Code in this notebook is tested under the following environment:

• OS: macOS 14.6.1 23G93 arm64
• Kernel: Darwin 23.6.0
• Shell: zsh 5.9
• Terminal: Apple_Terminal

• Shell Notebook
  • Announcement
  • Table of Content
  • Shebang
  • Variables
  • Passing Arguments to Scripts
  • Arrays
  • Arithmetic Operations
  • String Operations
  • If Statement
  • Case Statement
  • For Loop Statement
  • While Loop Statement
  • Break and Continue Statement
  • To-Do

• Format: #!<path_to_shell> (e.g., #!/bin/bash for Bash or #!/bin/zsh for Zsh)
• Position: The first line of the script.

Back to Top

• Definition: A variable can contain a number, a character or a string of characters.

  STUDENT_AGE=18
  StudentGender="F"    # StudentGender=F
  studentName="Alice"  # studentName=Alice

• No space is allowed on either side of = when initialising variables.

• Double quotes ensure the string is treated as a single, literal unit, especially when it contains spaces or special characters.

• Naming Rules:

  • Variable names are case-sensitive.
  • Start with a letter or underscore.
  • Use only letters, numbers and underscores.
  • Lowercase for local variables, e.g. file_name="shell_notebook.md".
  • Uppercase for environment variables and constants, e.g. MAX_ITERS=10.

• Escaping Special Characters: Use \<special_character> to escape special characters.

• Parameter Expansion

  • Format: ${<variable>}

  • Functionality: Manipulate variable values.

    milk_price=3
    echo "Today's milk price is ${milk_price} dollars per bottle."

• Command Substitution

  • Format: `<command>` or $(<command>)

  • Functionality: Capture a command output as part of another command.

    current_working_path=`pwd`
    file_with_timestamp="file_$(date +%Y-%m-%d).txt"

Back to Top

• Format: Append arguments sequentially after the script name. For example:

  ./student.sh Bob 20 Male

• Accessing Arguments:

  • $0: The script name.
  • $1, $2, $3, etc.: The first, second, third, etc. arguments passed to the script.
  • $@: All the arguments passed to the script.
  • $#: The number of arguments passed to the script.

Back to Top

• Format: (<element_1> <element_2> <element_3> <etc.>)

  null=()
  students=("Jeff" "Jerry" "Jason" "Jimmy")

• Indexing: ${<array>[<index>]} or $<array>[<index>] for an indexed element; ${<array>[@]} for all elements

  • Notice that all arrays are 1-indexed. This behaviour may vary based on different configurations.

  echo ${students[1]}   # Jeff
  echo ${students[-1]}  # Jimmy
  echo ${students[@]}   # Jeff Jerry Jason Jimmy

• Modifying

  echo ${students[@]}   # Jeff Jerry Jason Jimmy
  students[2]="Alice"   # replace Jerry with Alice
  echo ${students[@]}   # Jeff Alice Jason Jimmy

• Length: ${#<array>[@]} or $#<array>[@]

  echo ${students[@]}   # Jeff Alice Jason Jimmy
  echo ${#students[@]}  # 4

• Deleting: <array>[<index>]=()

  echo ${students[@]}   # Jeff Alice Jason Jimmy
  students[3]=()	      # delete Jason
  echo ${students[@]}   # Jeff Alice Jimmy
  echo ${#students[@]}  # 3

Back to Top

• Syntax: ((<expression>)) or $((<expression>))

  n=10
  
  ((a = n + 1))   # a=$((n + 1)), 11 = 10 + 1
  echo $a         # 11
  ((b = n - 1))   # b=$((n - 1)), 9 = 10 - 1
  echo $b         # 9
  
  ((c = n * 3))   # c=$((n * 3)), 30 = 10 * 3
  echo $a         # 30
  ((d = n / 3))   # d=$((n / 3)), integer division, 3 = 10 / 3
  echo $d         # 3, quotient
  ((e = n % 3))   # e=$((n % 1)), modulo, 1 = 10 % 3
  echo $e         # 1, remainder
  ((f = n ** 2))  # f=$((n ** 2)), 100 = 10 ** 2
  echo $f	        # 100
  
  ((g = n++))     # assign n to g before adding 1 to n, 10 = 10++
  echo $g $n      # 10 11
  ((h = n--))	    # assign n to h first before subtracting 1 from n, 11 = 11--
  echo $h $n      # 11 10
  ((i = ++n))     # add 1 to n before assigning n to i, 11 = ++10
  echo $g $n      # 11 11
  ((j = --n))	    # subtract 1 from n before assigning n to i, 10 = --10
  echo $h $n      # 10 10

• Difference

  • ((<expression>)): Used for arithmetic operations, assignments, and conditional checks.
  • $((<expression>)): Used for inline calculations, variable assignments, and output.

  compare() { 
    if (($1 < $2)); then  # ((<expression>)) will return 0 (false) or 1 (true) 
      echo "$(($1)) is smaller than $(($2))"
    else 
      echo "$(($1)) is greater then $(($2))"
    fi
  }
  
  compare 2**3 3**2  # 8 (2^3) is smaller than 9 (3^2)

Back to Top

• String Length: ${#<string>}

• Extracting Substrings

  • ${<string>:<start>:<length>}: Extract a substring of length <length> starting after the index <start>.

  string="hello world"
  substring=${string:6:5}  # a substring of length 5 starting after the index 6
  echo ${substring}   # "world"

  • ${<string>:<start>>: Extract a substring starting after the index <start> and ending to the end of line.

    string="hello world"
    substring=${string:2}  # a substring starting after the index 2
    echo ${substring}   # "llo world"

• Replacing Substrings

  • Replacing the First Occurrence of a Substring with Replacement

    Format: ${<string>[@]/$<substring_to_replace>/$<replacement_of_substring>}

    string="To study or not to study, that is the question."
    substring="study"
    replacement="sleep"
    echo ${string[@]/$substring/$replacement}   # "To sleep or not to study, that is the question."

    • Replacing All Occurrences of a Substring with Replacement

      Format: ${<string>[@]//$<substring_to_replace>/$<replacement_of_substring>}

      string="To study or not to study, that is the question."
      substring="study"
      replacement="be"
      echo ${string[@]//$substring/$replacement}  # "To be or not to be, that is the question."

    • Deleting the First Occurrence of a Substring

      Format: ${<string>[@]/$<substring_to_replace>}

      string="To study or not to study, that is the question."
      substring=" study"
      echo ${string[@]/$substring}  # "To or not to study, that is the question."

    • Deleting All Occurrences of a Substring

      Format: ${<string>[@]//$<substring_to_replace>}

      string="To study or not to study, that is the question."
      substring=" study"
      echo ${string[@]//$substring}  # "To or not to, that is the question."

    • Replacing the Substring if It Is at the Start of a String

      Format: ${<string>[@]/#$<substring_to_replace>/$<replacement_of_substring>}

      string="To study or not to study, that is the question."
      substring="To study"
      replacement="To play"
      echo ${string[@]/#$substring/$replacement}  # "To play or not to study, that is the question."

    • Replacing the Substring if It Is at the End of a String

      Format: ${<string>[@]/%$<substring_to_replace>/$<replacement_of_substring>}

      string="To study or not to study, that is the question."
      substring="the question."
      replacement="not a question."
      echo ${string[@]/%$substring/$replacement}  # "To study or not to study, that is not a question."

• Capitalising/Uncapitalising

  • Capitalising/Uncapitalising All Letters

    Format: ${<string>:u} for uppercasing; ${<string>:l} for lowercasing

    string="Hello World"
    echo ${string:u}  # HELLO WORLD
    echo ${string:l}  # hello world

    • Capitalising the First Letter of Each Word

      Format: ${(C)<string>}

      string="i love shell"
      echo ${(C)string}  # I Love Shell

Back to Top

• Syntax

  # if
  if [ <if_expr> ]; then 
    <if_code>
  fi
  
  # if-else
  if [ <if_expr> ]; then 
    <if_code>
  else 
    <else_code>  
  fi
  
  # if-elif-else
  if [ <if_expr> ]; then 
    <if_code>
  elif [ <elif_expr> ]; then 
    <elif_code>
  else 
    <else_code>
  fi

  In each [ <expr> ], <expr> must be separated with one space on each side.

• Operators (more details on man test)

  File Operator	Description
  -d <file>	True if <file> exists and is a directory.
  -e <file>	True if <file> exists (regardless of type).
  -f <file>	True if <file> exists and is a regular file.
  -r <file>	Ture if <file> exists and is readable.
  -s <file>	True if <file> exists and has a size greater than zero.
  -w <file>	True if <file> exists and is writable. True indicates only that the write flag is on. <fiile> is not writable on a read-only file system even if this test indicates true.
  -x <file>	True if <file> exists and is executable. True indicates only that the execute flag is on. If <file> is a directory, true indicates that <file> can be searched.
  <file_1> -nt <file_2>	True if <file_1> exists and is newer than <file_2>.
  <file_1> -ot <file_2>	True if <file_1> exists and is older than <file_2>.
  <file_1> -ef <file_2>	True if <file_1> and <file_2> exist and refer to the same file.

  String Operator	Description
  -z <str>	True if the length of <str> is zero.
  -n <str>	True if the length of <str> is nonzero.
  <str>	True if <str> is not the null string.
  <str_1> = <str_2>	True if <str_1> and <str_2> are identical.
  <str_1> != <str_2>	True if <str_1> and <str_2> are not identical.
  <str_1> < <str_2>	True if <str_1> comes before <str_2> based on the binary value of their characters.
  <str_1> > <str_2>	True if <str_1> comes after <str_2> based on the binary value of their characters.

  Integer Operator	Description
  <int_1> -eq <int_2>	True if <int_1> and <int_2> are algebraically equal.
  <int_1> -ne <int_2>	True if <int_1> and <int_2> are not algebraically equal.
  <int_1> -gt <int_2>	True if <int_1> is algebraically greater than <int_2>.
  <int_1> -ge <int_2>	True if <int_1> is algebraically greater than or equal to <int_2>.
  <int_1> -lt <int_2>	True if <int_1> is algebraically less than <int_2>.
  <int_1> -le <int_2>	True if <int_1> is algebraically less than or equal to <int_2>.

  Logical Operator	Description
  ! <expr>	True if <expr> is false.
  <expr_1> -a <expr_2>	True if both <expr_1> and <expr_2> are true.
  <expr_1> -o <expr_2>	True if either <expr_1> or <expr_2> are true.
  ( <expr_1> )	True if <expr> is true. In some cases, parentheses need to be escaped. E.g., [ \( 1 \lt 2 \) -o \( "a" = "A" \) ]

• test <expr> is equivalent to [ <expr> ].

• Each evaluation has an exit status:

  • 0 if the expression is evaluated as true;

  • 1 if the expression is evaluated as false or missing;

  • >1 if an error occurred.

• $? holds the exit status of the last command.

  test "bash" = "zsh"  # compare two strings
  echo $?   # 1 (false)
  
  [-e /bin/zsh ]       # whether zsh exists
  echo $?   # 0 (ture)

Back to Top

• Advantage: When all decision branches depend on a single variable, the case statement handles more efficiently than the repeated if statement.

• Syntax

  case ${<variable>} in 
    <case_1>) 
      <case_1_code>
      ;; 
    <case_2> | <case_3>) 
      <case_2_code>
      ;;
    *)
      <case_default_code>
      ;;
  esac

  The Double semicolon ;; will terminate the case statement and jump to the end of the code.

  The asterisk symbol *) will be used as the default case, which will handle all the other cases that are not matched above.

• Example

  #!/bin/zsh
  
  echo "Enter the name of a month:"
  read month  # read a month name from the terminal input
  
  case $month in 
    "February") 
      echo "There are 28/29 days in $month." 
      ;; 
    "April" | "June" | "September" | "November") 
      echo "There are 30 days in $month." 
      ;; 
    "January" | "March" | "May" | "July" | "August" | "October" | "December") 
      echo "There are 31 days in $month." 
      ;; 
    *) 
      echo "Please check if you entered the correct month name." 
      ;;
  esac

Back to Top

Syntax

for <arg> in <array>; do 
  <code>
done

Example:

#!/bin/zsh

students=("Jeff" "Jerry" "Jason" "Jimmy")
# The following loop will print all the student names one by one
for student in ${students[@]}; do 
  echo $student 
done

Jeff
Jerry
Jason
Jimmy

Back to Top

Syntax:

while [ <condition> ]
do
  <code>
done

Example:

#!/bin/zsh

students=("Jeff" "Jerry" "Jason" "Jimmy") 
# While the array contains more than one name
while [ ${#students[@]} -gt 0 ]; do 
  # Print the remaining student names 
  echo ${students[@]} 
  # Delete the last studnet name 
  students[-1]=() 
done

Jeff Jerry Jason Jimmy
Jeff Jerry Jason
Jeff Jerry
Jeff

Back to Top

Once break is encountered, the loop terminates, and no further iterations are executed

Once continue is encountered, the remaining code in that iteration is skipped, but the loop itself continues with the next iterations.

Example:

#!/bin/zsh

# The following example will print all the even numbers less than 10
for num in {0..100}; do
  if (($num > 10)); then 
    break
  fi
  if (($num % 2 == 0)); then 
    echo $num
    ((num = num + 1))
  else
    ((num = num + 1))
    continue
  fi
done

0
2
4
6
8
10

Back to Top

• TODO