shell script

1. sha-bang #!

#!/bin/bash # Proper header for a Bash script.

2. define all variables (UpperCase) at top at refer them later using $

3. command line argument

• $0: command itself
• $1: 1st argument

4. `pwd`.    //everything between `` is evaluated first

5. $?           //the exit code of previous command is stored in it

6. $(command)      //run command and return value as variable

7. i/o redirection

• 0, 1, 2                    //stdin, stdout, stderr
• 1>filename            //redirect stdout to file
• 2>&1                     //redirect stderr to stdout
• >                            //redirect, overwrite
• >>                          //redirect append

8. sed (stream editor)

sed -n           //don't print input, always use

sed -i            //modify file in-place

echo "hello sed" | sed 's/sed/world/'                                       //hello world
sed 's+http://+https://www.cyberciti.biz+g' input.txt             //sed allow delimiter other than /
sed -i -e '/FOO/s/love/sick/' input.txt.                                    //if the line contain FOO
sed -i 's/foo/bar/gI' hello.txt                                                   //'I' to match all cases of foo

sed -n -e 's/^.*ch\([a-zA-Z0-9-]\{1,\}\).*$/\1/p' input.txt.     //+ does not work in sed, use \{1,\}

9. special character

• $                                              //access variable
• ${HOME}                              //return value of HOME
• $(command)                           //return output of command, maybe separated by space !!!  
• "$(command)"                        //return output of command as a whole
• [   ]                                          //condition, !!! need space after [ and before ], i.e. [ 1 -lt 2 ] 
• [ -z "$var" ]                             //$var is empty

10. awk

• program structure
  
  BEGIN {awk-commands}
  /pattern/ {awk-commands}      // body
  END {awk-commands}
  
  
• variable
  $0: a special variable in awk that represents the entire line
  
  
• command
  awk -f command.awk marks.txt
  awk '/a/{++cnt} END {print "Count = ", cnt}' marks.txt

reference
1. special char in shell scripting
2. tldp sed
3. tutorialspoint regex with sed
4. how-to-use-sed-to-find-and-replace-text-in-files

5. unix tutorials point
6. freeos unix guide

7. shell scripting crash course - beginner level