== 批量删除 ==
  find . -name "*.pdf"|xargs rm -rf
  find . -name "*mobi" -print0|xargs -0 rm -rf

== 删除空文件夹 ==
  find ./aliyunpan/202305/ -type d -empty -delete


== 批量加后缀 == 
  find . -type f |xargs mv {} {}.txt
  find . -type f | xargs -I F mv "F" "F".txt

== 查看有没有非txt的 == 
  find .  -type f|grep -v "txt"

== 将所有非txt结尾的文件都移到一个地方 == 
  find .  -type f -print0 |grep -v "txt"|xargs -0 -I F mv "F" "/Users/liuhui/Documents/20230127/tmp/"


== 修改后缀名TXT—>txt == 
  find . -type f | grep json |xargs -I F echo "F" "F" |sed 's/json/jsonl/2'  |xargs -I {} echo "mv" {} > tmp.sh
  for file_name in `ls *`; do mv ${file_name} AE${file_name}.json;done
  for file_name in `ls *`; do mv ${file_name} ${file_name}l;done


== 批量删除空文件 == 
  find . -name "*" -type f -size 0c | xargs rm -rf
  
== 批量删除大于1MB的文件 == 
  find . -name "*" -type f -size +1M |xargs rm -rf
  
== 查找100k-200k的pdf ==
  find . -name "*pdf" -type f -size +100k -size -100k

== 批量解rar压缩 == 
  for tar in *.rar; do unrar e -o+ $tar; done
  
== 批量解zip压缩 == 
  unzip '*.zip'

== 批量删除某个文件夹 ==
  nohup find . -name "zipout*" -type d|xargs rm -rf & 

== 转码 == 
  for file_name in `ls *.xml`; do ebook-convert ${file_name} .txt --txt-output-formatting=plain;done
  find . -name "*.xml" | xargs -n 1 -I {} ebook-convert {} .txt --txt-output-formatting=plain
  find . -name "*.xml" | xargs -i ebook-convert {} .txt --txt-output-formatting=plain


== xml转txt == 
  find . -name "*.xml" |xargs -I F echo "F" "F" |sed 's/xml/txt/1'  |xargs -I {} echo "html2text -from_encoding utf8 -o" {}
  html2text --no-wrap-link --ignore-emphasis --ignore-links --ignore-images --ignore-tables --decode-errors 'ignore' vbk2012.mdx.part20.txt utf8 >v20.txt

== 将开头n个文件移走 == 
  ls |head -n 10000| xargs -i mv {} ../20230115/


== 加密压缩文件夹 == 
  zip -r -e 20230133.zip ./20230133/

== 压缩 == 
  gzip -9 filename
  find . -type f | grep jsonl |xargs -I F echo "F" ">F" |sed 's/jsonl/jsonl.gz/2'  |xargs -I {} echo "gzip -9 -c" {}
  zip -v -9 -r demo.zip demo
  zip -r -9 -e ~/file/3_dir/raw/20230309.zip 20230309/ 
  zip -rq -9 -e 20230301.zip 20230301/ 
  unzip -P 253874 -q 20230301.zip -d ~/bt2/
  zip -r  ~/network/liuhui/error_log.zip ~/network/liuhui/error_log/


== git常用命令 == 
  git add -A . 来一次添加所有改变的文件 
  git status 
  git reset HEAD 后面什么都不跟的，就是上一次add 里面的内容全部撤销 
  git reset HEAD XXX 后面跟文件名，就是对某个文件进行撤销 

== 查看编码分布 == 
  awk -F ',' '{count[$2]++;} END {for(i in count) {print "        ",i,count[i]}}' check_result.csv

== 爬取github == 
  nohup python publicRepos.py --start 99000000 --end 100000000 --github_tokens_file token.txt &>tmp.log &
  nohup python publicRepos.py --start 199000000 --end 200000000 --github_tokens_file token.txt &>tmp.log &
  find . -name "*.jsonl" -type f  | xargs ls -l
  for file_name in `ls *.html`; do ebook-convert ${file_name} .txt;done 

== 爬取github文件 == 
  nohup python repo_list_filter_export.py ~/bt2/bak_file/chinese_202303/2023030 100000 &>tmp.txt &
  nohup /home/jovyan/file/github_tmp/gh_downloader/gh_jsonl2txt/gh_downloader-linux-amd64  &>tmp.txt &

== 编码识别 == 
  python ~/file/github_tmp/charset_mnbvc/convert_files.py -i /Downloads/20230109 -step 1 -r check_result.csv

== 分割文件 == 
  split -l 100000 -d -a 2 arxiv-metadata-oai-snapshot.json arxiv-metadata-oai-snapshot.json

== 批量kill == 
  ps -ef|grep arxiv|grep -v grep|cut -c 9-16|xargs kill -9
  ps aux | grep -v grep | grep programname | awk '{print $2}' | xargs kill

== 取出最多时间的 == 
  cat zip19.log |grep "耗时 "| sort -t " " -k 4 -n -r | head -n 10
  find . -name tmp.log|xargs ls -l|sort -t " " -k 8 -n -r

== 修改downloadstation 最大下载数 == 
  sudo cat /var/packages/DownloadStation/etc/settings.conf

== 统计当前文件夹下所有json文件行数 == 
  awk 'END{print NR}' *.json

== 进阶：结合xargs，统计22年10月所有json文件行数 == 
  ls -al | grep '202210'| awk '{print $9}'  | xargs -I {} awk 'END{print NR}' {} | awk '{sum+=$1} END{print "sum="sum}'

== 显示当前文件夹下size最大的前十个文件 == 
  ls -lSh  | head -n 10
  du -h ./ | sort -hr | head -n 10
== 显示当前文件夹下时间最新的前十个文件 == 
  ls -lt | head -n 10
  #倒序 
  ls -rlt | head -n 10

== 统计当前文件夹下有几种文件类型 == 
  find . -type f | sed -n 's/..*\.//p' | sort | uniq -c
  
== 删除最后一行文件 ==
  sed -i '$d' spider_log.jsonl
  
== docker相关 ==
  #查看历史docker
  docker ps -a
  #创建docker
  docker run -dit -v /mnt/disk1:/mnt/disk1 -v /mnt/disk2:/mnt/disk2 -v /mnt/disk3:/mnt/disk3 --name cpu huggingface/transformers-pytorch-cpu /bin/bash
  #进入docker
  docker exec -it cpu /bin/bash
  #关闭docker
  docker stop cpu
  #启动docker
  docker start cpu
  #删除创建的docker
  docker rm $(docker ps -aq)
  #修改root权限
  sudo docker exec -u 0 -it notebook2 bash
  passwd jovyan
  apt-get update
  apt-get upgrade
  apt-get install vim
  visudo
  jovyan ALL=(ALL:ALL) ALL
  
== 查看进程详情 ==
  lsof -p 1234
  
== 拷贝子文件夹到另一个地方 ==
  cp -R a/. b/
  mv a/* b/
  
== 本文件夹下最大的文件 ==
  find . -type f -exec du -a {} \; | sort -n -r | head -n 10
  
== 只出现在b文件中的行 ==
  comm -13 a b > c
  
== 远程同步文件夹 ==
  rsync --progress --partial -avz -e "ssh -p 22566" xxx@xxx.org:/home/data/ ./data/