Linux commands and shell scripting notes

Bash

Useful Linux commands


  • How many users are logged in: who
  • Who is the current user: whoami
  • Print the date: date
  • Show calendar: cal
  • Print working directory: pwd
  • Print files and directory of the current directory: ls
  • Manual help: man whoami
  • Create a new directory: mkdir directoryname
  • Create a new empty file: touch filename
  • Remove a directory: rmdir directoryname
  • Remove a file: rm filename
  • Remove a directory and its content: rm -r directoryname
  • Display list of available commands: help
  • Clear terminal: clearctrl+l
  • Exit session: “exit`
  • Display date: date
  • Display time: time
  • Print information: hello
  • Increase the font size: Ctrl, shift and +
  • Print 5 random numbers between 1 and 100: shuf -i 1-100 -n 5
  • Creating file using the shuf command: touch dir$(shuf -i 1-10 -n 1)/sunny.txt
  • Simple user: $ prompt
  • Root user/admin: # prompt
  • Switch from simple user to root user: sudo -i
  • Switch from root to simple user: exit

Linux Files Types


  • Directory file.
  • Normal file [binary, text].
  • Device file : every device is represented as a file.
  • To show the type of a file: file file.txt
  • Linux read files based on their content and not based on their extensions.

Terminal


  • Open the terminal: ctrl+alt+t
  • Close the terminal: ctrl+d
  • Show the file that represent the current terminal: ttl —> /dev/pts/0
  • Show files: ls -l -l means long listing.

File Navigation System


  • Display hidden files: ls -a -a means all.
  • Home directory: /home/user1/home contains the home users directories.
  • Go directly to the home directory: cd or cd ~
  • Go to the previous directory: cd -
  • / is the topmost root.

Linux sub directories


  • bin directory: it contains all binaries executable related to our linux commands, which touch will print /user/bin/touch
  • sbin directory: it means system bin, an it contains binary executable related to the super user.
  • etc directory: it contains system configuration information required by the operating system, for example the passwords of the users /etc/passwd, groups info /etc/group and hosts info /etc/hosts(ip addresses and dns names).
  • tmp directory: tmp means temporary, if any file or directory is required temporary, it is created under the tmp folder in the current session. the content of the directory will be deleted automatically after the shutdown.
  • dev directory: dev means device, all the device related files will be stored in this directory, using those files, we can communicate with the device, tty file for terminal related files, hd for hard drive file …
  • mnt directory, media directory(for mounting).
  • opt directory: opt means optional, it contains all the 3rd part software installation files.
  • lib directory: lib means libraries, which are required by applications.
  • var: var means variable data, variable data will be stored inside this directory, for example the log files.
  • usr: all users related software.
  • home: every user has a separate folder to hold his specific data like images, documents and so one.
  • root directory: home directory of the super user.
  • proc: proc means process, for each process a unique id is allocated, and a separate directory will be created inside the proc folder.
  • boot directory: it contains the required files to boot the linux system.

ls command and its options


  • List all files and directories of the directory dir1: ls dir1
  • List all files and directories of the current directory: ls
  • List content in reverse alphabetical order: ls -r
  • List content in a long format(more details on the content): ls -l
  • List content based on creation time: ls -t
  • Possibility to combine all the above options: ls -ltr
  • List also the hidden files: ls -a
  • List contents by type: ls -F (/, *, @)
  • List contents recursively: ls -R
  • List contents with total size(1 block = 1024kb): ls -s
  • List contents of the top 10 entries: ls -l /dir | head
  • List contents of the top 5 entries: ls -l /dir | head -5
  • List contents of the bottom 10 entries: ls -l /dir | tail
  • List contents page by page: ls -l /dir | more or ls -l /dir | less, less is more powerful(forward and backward direction).

Creation of directories


  • To create a directory in the same pwd: mkdir directoryName
  • To create multiple directories in the same pwd: mkdir dir1 dir2 dir3
  • To create multiple directories from the same pwd: mkdir -p dir1/dir2/dir3dir1, dir2 will be created only if they don’t exist already.
  • To create multiple directories mkdir -p dir1/dir2{dir3-1,dir3-2,dir3-3}/dir{1..31}, every dir3-x will contains 31 directories.
  • Show directory content as a tree data structure: tree

Removing directories rm and rmdir commands


  • To remove directories: rmdir dir1 dir2 (rmdir removes only directories, no file will be deleted, and it will works only if the directories are empty).
  • To remove directories: rm -r dir1, or rm -R dir1. For files no need to the -r option, -r for recursive operations.
  • To get confirmation before deleting the contents: rm -r -i dir1
  • Force removal -frm -rf dir1
  • Verbose option -vrm -rv dir1, to print the flow of the deletion process.

Copying moving and renaiming files and directories


  • To copy file content from one file to another: cp a.txt b.txt, if b.txt doesn’t exist, a new file will be created, if it already exists, the file will be overwritten. cp a.txt home/anotherDirectory/c.txt to copy content to another destination.
  • To copy files to directories: cp a.txt b.txt c.txt home/anotherDirectory/d.txt, to copy all files cp dir1/* dir2
  • To copy total directory: cp -r dir1 dir2, the content of dir1 will be copied to dir2
  • Renaming a file: mv a.txt b.txt, to rename a directory mv dir1 dir2, if dir2 already exists, dir1 wil be moved inside dir2
  • Moving files from one directory to another: mv dir1/* dir2mv a.txt b.txt dir2

Creation of files using cat touch gedit and vi


  • Using the cat command: cat > a.txt, a possibility to write content on the a.txt file, ctrl+c to save and exit, if the file is already available, the data will be overwritten, to perform append operation, cat >> a.txt
  • Using the touch command: touch b.txt, a new empty file will be created, if the file is already available, the data will be not be overwritten, only the last date of modification will be modified.
  • Using the gedit command: gedit c.txt, ctr+s to save, ctrl+q to quit the graphical editor.
  • Using vi editor: gedit d.txt, press i to enter the insert mode, press esc key, :wq to save and quit the editor. vim is an advanced version of the vi editor.

View files contents using cat tac rev head and tail


  • View content using cat command: cat < a.txt, where < symbol is optional. -n option to display line numbers, -n to skip blank lines numbering. cat a.txt b.txt will print the content of the two passed files.
  • Create and add content to a file: cat > a.txtadded contentctrl+d to exit the editing section.>> for append operation.
  • Copy content from one file to another using cat command: cat a.txt >> b.txt, using > will overwrite the content of the b.txt file.
  • Copy content from many files to one file: cat a.txt b.txt >> c.txt
  • View content of a file in reversed order: tac a.txt, the order here is the order of the file lines.
  • View content of a file in a horizontal reversed order: rev a.txt, if the content of a.txt is content, the result of the previous command will be tnetnoc
  • cat command is suitable for small files, if the files are huge, head and tailmore and less are the most recommended commands.
  • To view specified lines from the top of a file: head a.txt, first 10 lines head -n 10 a.txt, abbreviation: head -10 a.txt
  • To view specified lines except some specified lines: head -n -5 a.txt will display all lines except the last 5 lines.
  • To view specified number of character: head -c 5 a.txt will show the first 5 characters.
  • To view specified lines from the bottom of a file: tail a.txt last 10 lines, last 5 lines tail -n 5 a.txt, abbreviation: tail -5 a.txttail -c 100 a.txt will print the last 100 characters.
  • To view content from line 3 to line 7 using piping: head -7 a.txt | tail -5

View files content using less and more


  • To view file content page per page: more a.txt, press enter to display next line, press space bar to display the next page, press q to exit, more -d a.txt to display more info about the file. By using more command, we can view file content page per page only in forward direction.
  • To view file content page per page either in forward direction or backward direction: less a.txt, press d to display the next page (d means down), press b to display the previous page (b means backward).

Creation of hidden files and directories


  • If any file name starts with ., such file is a hidden file.
  • Create hidden files: the name of the file should begin by a ., example: touch .a.txtcat > .b.dat
  • Display hidden files: ls -a
  • Create hidden directories: mkdir .dir1
  • Conversion of Normal/hidden files: using renaiming operation, mv .a.txt a.txt

Copying moving and renaiming files and directories using cp and mv commands (2)


  • cp a.txt b.txt, if b.txt exists, its content will be overwritten directly, to get confirmation message, cp -i a.txt b.txt
  • Copy multiple files content to one destination file: cat a.txt b.txt > c.txt
  • Moving and renaming files: mv oldname.txt newname.txt
  • Moving and renaming directories: mv olddirname newdirname, if olddirname exists already, then olddirname will be moved to newdirname, otherwise the renaming operation will take place.
  • Moving selected files to a directory: mv a.txt b.txt dir1
  • Moving all files of one directory to another directory: mv dir1/* dir2
  • Moving using absolute path: mv ~/dir1/* ~/dir2~/dir2 equivalent to /home/userName/dir2

Comparing content of files (diff sdiff…)


  • cmp command: comparing files byte by byte, cmp file1.txt file2.txt, if the files content are the same, no output, otherwise information about the first found difference, byte number and line number will be shown. Example: echo "hello" >> file1.txtecho "hello2" >> file2.txtcmp file1.txt file2.txt, output will be file1.txt file2.txt differ: char 6, line 1
  • diff command: it will show all differences, diff file1.txt file2.txt
  • sdiff command: all differences in a parallel comparison.
  • vimdiff command: advanced tool to show the differences, it will highlight all differences in a vim editor, sudo apt-get install vim to install the tool, two windows will be shown, ctrl+w+w to go to the other window, :q to close current window, :qa to close all windows, :qai close and ignore all changes.

Soft and hard link


  • In Windows, we have only soft link files (shortcut|raccourci), in linux we have also hard link files.
  • Hard link file is just another name of the same file. We can create hard link files using the following command ln sourcefile.txt hardlinkfile.txt
  • If we deleted the source file, the hard link file wil still exists. If we modify the content of the (sourcefile|link hard file), the update will be reflected also in the (sourcefile|link hard file).
  • Soft link file is just a shortcut of the original file (same as windows). If we delete the original file, the soft link file will be broken. We can create soft link files using the following command ln -s sourcefile.txt softlinkfile.txt
  • Change made in the original file will also be reflected to the Soft link file.
  • For directories, it is not possible to create hard link.

Sort commands


  • Sort command, to sort files content line by line in an alphabetical order: sort a.txt will only print the sorted order. sort a.txt > b.txt will redirect the sorted order output to the new file. sort a.txt > temp.txt and mv temp.txt > a.txt will make change on the same source file.
  • Sort in a reverse alphabetical order: sort -r a.txt
  • The spaces in the beginning of a line are ignored, blank line came first when sorting, Upper case before lowercase, If the file contains numbers, the numbers will come before letters.
  • If we want only unique lines, no duplicate, we should use -u option, sort -u a.txt
  • If we want to sort based on numerical values: sort -n a.txtsort -nr a.txt
  • Sort based on the column line number: ls etc/ -l | head -7 | sort -rk 5 will show the info of the top 7 files in etc folder and will sort them recursively based on the fifth column (length of file).
  • Sort based on the column lines in a file composed by lines in the following format token1:token2:token3: is called a separator. To sort such file: sort -k 3 -t ":" a.txt

Uniq commands


  • Find unique content sort -u a.txt, but uniq command is more powerful, uniq a.txt works only on sorted files, sort a.txt | uniq.
  • With uniq command, we can use multiple options:
    • -d: to display ony duplicate lines.
    • -c: to display numbers of occurrences of each line.
    • -i: to ignore case while comparing.
    • -u: to display ony unique lines.

Input and Output


  • command take some input and produces some output:
    • Input in two forms: stdin or command line argument, stdin example: echo "hello" >> file.txt, command line argument example: touch file.txt
    • Output in two forms: stdout or stderror, stderror in cas when the terminal print arrows.
  • Standard Input, Standard Output and Standard Error are data stream and can be redirected from one place to another place. Hence, piping and redirection are possible on those data streams.
    • stdIn associated with the number 0, by default it is connected with the keyboard.
    • stdOutput associated with the number 1, connected with the terminal.
    • stdError associated with the number 2, connected with the terminal.

Redirecting stdInput stdOutput and stdError


  • Redirecting stdOutput: We can perform redirection using > and >>cat 1> output.txt
  • Redirecting standard error: Instead of terminal we can redirect messages from terminal to another place, a file as an example. cal jkbjkcabjkac 2> log.txt, 2 stands for stdError.
  • Redirecting standard input: we can use the < symbol to perform input redirection, cat 0< a.txt, 0 is optional.
  • Examples:
    • cat 0<a.txt 1>copy.txt 2>error.txt
    • cat <a.txt >copy.txt 2>error.txt
    • cat <a.txt &>copy.txt, means either output or error direction to the copy.txt file.
    • Redirect the documentation content of the ls command to some output file: man ls > newfile.txt.

Piping


  • ls -l /etc | more, the output of the first command will be the input of the second command.
  • ls -l /etc > somefile.txt | more, no piping because the redirection symbol breaks the piping.
  • tee command: if we want to save the output of intermediate command and want to pass that output as input to next command, ls -l /etc tee somefile.txt | more
  • rm command can’t take data stream as input, but can only get arguments from the command line terminal, for that purpose, the command xargs can be used to converts data stream to arguments. Example: cat /etc | xargs rm
  • pipe symbol | can be used to link two commands, > is useful when redirecting the output to another place.

Execute multiple command


  • ls -l /etc | more: the preceding commands are dependent commands, to run multiple independent commands in one line, we can use the ; symbol or by using &&.
  • By using ;: e.g. cmd1;cmd2;cmd3, if cmd2 fails, cmd3 will be executed.
  • By using &&: e.g. cmd1&&cmd2&&cmd3, if cmd2 fails, cmd3 will not be executed.

Commands aliasing


  • Aliases are nicknames given to a given command.
  • Syntax: alias nickname="original command name", example: alias cls=clear. Spaces are not allowed before and after the = symbol.
  • To list the existed aliases: alias, to delete a given alias: unalias cls, to delete all aliases unalias -a
  • To know the original name of an alias: type cls
  • Aliases are temporary unless we persist them permanently:
    • Editing the .bashrc file, It could be found inside the user home directory.
    • Create our new aliasing file, the name of the file should be .bash_aliases and should be located in the user home directory.

Regular expressions


  • If we want to represent a collection of strings according to a particular pattern, then we should go for Regular expressions.
  • Some reg expressions rules:
    • * —-> 0 or more character.
    • ? —-> 1 character.
    • [abc] —-> a or b or c.
    • [!abc] —-> only one character but not a or b or c.
    • [a-z] —-> any lower case character from a to z.
    • [A-Z] —-> any upper case character from A to Z.
    • [a-zA-Z] —-> any alphabet from a to z or from A to Z.
    • [0-9] —-> any digit from 0 to 9.
    • [a-zA-Z0-9] —-> any of the two above.
    • [!a-zA-Z0-9] —-> negation of the above.
  • Examples: ls *ls *.*ls *.txtls a*ls a*t.*ls a?.*.
    • ls [abc]*: aaaaaa, bbbb, ….
    • ls [!abc]*: should not start with a, b or c.

locate


  • We can use locate to locate files and directories in our system.
  • locate file.txtlocate *.txt: to locate the given entries, locate command look in a database to locate the given entries, the database is updated only one time a day, so to update the database we should use the following command: sudo updatedb
  • locate -S: to get the statistics of the database, locate -i FilE.txt to ignore case sensitivity. locate --limit 5 a.txt to limit the search to 5 entries.
  • locate --existing b.jpeg, or with -e option: check if the given file exist or not.
  • Advantage of using a database is related to performance.

find command


  • It provides more search option comparing to the locate command. contrary to the locate command, the find command will search directly in the file system.
  • We can search only files, only directories, search by name, search by size…
  • find: by default, it will look for all files and directories from the current directory and below in the linux file system.
  • find /home: look for all files and directories from the current directory and below.
  • find -maxdepth 2: will limit the search only in the 2 levels.
  • By default the find command will also look for hidden files.
  • find -type f: will limit the search to the files.
  • find -type d: will limit the search to the directories.
  • find . -type f -name d?.*: look for files inside the current directory and below with the given name that match given regular expression, -iname option to ignore case.
  • find . -type f -size +200k: look for files that have size greater than 200 kbytes, -size -200k for less than, -size 100k for equality size.

Compression and uncompression of files (episode 33)


Creation of archive files

  • tar command: tape archive, to group multiple files and directories into a single archive, tar -cvf demo.tar file1 file2 dir1-c create, -v verbose,-f named file.
  • tar -tvf demo.tar: print table of content of the tar file.
  • tar -xvf demo.tar: extract the given tar file.

Apply compression algorithm on archive files (gzip bzip2)

  • gzip demo.tar: To compress a given file, demo.tar.gz will be created.
  • gunzip demo.tar.gz: To decompress a given compressed file.
  • bzip2 and bunzip2 as a second alternative to compress and decompress files.
  • Archiving and compressing (gzip) in one command: using the -z option, tar -czvf demo.tar file1 file2 dir1
  • Decompressing (gunzip) and extracting the archive in one command: using the -x option, tar -xvzf demo.tar.gz

grep command


  • Globally search a regular expression and print it, global regular expression print, global regular expression parser.
  • locate and find help to find required files and directories, grep command help to find content within the file.
  • Syntax: grep <pattern> filname, and it will print all matched lines.
  • grep someword filname1.txt filname2.txt: look for the pattern in the given files.
  • grep someword *: look for the pattern in the current directory and not in the sub-directories, to ignore case we should use the -i option, --color to display the result in a colored form.
  • grep someword -c *: to print how much time the pattern is found.
  • grep someword -n *: to print the lines number where the pattern is found.
  • grep someword -l *: to print only the file names where the pattern is found.
  • grep someword -v *: to print the lines where the pattern is not found.
  • grep -w 'word1' *: will search for the exact word, aword100 will not be valid, only word1 will be a correct answer.
  • Display before, after and surrounding lines including search result: -A means After, -B means Before, -C means Before and After. grep 'word1' -B3 *
  • Search multiple contents in a file: egrep '(word1|word2)' *grep can understand only some patterns; for this reason we use egrep
  • grep -o 'word1' *: will print only matched pattern.
  • egrep -R '(word1|word2)' *: will execute recursively, it means also checking the sub-directories.

Regular expressions patterns


Character Patterns

  • grep 'd*' *: display all lines which contains d followed by any number of characters.
  • grep 'c[aei]b' *: all lines wich contains cabceb or cib
  • grep 'b..x' *. means any character, it will search for all 4 letter words beginning with b and ending by x.

Word Patterns

  • grep 'linux' demo.txt: display all lines which contains the word linux, for example linuxworld will be valid.
  • grep '\<linux\>' demo.txt: display all lines which contains exactly linux, so linuxworld is invalid.
  • grep '\<linux' demo.txt: display all lines which starts with linux.
  • grep 'linux\>' demo.txt: display all lines which ends with linux.

Line Patterns

  • ^: line starts with, grep '^d' demo.txt will print all lines that start with d.
  • $: line ends with.
  • \<the\>: lines that start exactly with the word the
  • ^[aeiu]: lines that start with aei or u
  • ^[^aeiu]: lines that do not start with aei or u
  • ^....$: lines that contains only 4 characters.
  • ^\.: lines that starts with the . symbol.
  • ^$: blank lines, to remove blank lines, grep '^$' file.txt > temp.txtmv temp.txt file.txt.

Additional Patterns available only when using the egrep command

  • |: it matches any of the passed string, egrep '(linux|java|docker)' *.
  • {m}: it matches exactly m occurrences of the preceding character. egrep '[0-9]{10}' * will match every 10-digit mobile numbers.
  • {m, n}: it matches minimum m occurrences and maximum occurrences of the preceding character. egrep '[0-9]{8, 10}' * will match every 8-digit, 9-digit and 10-digit mobile numbers. {m, } means no restriction for the maximum occurrences.

The cut command


  • we use the cut command to extract specific data from files, the file can be either normal or tabular.
  • cut -c 9 emp.dat: print the 9th character in each line.
  • cut -c 3-5 emp.dat: print the 3th4th5th character in each line, -c 3- the 3th to the last character, -c -3 will print the 3 first characters by each line.
  • cut -c 3-5 7-9 emp.dat: will display the 3th4th5th characters and also the 7th8th and 9th
  • Display content based on delimiter: name|username|password, to print the second column we can use cut -d "|" -f 3 tabularfiled option for delimiter and f for field, -f 1-3 to get the columns 1, 2 and 3, -f -2 to get up to the second column, -f 1, 3 to get only the first and third column, to get all columns except the third column, we can use cut -d "|" --complement -f 3 tabularfile

The Linux File Permissions Concept


File persmmisions describe the allowed operations by various users.

  • User categories.
  • Permission types.
  • Operations allowed related permissions.

User categories

  • All users are divided into 4 types:
    • User/owner: represented by u, the user who created the file.
    • Group: represented by g.
    • Others: represented by o.
    • All: represented by a.

Permission types

  • r: Read, on files we can view the content, on directories we can view the content of the directorties; for example ls command.
  • w: Write, on files we can edit the content of files, on directories we can create and delete files.
  • x: Execute, on files we can execute the file just like a program, on directories we can enter into a directory; for example using cd command. Keep in mind to execute a file, both read permission and execute permission are required.
  • -: No permission.
  • By default, the owner of the file get only read and write permissions.
  • Without execute permission, read and write permissions on directories are useless.

Operations allowed related permissions

  • +: add a partcular permission to a user|group|others|all.
  • -: remove a partcular permission from a user|group|others|all.
  • -: assign a set of permissions to a user|group|others|all.
  • chmod command:
    • It means change mode. we can use it to change files or directories permission.
    • Syntax: chmod <user_category><permission>, example: chmod u+x script.sh, gives execute permission to the owner on the given file.
    • chmod u+x,g+w,o-r demo.txt: gives the owner execute permission, gives write permission to the group and remove read permission from others.
    • chmod u=rw,g=rw,o=r demo.txt: we assing read and write permission to the owner and the group, the read opertion to others.
    • chmod a=- demo.txt: we retrieve all permissions from all.
    • chmod a=rwx demo.txt: we give all permissions to all.
    • r, w, x are symbolic permissions, we can also specify permissions by using octal numbers.
      • 0000 means No permission.
      • 1001 means only execute permission.
      • 2010 means only write permission.
      • 3011 means write and execute permission.
      • 4100 means only read permission.
      • 5101 means only read and execute permission.
      • 6110 means only read and write permission.
      • 7111 means read, write and execute permission.
      • Note that 4 means read permmision, 2 means write and 1 means execute.
      • Example:
        • chmod u=rw,g=wx,o=w demo.txt could be also written as chmod 632 demo.txt.
        • chmod 77 demo.txt: is the same thing as chmod 077 demo.txt.
        • chmod demo.txt is not allowed.

Users and groups

  • sudo addgroup coursegroup: to create a group with name coursegroup, we can check if the group was added by checking the following file etc/group.
  • sudo adduser --ingroup coursegroup user1: to create a new user in the group coursegroup. We can check if the user was succefuly added by checking the following file etc/passwd.
  • sudo adduser user2: when executing this command, a new user will be created user2, a group with the same name will also be created, and the new user will be added to that group.
  • su - user1: change current linux user to user1.

Umask

  • umask means user mask. Based on the umask value, default permissions will be there for files and directories. The default umask value is 022.
  • Default permissions for files: 666– umask value, 666022=644, it means read and write for the user, read for the group and the others.
  • Default permissions for directories: 777– umask value, 755022=644, it means read, write and execute for the user, read and execute for the group and the others.
  • To congigure ou own umask value, by using the umask command: umask 002.
  • Se case:
    • For newly created files, default permission should be 444, what should be the mask value: 444=666-umask=222.
    • The most used umask values are:
      • 022644 on files means read and write for the user, read for the group and others.
      • 002664 on files means read and write for the user and the group, read for the others.
      • 077600 on files means read for the user, nothing for the group and others.
      • 007660 on files means read and write for the user and the group, nothing for others.

Working with editors


  • There are multiple editors: graphical editor (gedit), vi editor, nano editor…
  • gedit: same as windows note pad.
    • gedit file1.txt: if the file exist already, it will be opened for editing, otherwise a new file will be created and opened for editing. Note that gedit can work only on the desktop version.
  • vi or visual editor: it is a unix based editor, on the contrary to gedit and nano that are linux based. We can use it to create new files and to edit the contrent of the existing ones.
    • vi file2.txt: if the file exist already, it will be opened for editing, otherwise a new file will be created and opened for editing.
    • To save the file: :wq (w means write, q means quit).
    • There are 3 mode of vi:
      • Command mode: It is the default mode, in this mode you can use any vi command, from command mode we can enter into insert mode by using multiple ways, for example we can use i. To enter into exit mode we press : key.
      • Insert mode: In this mode we can insert/modify/append data, to return to the command mode we can press Esc key.
      • Exit mode: To insert the exit mode from the command mode, we press :, in this mode we can exit the vi editor. To return to the command mode we press Esc
    • How to insert/append data when we are in the command mode, we can press the following keys:
      • A: To append data at the end of the line.
      • I: To isert data at the beginning of the line.
      • a: To append data to the right side of the cursor position (just after the cursor position).
      • i: To insert data to the left side of the cursor position (just before the cursor position).
    • To delete data when we are in the command mode, we can press the following keys:
      • Delete characters and words:
        • x: To delete the current character.
        • nx: To delete the n characters starting from the current position.
        • X: To delete the previous charachter.
        • nX: To delete the n previous charachters.
        • dw: To delete the current word.
        • ndw: To delete the n words starting from the current position.
      • Delete lines:
        • dd: To delete the current line.
        • ndd: To delete n lines.
        • d$: To delete from the current position to the end of line.
        • d^: To delete from the beginning to the current position.
        • dgg: To delete from the beginning of the file to the current position.
        • dG: To delete from the current position to the end of the file.
    • To replace data when we are in the command mode, we can press the following keys:
      • r: To replace the current character.
      • R: To replace multiple charatcers from the current position.
      • S or cc: To replace a line.
      • O: To open a new line above te cursor position.
      • o: To open a new line below te cursor position.
    • To copy/paste data when we are in the command mode, we can press the following keys:
      • yy: To copy one line, y for yanking.
      • nyy: To copy n lines.
      • yw: To copy a word.
      • nyw: To copy n words.
      • y$: To copy from current cursor to the end of line.
      • y^: To copy from the beginning of the line to the cursor position.
      • p: small p, To paste below the cursor position.
      • P: cpaital P, To paste above the cursor position.
    • Navigation when we are in the command mode:
      • k: up.
      • j: bottom.
      • l: right.
      • h: left.
      • Arrow keys can replace the above keys.
      • $: go to the end of the current line.
      • ^: go to the beginning of the current line.
      • H: go to the beginning of the current page.
      • M: go to the middle of the current page.
      • L: go to the end of the current page.
      • b: backword, back to the beginning of the word.
      • nb: back to the beginning of the nth word.
      • e: end of the currentrword.
      • ne: end of the third word.
      • w: forward, forwarding to the beginning of the next word.
      • nw: forward, forwarding to the beginning of the nth next word.
      • G: move to the last line.
      • ctrl+f: page down.
      • ctrl+b: page up.
      • u: undo operatio, to undo the previous operation.
    • Exit mode commands:
      • :w: save file data.
      • :wq: save file data and quit from the editor.
      • :q: quit from the editor.
      • :q!: ignore changes and quit from the editor.
      • :set nu: set line numbers on the editor.
      • :set nonu: to remove line numbers from the editor.
      • :n: place the cursor to the nth line.
      • :$: place the cursor to the laste line.
      • :!<unix_command>: to execute any unix command, :!ls:!date, …

Checking Resource Usage


  • free -m (memory usage)
  • htop (cpu usage, live mode)
  • df -mh (disk space usage)
  • uptime (how long the system has been running)

Package Management(ubuntu)


  • apt update: update repository(no instalation)
  • apt list: list all packages, with --upgradable (list only the packages that can be upgraded)
  • apt search packageName: to search a given package.
  • apt install packageName: to install a given package.
  • apt remove packageName: to remove a given package.
  • apt autoremove: to remove packages no longer needed/used.
  • apt upgrade: to upgrades all local packages.
  • apt dist-upgrade: to upgrades all local packages, even installing new ones or removing existing packeges more info.

Managing systemd units


  • systemctl status jenkins: systemctl is included in systemd, the preceding command check the status of the jenkins service.
  • systemctl disable jenkins: do not start jenkins automatically after the boot.
  • systemctl enable jenkins: start jenkins automatically after the boot.
  • systemctl stop jenkins: stop the service jenkins.
  • systemctl start jenkins: start jenkins.
  • systemctl restart jenkins: restart jenkins.

Viewing Logs


  • cat var/log/syslog: print the content of the syslog file. The var/log directory contains logs of different services.
  • tail -n 20 var/log/syslog: print the last 20 lines from the syslog file.
  • tail -f var/log/syslog: print the last 10 lines from the syslog file, and show any live modification to the output.
  • journalctl -u jenkins: print only the jenkins service logs, note that we can use also the -f option.
  • cat var/log/syslog | grep jenkins print logs that contains the jenkins keyword.

Managing users


  • cat etc/passwd: to print the list of users with details like uisgid, default bash…
  • cat etc/shadow: to check the hashed password of the users.
  • cat etc/group: to print the list of groups.
  • adduser batman: to add a new user batman.
  • su - batman: switch to the batman user, logout to come back to the main user.
  • passwd: to change the password of the current user.
  • sudo passwd brahim: to change the password of brahim.
  • sudo userdel -r brahim: to delete the user brahim, also -r will delete the home directory of brahim.
  • sudo groupadd players: to add playrs as a new group.
  • groups: to print the groups where the current user is member.
  • groups zaki: to print the groups where the user zaki is member.
  • sudo usermod -aG players brahim: to add brahim to the players group.
  • sudo gpasswd -d zaki players: to remove the user zaki form the group players.
  • sudo groupdel players: to delete the grous players.

SSH


  • print actives network connections: netstat -tulpn, for pid informations, use sudo netstat -tulpn
  • print where the ssh program is lcoated: which sshd
  • check if ssh is running: systemctl status ssh
  • to connect to a remote machine via ssh: ssh username@1.2.3.4 where 1.2.3.4 is the ip adress of the remote machine, the user will be asked to enter the password of the guven user. it is recommended to disallow password authentication and root access to the machine, a better way is to use public/private key authentication.
  • ssh -i publickey.pem username@1.2.3.4 to log in using ssh with the given public key.
  • ssh -v -i publickey.pem username@1.2.3.4 to log in using ssh with the given public key in debug mode, we get more details on the terminal, with that debug mode, if there is some issue to connect via ssh, the output can be helpful.
  • ssh -p 33 username@1.2.3.4: to connect via ssh on a customized port, normally the default port is 22.
  • default configuration of ssh in the server: /etc/ssh/sshd_config.
  • to check the logs of authentication to the server: /var/log/auth.log.
  • on the client side, when connecting via ssh to a remote server, a folder named .ssh will be created and will contains the following files:
    • known_hosts: hosts that the client already connected to via ssh.
    • id_rsa: private key.
    • id_rsa.pub: public key.
  • ssh server configuration:
    • the config file to edit is /etc/sshd_config.
    • we can edit some configurations like port number, prohibit root login, no user/password authentication….
    • We can allow only some users or groups to connect via ssh with AllowsUsers groupname user1 user2
  • keys generation:
    • ssh-key-gen: to generate new keys, by default they will be placed inside .ssh folder, id_rsa as default name of the private key, id_rsa.pub as default public key name, the private key should be kept private.
  • ssh-copy-id -i /.ssh/id_rsa.pub username@1.2.3.4: to copy the public to the remote ssh server, with that further logging will succeed without passing password or publickey.
  • ssh-copy-id will copy our public key to the remote /.ssh/authorized_keys file in the ssh server. In the authorized_keys file, every line represent a client ssh public key.

Bash history


  • history: to print history of typed commands, every command will have an number associated. we can run the command with number 10 by typing !10.
  • to not save typed commands on the history, we type a space before typing the command name.

Process control commands


background/foreground mode

  • run a process in background mode: man ls &, the man program will run in the background mode, we can also send a process to bakground by typing ctrl+z when it is in a running mode.
  • to see how many jobs we have (background running process): jobs
  • to return a background process to the foreground: fg without parameters, automatically the job with + sign will be brought to the background, we can also pass the identifier of a process.

ps command:

  • ps -e: print all running processes.
  • ps -f: print all running processes with full details.
  • ps -aef | grep ls: to get info about a specific process.
  • ps -p 1: to get info about a process based on a given PID.
  • ps -u zaki: to get info about a process based on a given UID.

top command:

  • top command will display the complete active process, cpu and memory information, shift+p will order the output based on cpu usage, shift+m will order based on memory usage, c to print the absolute path of the process, k to kill a process.
  • top -u zaki: print only processes owned by a given user.

kill command:

  • kill 1245 kill the process with the pid 1234, you should be able to kill a process only if youa are the owner of that process.
  • kill -9 5678: forcing to kill the process with sigkill signal.
  • killall vim: killing all the vim process.

Network linux commands


  • hostname: to print the name of the host, with -d option it prints the domain name which the machine belongs to, with -f option, it prints the fully qualified domaine name, with –i, it prints the ip address of the machine.
  • ping: used to cchechk the establishement of the connection and also the speed of that connection, ping www.google.fr.
  • netstat: list detailed information about connection to and from the host, with -t option we list only tcp connections, with u only udp connections and with -l we list only listening ports, with -p we list also the pid/program name information.
  • netstat -c: display information continiously (live)
  • netstat -r: display the kernel routing table.
  • netstat -ap | grep ssh: check on which port ssh is running
  • netstat -s: print connection statistics.
  • netstat -lx: print all unix listening ports.
  • netstat -i: show network interfaces, with -e, more extended output.
  • ifconfig -a: view all network configuration & setting.
  • ifconfig eth0: to view specific network setting.
  • ifconfig eth0 up: enabling eth0 interface.
  • ifconfig eth0 down: disabling eth0 interface.
  • nslookup: to get ip address from dns name or vise-virsa: nslookup www.google.fr.
  • traceroute: utility to view the number of hopes to reach a given host.

man page


  • man is the builtin help system in Linux, man is short for manual, a man page is a documentation.
  • man man: to see the man page of the man command.
  • when on the man page we can type:
    • h for displaying help, q to quit.
    • y to go backward one line.
    • b backward one window.
    • g go to the first line.
    • G go to the last line.
  • searching :
    • by typing /keyword: search the keyword forward in the page.
    • by typing ?keyword: search the keyword backward in the page, type n to go to next found keyword, N in the reverse order.
  • ls man page:
    • Synopsis: ls [OPTION]... [FILE]... : means that we can pass many optional option and many optional file to the ls command.
  • man -k ls: search the manual page name for the keyword ls.
  • man 2 mkdir: open the sectioné of the man page of the mkdir command.
  • The man-pages manual describes the convention that should be employed when writing man pages.
  • some shell commands(built-in shell commands) don’t have a specific man page, we can use help to get more details, for example help while to get details about the while shell keyword.
  • we can get brief help for a given command by passing --help option, ls --help will output details about the ls command.

curl command


  • curl is a tool to transfert data between from or to server using one of the supported protocols.
  • curl www.google.fr
  • curl -i www.google.fr: extra information (headers).
  • curl -d “data inside the body post request” www.google.fr: post request.
  • curl -X PUT -d “data inside the body post request” www.google.fr: put request.
  • curl -X DELETE google.fr: to send a delet request.
  • curl -u username:password needauthentication.com.
  • curl -o resp1.txt www.google.fr : to save the response received to the resp1 file.

Stackoverflow questions


Shell scripting What is Shell?


  • Shell is responsible to read the command provided by the user.
  • Shell will check if the command is valid or not.
  • Shell will check if the command is properly used or not.
  • If everything is ok, the shell will interpret that command into kernel understandable form and pass that command to the kernel. The kernel is responsible to excute that command with the help of the hardware.

Types of Shell?


  • Bourne Shell: was developed by Stephen Bourne, it came with the first version of Unix. By using sh command we can access this Shell.
  • Bash Shell: Bash for Bourne Again SHell. It is an advanced version of Bourne Shell (sh), this is the default Shell for the most linux distributions, by using bash command, we can access this Shell.
  • Korn Shell: developed by David Korn, mostly used in IBM AIX OS, by using ksh command, we can access this Shell.
  • CshellTShell,ZShell, …

Scripts and Shell info


  • To cretae a script, we can create a file named myscript.sh, we can add commands into it.
  • To execute the script, we should give the user execute permission on the given script, and we run the script using ./myscript.sh, note that wen executing the preceding command , the defaut shell will be executed (Bash Shell), to use Bourne Shell for example, we can type sh ./myscript.sh.
  • How to check default shell in our sysrtem: echo $0, or echo $SHELL, not that LINUX IS CASE SENSITIVE, we can also check for this info by checking the following file cat /etc/paswd; this file contains the configuration of the system users.
  • How to check all available shell in our system: cat /etc/shells
  • To switch from one shell to another: we type sh to switch to the sh shell.
  • What is a shell script: a sequence of commands and programming features saved in a file. the programming features like control statement, loops, …
  • Executing a script using the bash shell: /bin/bash /home/zaki/myscript.sh, we can use directly the bash command if it can be found inside a location that exists in $PATH variables. If the myscript.sh file exists in the same directory as the user, we can run it with bash ./myscript.sh or ./myscript.sh if the bash is the default Shell in the system. If we want to use the sh Shell, sh ./myscript.sh
  • Inside a script file, we can specify the interpreter, which is responsible to execute the script. We can do that using Sha-Bang #!, for example: adding #! /bin/sh in the first file of the script to specifiy the sh shell.
  • By default, for any command or script, the OS will check locations specified by the $PATH varaiable. So if we add our scipt location to path variable locations, we can then run the script from anywhere just like a command.
  • EXPORT PATH=$PATH:/home/user/scripts: will append the new location (/home/user/scripts) to the PATH variables, not that the result of this operation is temporary and available only at session level.
  • To set the PATH permanently at a user level: we should add EXPORT PATH=$PATH:/home/user/scripts to the hidden file .bashrc, so whenever a new session begin (session means a new command line terminal opening operation), the .bashrc file will executed and the path will be adjusted.

Variables in Shell programming


  • In Shell programming, we don’t have the notion of data types, every value is text or String type.
  • All variavbles are divided into two type:
    • Predifined variables (environement variables).
    • User defined variables

Predifined variables (environement variables)

  • They are mostly used by the system internally, but we can use them in our script. we can get all environement variables either using env or set
echo "user Name: $USER"
echo "USER HOME directory: $HOME"

User defined variables

variables name
  • It is recommanded to use only upper case characters.
  • If a variable name is composed by multple words, it should be separated with the _ symbol.
  • Should not begin with digit.
  • We should not use special symbols: -@#$, …

readonly A: To define a variable A as readonly.

variable scopes

  • There are 3 scopes available for variables:
    • Session scope.
    • User scope.
    • System scope.
  • Session scope: variables declared in the terminal are said to be session scope, once we close the terminal, all those variables are gone.
  • User scope: for each user, there is a special file .bashrc, variables declared inside this file are available for that user in every session. user scoped variables are not available to other users.
  • System scope: the variables are availbe to all users of the system. To declare system scoped variable, we can declare them inside the following file /etc/profile. Example: export global_variable=globalValue, to validate the changes of the system scope variables we should restart the OS.

Variable substitution

  • Accessing the value of a variable by using the $ symbol is called variable substitution.
  • Syntax: $variableName or ${variableName}herewecanappendsomething
  • Example:
name=zaki
echo $name
echo "${name}"
  • Note that echo $name will not work as expected, and the variable substitution will not work.

Command substitution

  • Command substitution means execute command and substitute its result.
  • Example:
currentDate=$(date)
echo $currentDate

Or using the old way

currentDate=`date`
echo $currentDate
  • We use command substitution whenever we need to use the result of a linux command. Example: echo "the current working directory is $(pwd)"

Command Line arguments

  • To run a specified script with arguments: ./script.sh hello world, It is friday
  • inside the script, we can get:
    • The numbers of arguments: $#
    • The script name: $0
    • The first argument: $1
    • The second argument: $2
    • All arguments: $* or $@.
    • With $@, we get space between each two argments $1 $2, with $* we get special character between each two characters $1c$2, the c is the first character in the IFS(Internal Field Separator), the default is tha space charcter.
    • To check default IFS: set | grep "IFS"
  • echo -n "APPLE" | wc -c Vs echo "APPLE" | wc -c
  • Example using variable/command substituion and a command line arguments
echo "hello"
len=$(echo -n $1 | wc -c)
echo "the lengh of the given parameter is $len"

Read dynamic data from the user

  • Without prompt message: read a b, to read two values from the user.
  • With prompt message:
echo "Enter the first value"
read A
echo "Enter the second value"
read B
echo "A is $A"
echo "B is $B"
  • In place of the above code, we can use the following approach: read -p "enter the first value:" A, the entered value will be stored to the A variable.
  • The -p for displaying message, -s to hide the user input. read -s -p "enter the password value:" pass, note that the -s should be provided before the -p option.
  • Write a script that read a file name from the user, and remove bank lines from that file?
read -p "Enter the file name:" filename
grep -v "^$" $filename > temp.txt
mv temp.txt $filename
  • Write a script that read a file name from the user, and remove duplicate lines from that file?
read -p "Enter the file name:" filename
sort -u $filename > temp.txt
mv temp.txt $filename

Operators

  • Arithmetc operators: + - * / %
  • Relational operators (numeric comparison operators), they return boolean value:
    • gt: greatet than.
    • ge: greatet than or equal.
    • lt: less than.
    • let: less than or equal.
    • et: equal to.
    • gt: Not equal to.
  • Logical operators:
    • a: Logival AND
    • o: Logival OR
    • !: Logival NOT

How to perform matematical operations

  • There are multiple ways to perform matematical operations.
  • By using expr keyword
    • sum= expr $a + $n
    • sum=$(expr $a + $n)
    • Both the $ symbol and the space are mandatory.
  • By using let keyword
    • Most used approach.
    • let sum=$a+$b
    • The $ symbol is optional, we should not put space before or after the + symbol.
  • By using (())
    • sum=$(($a + $n))
    • Both the $ symbol and the space are optional.
  • By using []
    • sum=$[$a + $n]
    • Both the $ symbol and the space are optional.
  • All the above approaches will work only for integer arithmetic operations, for floating point calculations we should use the bc command.
  • bc command for binary calulator. From the terminal we can start using the bc command.
read -p "enter the first floating number" a
read -p "enter the second floating number" b
sum=$(echo $a+$b | bc)
echo "the sum is $sum"
  • Write a script to sum read 4 given digits:
read -p "Enter any 4 digits without space" n
a=$(echo $n | cut -c 1)
b=$(echo $n | cut -c 2)
c=$(echo $n | cut -c 3)
d=$(echo $n | cut -c 4)
sum=$[a+b+c+d]
echo "the sum is $sum"

Control statements

if statement
  • Simple if statement:
  • Syntax:
        if [ condition ]
        then
          action
        fi
  • Example: read a name from the user
        read -p "enter your name :" name
        if [ $name = "Sunny" ]
        then
          echo "Your name is Sunny"
        fi
  • Note that the instruction a=b is an assignement and a = b is a comparison instruction, also the spaces between [] are mandatory.
  • if else statement:
  • Syntax:
        if [ condition ]
        then
          action1
        else
          action2
        fi
  • Example: read a name from the user
        read -p "enter your name :" name
        if [ $name = "Sunny" ]
        then
          echo "Your name is Sunny"
        else
          echo "Your name is not Sunny"
        fi
  • Nested if statement:
  • Syntax:
        if [ condition1 ]
        then
          if [ nestedcondition ]
          then
            action1
          else
            action2
        else
          action3
        fi
  • Ladder if statement:
  • Syntax:
        if [ condition1 ]
        then
          action1
        elif [ condition2 ]
        then
          action2
        elif [ condition3 ]
        then
          action3
        fi
  • Example: read a name from the user
        read -p "enter your name :" name
        if [ $name = "Sunny" ]
        then
          echo "Your name is Sunny"
        elif [ $name = "Bunny" ]
        then
          echo "Your name is Bunny"
        else
          echo "Your name is unknown"
        fi
  • Question1: Write a script to find the greater of two numbers:
  read -p "Enter the first number" a
  read -p "Enter the second number" b
  if [ $a -gt $b ]
  then
    echo "the first number is greater than the second"
  elif [ $a -lt $b ]
  then
    echo "the second number is greater than the first"  
  else
    echo "the first number is equal to the the second"  
  fi
  • Question2: Write a script to find the greanter of three numbers:
  read -p "Enter the first number" a
  read -p "Enter the second number" b
  read -p "Enter the third number" c
  if [ $a -gt $b -a $a -gt $c ]
  then
    echo "the first number is the biggest"
  elif [ $b -gt $c ]
  then
    echo "the second number is the biggest"  
  else
    echo "the third number is the biggest"  
  fi  
  • Note the use of the -a in the first condition, it perfroms the AND logical operation, -o for the logical or operation, ! for logical not.
  • Question3: Write a script to find if a given number is even or not:
  read -p "Enter the first number" a
  if [ $[a%2] -eq 0 ]
  then
    echo "Yes it is even"
  else
    echo "No, it is not"  
  fi  
Exit codes / Status codes
  • Every command return some value after execution, This returned value is the exit code or status code, we can find exit code by using the $ symbol.
  • The status code value have a range between 0 and 255, the value 0 means that the command/script executed successfuly, otherwise not.
  • Example:
  rm inexistentfile
  echo $?
  • The above code will print 1, as the command rm didn’t execute successfuly.
File test options
  • The -e option return true if a given file/directory exists.
  • Script to test if a file exists in the current working directory:
 read -p "enter file name to test" name
 if [ -e $name ]
   echo "exists"
 else  
   echo "does not exist"
 fi  
  • The -f option returns true if a given file is a regular file.
  • The -d option returns true if a given file is a directory.
  • The -s option returns true if a given file is not empty, -s for size.
  • The -l option returns true if a given file is a link file.
  • The -b option returns true if a given file is a block special file.
  • The -c option returns true if a given file is a character special file.
  • The -r option returns true if the current user has read permiision on the given file.
  • The -w option returns true if the current user has write permiision on the given file.
  • The -x option returns true if the current user has execute permiision on the given file.
  • The -o option returns true if the current user is the owner of the given file.
  • file1 -ot file2 returns true if file1 is oltder than file2 (last modified time).
  • file1 -nt file2 returns true if file1 is newer than file2 (last modified time).
  • The -z option returns true if a given string is empty.
  • Examples:
    • Write a script to test whether a given file is regular file or directory:
#! /bin/bash
read -p "enter file name to test" name
if [ -e $name ]
then
  if [ -f $name ]
  then
    echo "it exists and it is a regular file"
  elif [-d $name ]  
    echo "it exists and it is is a directory"
  else
    echo "it is something else"
  fi  
else
  echo "does not exist"
fi  
  • Write script to comapre two files
  read -p "enter the first file name" file1
  read -p "enter the second file name" file2
  result=$(cmp $file1 $file2)
  if [ -z "$result" ] then
    echo "they are same"
  else  
    echo "they are not"
  fi
String test options
  • str1 = str2 returns true if both strings are equal.
  • str1 != str2 returns true if both strings are not equal.
  • -z str1 returns true if str1 is empty.
  • -n str1 returns true if str1 is not empty.
  • str1 > str2 returns true if str1 is alphabitically greater than str2.
  • str1 < str2 returns true if str1 is alphabitically lesser than str2.
  • Note that the symbol < is also a redirection symbol, so we should use the backslach str1 \< str2
  • [ 0 ][ 1 ] returns true all the time.
  • Script to test if the logged user is root or not and provide 5 current rnning processes information:
user=$(whoami)
if [ $user = "root" ]
  echo "you are root"
  ps -ef | head -6
else
  echo "you are not root"

ps for process status, -e every process, -f full listing.

case statement
  • Syntax:
  case $variable in 
    option1)
          action-1
          ;;
    option2 )
          action-2
          ;;
    optionn)
          action-n
          ;;
    *)
          default action
          ;;
  esac 
  • The space in option2 ) is optional, ;; is used to break from the case statement, it is optional in the default option. It is recommended if you use the default statement to put it as a last step in the case statement.
  • Write a script to read a digit from 0 to 5 and print its string equivalent in the console using the case statement:
 #! /bin/bash
 read -p "enter any digit from 0 to 9" digit 
 case $digit
  0)
    echo "zero"
    ;;
  1)
    echo "one"
    ;;
  2)
    echo "two"
    ;;
  3)
    echo "three"
    ;;
  4)
    echo "four"
    ;;
  5)
    echo "five"
    ;;
  *)
    echo "please enter a number between 0 and five"
 esac   
  • For string case statement:
    case $String
      "AL")
        echo "zero"
        ;;
      "BL")
        echo "one"
        ;;
  • Using regular expressions:
    case $Character
      [abc])
        echo "a, b or c"
        ;;
      [^abc])
        echo "any character except a and b and c"
        ;;
      [a-z])
        echo "any lower case character"
        ;;
      [A-Z])
        echo "any upper case character"
        ;;
      [a-zA-Z])
        echo "any alphabet character"
        ;;
      [0-9])
        echo "any digit from 0 to 9"
        ;;
      [a-zA-Z0-9])
        echo "any alphanumeric character"
        ;;
      [^a-zA-Z0-9])
        echo "any character except alphanumeric character"
        ;;            
while loop
  • Syntax
  while [ condition ]
  do
    body
  done  
  • Write a script to print numbers from 0 to 9:
   #! /bin/bash
   i=0   
   while [ $i -le 9 ]
   do
     echo $i
     sleep 2          
     let i++
   done  

let i++ could also be written as i=$[i+1]sleep 2 for sleeping two seconds.

  • Write a script to display time every second:
 while [ true ] 
 do
   clear
   echo -e "\n\n\n\n$(date +%H:%M:%S)"
   sleep 1
 done

The -e option to process the scape characters like \n, we can replace echo -e by printf "\n\n\n\n$(date +%H:%M:%S)"

To put the cursor in some position, we can use the following command tput cup x y where x and y are the row and the column numbers respectivly.

break and continue
  • To break the loop execution
  i=1
  while [ $i -le 10 ] 
  do
    if [ $i -eq 5 ]
    then 
      break
    fi  
    clear
    echo "$i"
    let i++
  done
  • To skip the current iteration and continue for the next iteration
  i=1
  while [ $i -le 10 ] 
  do
    if [ $i -eq 5 ]
    then 
      continue
    fi  
    clear
    echo "$i"
    let i++
  done
  • Write a script to reverse a given string
  read -p "please enter a String: " word
  length=$(echo -n $word | wc -c)
  output=""
  while [ $length -ge 1 ]
  do
    lastcharcter=$(echo -n word | cut -c $length)
    output=$output$lastcharacter
    let length--
  done
  echo "the reversed character $output"  
  • Write a script to read a file line by line
  read -p "please enter the file name: " fname
  if [ ! -e $fname ]
  then
    echo "the file dosn't exist"    
    exit 1
  fi  
  if [ -d $fname ]
  then
    echo "it is a directory, and not a file"    
    exit 2
  fi
  if [ ! -s $fname ]
  then
    echo "it is empty"    
    exit 3
  fi
  count=1
  while read line
  do
    echo "   $count  $line"
    let count++
  done < $fname
for loop
  • Syntax
   for variable in item1 item2 item3... itemN
   do
     action
   done
  • Print numbers from 1 to five
   for i in 1 2 3 4 5
   do
     echo "$i"
   done  
  • Print every number from 1 to 100 that is divisible by 10
   for i in {1..100}
   do
     if [ $[i%10] -eq 0 ]
     then
       echo "$i"
     fi  
   done  
  • Display all file names present in the current working directory
  #! /bin/bash
  for fname in *
  do
    if [ -f $fname ]
    then
      echo "$fname"
    fi  
  done
  • Diplay the command line arguments passed to the script
  #! /bin/bash
  if [ $# -eq 0 ]
  then
    echo "no arguments"
    exit 1
  fi  
  for arg in $@
  do
    echo "$arg"
  done
  • Diplay the third value of each row of the following file
    100:anny:1250:A
    200:bnny:2250:B
    300:snny:3250:C
    #! /bin/bash
    for row in $(cat abovefile.txt)
    do
      value=$(echo $row | cut -d ":" -f 3)
      echo value
    done
  • alternative syntax of for loop
    #! /bin/bash
    for ((i=0; $i<5; i++))
    do
      echo $i
    done 
  • Inside the expression ((…)), the spaces and the $ symbol are not required.
  • Write a script to test if a number is prime
    #! /bin/bash
    read -p "enter a number grater than 1" n
    counter=0
    for ((i=2; $i<=n/2; i++))
    do
      if [ $[n%i] -eq 0 ]
      then
        echo "it is not a prime number"
        exit 1
      fi  
    done  
    echo "it is a prime number"

Arrays concept

How to create arrays
  • If we know elements at the beginning: courses=("java" "shell" "linux")
  • If we don’t know elements at the beginning: courses[0]="java"courses[1]="shell", we are not required to declare the courses array separately, but we can do it using the follwing code declare -a courses. The index values need not be consecutive, we can take randomly.
How to access arrays
  • We can access array elements by using index which is 0 based.
  • echo ${courses[0]} will print the first element of the array.
  • echo ${courses[@]} will print all elements of the array with space separator.
  • echo ${courses[*]} will print all elements of the array spaced with IFS.
  • echo ${!courses[@]} will print all indices where elements are available.
  • echo ${#courses[@]} will print the number of elements present inside the array.
  • echo ${#courses[0]} will print the length of the first element.
Examples
  • Write a script to create an array with some elements and print all its elements by using while loop, for loop and the alternative for loop
    • Using while loop
      #! /bin/bash
      declare -a fruits
      fruits={"A", "B", "C"}
      index=0
      while [ index -lt ${#fruits[@]} ]
      do
        echo ${fruits[$index]} 
        let index++
      done  
  • Using for loop
      #! /bin/bash
      declare -a fruits
      fruits={"A", "B", "C"}
      for fruit in ${fruits[@]}
      do
        echo ${fruit} 
      done  
  • Using alternative for loop
      #! /bin/bash
      declare -a fruits
      fruits={"A", "B", "C"}
      for((index=0; index<${#fruits[@]; index++))
      do
        echo ${fruits[index]} 
      done  

Shell Script Functions

How to define a function
  • First way
  function f1()
  {
      a=$1
      b=$2
      echo "you passed $a and $b" 
  } 
  • We call the function by the flllowing command f1 5 6
  • The function keyword is optional, if we need to pass parameters, it is not possible to declare them in the function signature, but we can access them using $1 for the first parameter, $2 second, …
  • Write a function that print all passed parameters:
#! /bin/bash
shellfunction()
{
    if [ $# -eq 0 ]
    then
      echo "please you should pass parameter to the function"
    else
      for param in $@
      do
        echo $param
      done
    fi    
}
# calling the above function
shellfunction 1 254 4
  • Write a function that print the factorial of a given integer number:
#! /bin/bash
factorial(){
  if [ $1 -le 0 ]
  then
    echo "please enter a positive integer"
  else
    let factorial=1;
    for ((let i=2; i<$1; i++))  
    do
      factorial=$factorial*i
    done
    echo $factorial
}
read -p "enter any positive integer" n
factorial $n
Varibale scopes
  • The below examples are valid programs, by default every declared variable has a global scope, even those declared inside a function. Before using a variable, it should be declared already.
  #! /bin/bash
  f1(){
    echo $x
  }
  x=10
  f1
  #! /bin/bash
  f1(){
    x=10
  }
  echo $x
  f1
  • If we want to declare a variable local to a function, we use the local keyword.
    #! /bin/bash
    f1(){
      local x=10
    }
    f1
    echo $x  #no access to the x variable, en empty value will be printed.
    #! /bin/bash
    f1(){
      local x=10
    }
    f1
    x=20
    echo $x   #will print 20.
Return statement
  • Every function in shell scripting returns some value, the dafault return value is the exit code/status code of the last command inside function. But based on our requirement, we can return values explicitly by using the return statement return <exit_code>, the allowed values are in the range [0, 255]. we can get the return value of the function bu using $?.
  • The difference between exit and return in context of a function, the exit will end the program, but return will only returns from the function and the program can continue executing.
break vs exit vs return
  • break: we can use break only inside loops to break loop execution.
  • exit: we can use exit anywhere, to terminate script execution.
  • return: we can use return only inside function to stop the execution of the function.
How to call functions present in another script

utilistyscript.sh

      #! /bin/bash
      add()
      {
        echo "$1 + $2= $[$1+$2]"
      }

main.sh

      #! /bin/bash
      . ./utilistyscript.sh 
      add 5 10

the line . ./utilistyscript.sh for including the script utilistyscript.sh, the script should be placed in the given path.

Various topics

  • Hox to copy files over ssh: awnser