Crontab

crontab is a useful tool, I really regret that I don’t familar it earlier.

It’s usage really simple, and these two are my frequent using:

crontab -e      #edit crontab file that reside in /var/spool/cron
crontab -l      #list current crontab job

It basic syntax as follow, also really simple crontab-layout.jpg

example of usage:

  1. delete file foo every minute * * * * * rm foo
  2. delete file foo every 15 minutes 15 * * * * rm foo
  3. delete file foo every beginning of hour 0 * * * * rm foo
  4. delete file foo every minute after 3 hours * 3 * * * rm foo
  5. delete file foo every day at 18:30 30 18 * * * rm foo
  6. delete file foo every beginning of month 0 0 0 * * rm foo
  7. delete file foo on beginning of 1st,10th of month 0 0 1,10 * * rm foo

The usage really simple, I now use it to renew my SSL certification and daily update bt-tracker.

tr

tr is really useful when encountered situation that needs struggle with string. Three frequency ways of using tr

  1. shrink multiple characters into single one

    tr -s '[string]' e.g. echo "ssssssspaaaaace" | tr -s 'sa' would convert “ssssssspace” to “space”

  2. delete specific character. I usually use it to delete white space or delimiter

    tr -d '[string]' e.g. echo "blog.garhve.com" | tr -d '.o' would convert url to “blggarhvecm”

  3. convert specific character to another one.

    tr '[string1]' '[string2]' e.g. echo "woopwon | tr "wo" "fe" would result “feepfen”

cut

I use cut mostly to get word from a string, especially get relative path from absolute path. Because I always want to loop to get same sub-directories file from different main directory, cut helps a lot.

for now, I only use it one way

echo string | cut -d '[character]' -f position e.g. echo path/to/most/inner/file | cut -d '/' -f1 this will give me word before first ‘/’, which is ‘path’

Often, rev will co-work with cut to get last one word

e.g. echo blog.garhve.com | rev | cut -d '.' -f 1 | rev this will give word after last ‘/’, which is com.

a worth noting here is that the rev command needs to appear twice because it usage is not so intuitive, it reverse whole string

echo "hello world" | rev will get ‘dlrow olleh`

echo "hello world" | cut -d ' ' -f1 | rev will give result of ‘dlrow’

echo "hello world" | rev | cut -d ' ' -f1 will give same result, as ‘dlrow’

echo "hello world" | rev | cut -d ' ' -f1 | rev will give expected result, as ‘world’

Daily update bt tracker

I already learn shell script for a while.. so I wrote a simple script to test whether I really got used to it, but result is obviously, I need more and more practice to memorize commands.

#! /bin/sh
#bt-tracker.txt
site=https://raw.githubusercontent.com/ngosang/trackerslist/master/trackers_all.txt

file=/path_to_aria_directory/aria2.conf

Addr=user@addr

# Get bt-tracker and format it to fulfill aria needs, then store in variable
bt=$(curl $site | tr -s '[:space:]' | tr '[:space:]' ',')

# sshpass is a software, that can allow me pass ssh password as argument
# so that I don't need to wait prompt
# 410 is the line of bt tracker resides, I now no other way to replace it.
$(sshpass -p 'password' ssh -T $Addr "sed -i '410d' $file | echo $bt >> $file")

Usage of tr and cut, expansion of ‘command tips’ is needed

Ways to find files or specific string in files

In order to look up C definitions, I need to know where linux stores header files or which files store definitions I need. So that here comes up some methods to fulfill this need:

  1. find 'path' -name 'file_name'

e.g. find / -name stdio.h will return multiple location that stdio.h resides. this could expand to look for others

image.png

more info could see man find

  1. grep -rnw 'path' -e 'pattern'

e.g. `grep –include=*.{h,c} -rnw / -e “from_kuid” will return string and filename that contains string.

image.png

  • -r or -R is recursive,
  • -n is line number, and
  • -w stands for match the whole word.
  • -l (lower-case L) can be added to just give the file name of matching files.
  • -e is the pattern used during the search

Along with these, --exclude, --include, --exclude-dir flags could be used for efficient searching:

  • This will only search through those files which have .c or .h extensions:
grep --include=\*.{c,h} -rnw '/path/to/somewhere/' -e "pattern"
  • This will exclude searching all the files ending with .o extension:
grep --exclude=\*.o -rnw '/path/to/somewhere/' -e "pattern"
  • For directories it’s possible to exclude one or more directories using the --exclude-dir parameter. For example, this will exclude the dirs dir1/, dir2/ and all of them matching *.dst/:
grep --exclude-dir={dir1,dir2,*.dst} -rnw '/path/to/somewhere/' -e "pattern"

more info could see man grep.