDR).To4()
        sa := &syscall.SockaddrInet4{Port: PORT}
        copy(sa.Addr[:], ip4)  #这个sa.Addr[:] 冒号啥写法呢？

2）for { 这里面是epoll 吗？}

A:
1.
copy(sa.Addr[:], ip4)  #这个sa.Addr[:] 冒号啥写法呢？

sa.Addr 是一个数组，
sa.Addr[:] 将这个数组临时转为切片，底层数组指向sa.Addr数组
方便copy进行切片复制

2. 就是个for{}死循环，防止代码往下执行

来自：
https://mp.weixin.qq.com/s/dRADQ5kUCT80dX7oEAAXyA
https://mp.weixin.qq.com/s/kFHw73w_IIqjKeicesCpZw

背景：[实践OK]Macbook 升级到catalina 后的一些/data目录不见了等问题
:https://jackxiang.com/post/10397/,现在升级后，这个玩法被苹果给屏蔽更深了，怎么办?有人做这个，我不太敢接着搞下去了：
Big Sur 新增了 Signed System Volume 机制，对系统所在的 APFS Volume 增加了更多的保护

Reboot into recovery mode and run:重启mac，按住Command+R，等到系统进入安全模式。
命令 csrutil disable  ,sur还得加上这个：
csrutil authenticated-root disable

Reboot back into MacOS #正常重启进入系统
Find your root mount's device - run mount and chop off the last s, e.g. if your root is /dev/disk1s5s1, you'll mount /dev/disk1s5
Create a new directory, for example ~/mount
Run：sudo mount -o nobrowse -t apfs /dev/disk1s5 ~/mount  ，我这想:sudo mount -o nobrowse -t apfs /dev/disk1s5 /var/data
mount /dev/disk1s5s1 on / (apfs, sealed, local, read-only, journaled) sudo mount -o nobrowse -t apfs /dev/disk1s5s1 ~/mount
/dev/disk1s5s1 :https://iosre.com/t/big-sur/18051

升级到Big Sur后，以写模式挂载root报错（SIP已disable）

sudo mount -uw /

mount_apfs: volume could not be mounted: Permission denied

mount: / failed with 66

Big Sur 新增了 Signed System Volume 机制，对系统所在的 APFS Volume 增加了更多的保护

Reboot into recovery mode and run:
csrutil authenticated-root disable

Reboot back into MacOS
Find your root mount's device - run mount and chop off the last s, e.g. if your root is /dev/disk1s5s1, you'll mount /dev/disk1s5

Create a new directory, for example ~/mount
Run：sudo mount -o nobrowse -t apfs /dev/disk1s5 ~/mount

Modify the files under the mounted directory

Run：sudo bless --folder MOUNT_PATH/System/Library/CoreServices --bootefi --create-snapshot

Reboot your system, and the changes will take place

缺点：

修改后，无法再启用SSV机制，但是可以启用SIP机制，因为系统已修改，导致签名不一致，若启用SSV，则校验失败，报错禁止启动。

csrutil enable

csrutil authenticated-root disable



Big Sur 引入的 Signed System Volume 机制会给 System APFS Volume 的每个文件 /文件夹都设置一份 hash，在 bootloader 启动后，加载内核前会进行完整性校验，校验不通过的话会拒绝引导，并提示用户重装 macOS 。使用 mount 查看挂载点会发现 /多了个 sealed 标志。系统默认会从 SSV snapshot 启动，对用户无感知。

Big Sur’s Signed System Volume: added security protection


来自：https://www.quwenqing.com/archives/1597.html

背景：Chrome您的连接不是私密连接解决办法--一个比较实用的技巧分享，有的勾选确认，发现没有勾选按钮。怎么办？
方法：解决：就是在当前页面用键盘输入  thisisunsafe  ，不是在地址栏输入，就直接敲键盘就行了，页面即会自动刷新进入网页。

阅读全文

Mac / OS X 如何快速显示桌面：F4
Mac苹果电脑怎么设置Launchpad快捷键？F11
=============================
Mac / OS X 如何快速显示桌面（Desktop）？
https://www.zhihu.com/question/21448506

  【收藏】Mac苹果电脑怎么设置Launchpad快捷键？
来自：https://www.zhihu.com/question/21448506

阅读全文