*nix Command Line and Shell (Bash) Programming

*nix Command Line and Shell (Bash) Programming


These notes are for general command line usage and shell programming.
There is a separate notes document for developer tools on the command line.


-------------------- Command Line --------------------


*nix systems' file structure is always a single tree with a root (/) directory, unlike Windows file
systems that have each physical hard-drive be a different drive, therefore a different tree.

You type commands into a command line using a shell. A shell is simply the program you
use on the command line to enter commands to be run by the system. There are many different
shells in the *nix environment. The most popular Linux shell is Bash, and it is the default
shell used when you open a command line terminal. The four basic shells are:
    Bash    (Bourne Again Shell)
    sh      (Bourne Shell)
    csh     (C Shell)
    ksh     (Korn Shell)
The Shell Scripting Notes further down in this document are for Bash scripting.

Most programs are installed in the /usr/bin directory.

Most commands operate like this:        command -options arguments


Command Completion:
    Pressing tab once will complete a command or filename if it is the only command or filename that
    starts with the letters you've already typed.
    If you can't remember exactly what a certain command is, if you can at least start the command,
    then just press tab twice and all commands with those starting letters will be output.

- vs -- options:
    Options for a command can come in a shortened form using -option or in long form using --option.
    This shorthand "-" version is usually used on the command line itself, but for shell scripting
    it is better to using the long form "--" version because then it will be more clear what the
    option is when the script is viewed later on.

File Names
    File names that being with a period are hidden.
    *nix systems have no concept of file name extensions, the programs used probably do care.
    *nix systems allow spaces and special characters in file names but you should not use spaces (use
    underscores instead) and you should limit the special characters to periods, dashes, and underscores.

A note on paths:
    When you type in a command without specifying a path name then the system does not search the
    entire computer, it only searches a list of directories it maintains where executable files are
    kept. If it does not find the program in those directories, it will issue the "command not found"
    error message. The list of directories is called your path and is contained in the Path Variable.
    You can view the directories in your path with "echo $PATH", which outputs a colon-separated
    list of the path names contains in your path.
    You can add directories to your path with the command:      export PATH=$PATH:the_directory
    But a better way is to edit your .bash_profile or .profile file to include the above command, then
    the command will be done automatically every time you log in.
    Most Linux distributions encourage a practice of having a bin directory as a subdirectory of your
    home directory, and putting all the programs you personally use in there. So you can create this
    bin directory in your home directory, then add that directory to your path in the .bash_profile
    file. Now any shell scripts you put in there will be executed just by typing in the name of the
    program at the command prompt, no matter what directory you are in.
    When a command is entered, the built-in commands are checked first to find a match, if
    no match if found in the built-in commands then the PATH environment variable is checked
    for the command. If the command isn't found there then the command canonot be executed
    on the machine.


File System
    /       root directory
    /boot   in linux the linux kernel and boot loader files are kept here, kernel is vmlinuz
    /etc    contains configuration files for the system, all file in here are text files.
            /etc/passwd - contains essential information for each user, this is where users
                          are defined. User information is shown in the following columns:
                          username, password (encyrpted, displayed as just an "x"), UserID,
                          GroupID, GECOS (usually the person's full name or a description of
                          what the account is for), home folder location, and the shell.
            /etc/group - contains essential information for each group on the system, this is
                         where groups are defined. Group information is shown in the following
                         columns:   name of group, group password (encrypted, displayed as just
                         an "x"), GroupID, extra member of the group (separated by commas).
            /etc/shadow - this is where the encrypted passwords for all users are held. They
                          used to be shown in /etc/passwd, but now they are in this file for
                          more security (this file isn't even accessible to normal users, only
                          to the root).
            /etc/fstab - contains a table of devices that get mounted when your system boots
                         and defines the disk drives.
            /etc/hosts - lists the network host names and IP addresses that are intrinsically
                         known to the system.
            /etc/init.d - contains the scripts that start various system services typically
                          at boot time.
    /bin, /usr/bin      these two directories contain most of the programs for the system.
                        The /bin directory has the essential programs the system requires to
                        operate. The /usr/bin directory contains applications for the
                        system's users.
    /sbin, /usr/sbin    the sbin directories contain programs for system administration,
                        mostly for use by the superuser.
    /usr    contains a variety of things that support user applications.
            /usr/share/x11 - supports files for the X windows system.
            /usr/share/dict - dictionaries for the spelling checker (commands: look and aspell).
            /usr/share/doc - various documentation files in a variety of formats.
            /usr/share/man - the man pages are kept here
            /usr/src - source code files. If you installed the kernel source code package, you
                       will find the entire linux kernel source code here (in linux machines).
    /usr/local      this directory and its sub-directories are used for the installation of
                    software and other files for use on the local machine. So software that
                    is not part of the official distribution usually goes here (as opposed to
                    /usr/bin). When you install programs on the system they should be installed
                    in one of the /usr/local directories, usually the /usr/local/bin directory.
    /var    contains files that change as the system is running
            /var/log - contains log files, these are updated as the system runs. You can view
                       the files in here from time to time to monitor the health of the system.
            /var/spool - holds files that are queued for some process, such as mail messages
                         or print jobs.
    /lib    the shared libraries (similar to DLLs in Microsoft Windows).
    /home   this is where users keep their personal work. In general this is the only place
            users are allowed to write files.
    /root   this is the superuser's home directory.
    /tmp    a directory in which programs can write their temporary files
    /dev    a special directory because it doesn't contain files in the usual sense. It contains
            devices that are available to the system. On *nix systems devices are treating like
            files. You can read and write devices as though they were files.
    /proc   this directory is also special in that it does not contain files, in fact it doesn't
            really exist at all, the entire directory is virtual. It contains little peep holes
            into the kernel itself. There are a group of numbered entries in this directory
            that correspond to all the processes running on the system. (I don't think this is
            on Macs, or it is named differently).
    /media, /mnt    the /media directory is normal directory used in a special way, it is used
                    for mount points. The process of attaching a device to the directory tree
                    is called mounting. For a device to be available it must first be
                    mounted. And on systems that require manual mounting of removable devices,
                    the /mnt directory provides a convenient place for mounting these
                    temporary devices.


Wildcards
    Wildcards can be used when specifying filenames, they allow you to select filenames based on
    patterns of characters so that you can rapidly select groups of filenames instead of typing
    out every single separate file you want to do something with. You can use wildcards with any
    command that accepts filename arguments.

    *               matches any characters, i.e. *.html refers to all html files in a directory
    ?               matches any single character
    [characters]    matches any character that is a member of the set of characters. The set of
                    characters can also be expressed as a POSIX character class as one of the
                    following:
                                    [:alnum:]       alphanumeric characters
                                    [:alpha:]       alphabetic characters
                                    [:digit:]       numerals
                                    [:upper:]       uppercase alphabetic characters
                                    [:lower:]       lowercase alphbetic characters
    [!characters]   matches any character that is not a member of the set of characters

    Some examples of using wildcards:
        *           all filenames
        g*          all filenames the begin with a "g"
        b*.txt      all filenames that beging with a "g" and end with ".txt"
        Data???     any filenames that beings with the characters "Data" followed by exactly three
                    more characters
        [abc]*      any filename that begins with an "a", "b", or "c"
        [[:upper:]]*    any filename that beings with an uppercase letter
        BACKUP.[[:digit:]][[:digit:]]   any filename that begins with the characters "BACKUP"
                                        followed by exactly two numerals
        *[![:lower:]]   any file name that doesn not end with a lowercase letter


Commands
    Commands can be one of 4 different kinds:
        1. An executable program
            like all those file in /usr/bin. Within this category, programs
            can be compiles binaries such as programs written in C and C++ or programs written
            in scripting languages such as the shell, Perl, Python, etc.
        2. A command built into the shell itself
            Bash provides a number of commands internally called shell builtins. The cd command,
            for example, is a shell builtin.
        3. A shell function
            These are miniature shell scripts incorporated into the environment.
        4. An alias
            Commands that you can define yourself, built from other commands.


Text Editors
    In Unix-like systems there are several text editors that are generally used. Nano or pico is simple
    text editor that is good if you don't want to have to learn a bunch of stuff to use the text editor.
    Vi or Vim (Vi iMproved) is a popular choice that comes installed in all versions of Unix-like systems
    and it has tons of keyboard shortcuts to help editing go faster. However Vi or Vim is very hard for
    beginners and you have to read up on it before you even try to start using it. Sames goes for Emacs,
    which is the other really popular and powerful text editor on *nix systems.
    To start editing or create a file using one of these text editors just type the name of the text
    editor followed by the name of the file you want to edit or create.

Changing the command line prompt
    Can change the command line prompt in Bash by changing the PS1 variable.
    i.e.    PS1="new_prompt"
    Just type that into the command line and press enter.
    You can see what the current value of the prompt is by typing: echo $PS1

SuperUser (the Root)
    The SuperUser for a system, called the Root, is a user who can do anything on the system. The
    Root cannot be locked out of any files or directories. The Root has ultimate control of the
    system. The Root's home directory is /root. A normal user can switch to being the Root with
    the command "su", but you have to know the Root's password. The "sudo" command is used instead
    of "su" usually because it is safer than "su". The sudo command only lets a user run a single
    command as root at a time, and the System Admins can restrict what root-only commands are
    allowed for certain people. Only System Admins should have Root access. No normal user should
    have Root access because the Root can do anything and therefore a user with access could
    destroy important data or cripple the system.
    The Root always has a UID of 0. You can effectively duplicate the root account just by setting
    the UID of any other account to 0 in /etc/passwd - this is a way to hack the system.
    Some Unix systems have a special group called wheel, only group members of wheel may use 'su'
    to Root and use the Root capabilities.



----COMMANDS----

whoami
    Outputs the name of the current user.

who
    Shows a list of everyone that is currently on the system (along with some other information).
        Syntax:         who

hostname
    Outputs the name of the system you are using.

pwd
    Print working directory.


ls
    List file and directories in the current directory.
    You can list the contents of multiple directories at once
    ls -l                   <- list the files in the working directory in long format
    ls pathname pathname    <- list contents of multiple directories
    ls -a                   <- list all files (including hidden files)
    ls -la                  <- list all files in long format
    ls -F                   <- identifies directories from files by ending directory names with a "/"
    Long format shows the file permissions, the owner, the group, the size in bytes,
    modification time, and file name.
    When a file's first character in its file perissions is "l" it means that the file is a
    symbolic link. Also the filename will show the file (symbolic link) points ("->") to another
    file (the file that is run by the symbolic link).
    ls -l:
        The ls -l command displays 9 columns of information: the permissions, number of links,
        owner of the file, what group the file belongs to, sizes in bytes, month file was created,
        day file was created, time file was created, name of the file.
        The different possible file types that are displayed by the first character in the
        permissions column are as follow:
                -       plain file
                d       directory
                l       link
                p       pipe
                b       block device            // device files that read info in blocks
                c       character device        // device files that read info in a stream


cd
    Change the current directory.
    Syntax:     cd pathname
    cd, cd ~, cd ~userName      <- these change the working directory to the home directory
    cd -                        <- this changes the working directory to the previous one
    The pathname can be an absolute or relative pathname.

pushd
    Changes the working directory, saving the current directory to a directory stack so that
    you can return to it using the popd command.
    This is what you use instead of cd if you just need to go to another directory to do
    something then you want to come back to your current directory.
        Syntax:     pushd directoryPath

popd
    Changes the working directory back to whatever directory was last pushed to the stack
    using pushd.
        Syntax:     popd

touch
    Creates an empty file
    Syntax:     touch filename

more
    View text files starting from the top of the file. Supports paging which means if the
    text is larger than the console window you can press a button to move up or down in the
    output document. Press "q" to leave the contents that less has displayed and get back to
    the cursor on the command line. Similar to "less" as shown below, but less is more
    popular to use because it has more functions.
    Syntax:     more file_name


less
    View text files starting from the top of the file. Supports paging which means if the
    text is larger than the console window you can press a button to move up or down in the
    output document. Can also use the arrow keys to move by line up and down in the document.
    Press "q" to leave the contents that less has displayed and get back to the cursor on
    the command line.
    Syntax:     less file_name
    Controlling less:
        Spacebar to scroll down by one full page
        "b" to scroll up by one full page
        Enter to scroll down by one line
        "q" to exit less
        "G" to go to end of the text file
        "1G" to go to beginning of the text file
        "/characters" to search for sequence of characters from current point to end of file
        "n" to repeat the previous search
        "h" to display a complete list of less commands and options


file
    Classify a files contents. Tells you what kind of data a file contains.
    Syntax:     file filename
    Output of a python file for instance will be:
            filename: a path script text executable
    Some file types are:
        ASCII text
        Bourne-Again shell script text
        HTML document text
        Zip arhive data
        JPEG image data
        GNU tar archive
        etc.

ln
    Creates a link to another file.
    There are two kinds of links, hard and soft (or symbolic).
    The ln command takes two arguments, first argument is the file to link to, second argument
    is the name of the new file being linked to the original file.

    Symbolic links are useful for say a file that the system uses but the system expects it to
    always have the same filename but say there are different versions of whatever thing it is,
    and the symbolic link will make it so even if the version changes it will always have the same
    filename that the system can use, which points to the correct version. This would be done by
    just updating what file the link points to if a new version is to be used. To make a symbolic
    link you just include the -s option.
            Syntax for symbolic link:       ln -s file1 file2

    Hard links make an exact clone of the file, rather than just pointing to a file somewhere else
    on the system.
            Syntax for hard link:           ln file1 file2

    Differences between hard and symbolic links are as follows. Hard link won't show the "l"
    when you use ls -l, because it doesn't know it's a link. A hard link won't take up any
    disk space. Hard disk can't span filesystems (can't be used across partitions). A symbolic
    link does show the "l" identifier when displayed with ls -l, a symbolic link does take up
    space (albeit very little space since it is just a link to a file), and a symbolic link
    can be used in other partitions on the system.



cat
    Outputs contents of a file to the command line. "cat" stands for concatenate. Can output
    one or more files to the console window by listing one or more arguments.

cp
    Copies files and directories.
    Can be used to copy a file in the same directory, or to copy files to other directories.
    Syntax:
                cp file1 file2              copies file1 to a new filename (file2).
                cp file ... directory       copies multiple files (and/or directories) to a
                                            different directory.
    Examples of using cp:
                cp file1 file2              copies contents of file1 into file2, if file2 does
                                            not exist, it is created.
                cp -i file1 file2           same as above except the -i (interactive) option means
                                            that if file2 already exists, the user is prompted
                                            before it is overwritten with the contents of file1.
                cp file1 dir1               copies the contents of file1 into a file named file1
                                            inside of directory dir1.
                cp -R dir1 dir2             copies the contents of directory dir1, if dir2 does
                                            not exsit then it is created. Otherwise it creates a
                                            directory named dir1 within dir2.


mv
    Moves or renames files and directories depending on how it is used.
    The reason it can move or rename files is because you need to think about it as renaming
    an entire path, not just the filename. So you can rename the path in a way to move the file
    to a different directory while keeping the same file name, oryou can rename the path by
    keep everything the same except the last part of the path (the filename) which will just
    rename the file and not move it.
    When moving you don't have to put directory/filename if you aren't changing the filename,
    Unix knows what is a directory and what isn't so if your second argument is just a path to
    a directory it will just move the file there and keep the same filename.
    Syntax to rename a file:                                        mv filename1 filename2
    Syntax to move files/directories to a different directory:      mv file ... directory
    Examples of using mv:
                mv file1 file2              if file2 does not exist then file1 is renamed file2,
                                            if file2 does exist then its contents are replaced
                                            with the contents of file1.
                mv -i file1 file2           same as above except the -i option means that if file2
                                            exists then the user will be prompted before it is
                                            overwritten with the contents of file1.
                mv file1 file2 file3 dir1   file1,file2,and file3 are moved to directory dir1, if
                                            dir1 does not exists mv will exit with an error.
                mv dir1 dir2                if dir2 does not exist then dir1 is renamed dir2,
                                            otherwise the directory dir1 is moved within dir2.


rm
    Removes (deletes) files and directories.
    To delete directories you have to use the -r option, which stands for recursive and will
    recursively delete all files and sub-directories within the specified directory. If the
    directory is empty though you don't need to use the -r option.
    Syntax to delete a file:            rm file ...
    Syntax to delete a directory:       rm -r directory
    Examples of using rm:
                rm file1 file2              deletes file1 and file2.
                rm -i file1 file2           the -i (interactive) option gives the user a prompt
                                            before each filed is deleted
                rm -r dir1 dir2             directories dir1 and dir2 are deleted along with all
                                            their contents
                rm *~                       delete all files in the current working directory that
                                            end with the "~" character. Some applications create
                                            backup files using this naming scheme, using this
                                            command will clean them out of a directory.

    Tip for using rm: You can do massive damage to the system by deleting stuff on accident,
    especially when using wildcards. One way to safeguard yourself is to construct the command
    with ls before you use rm, this way you can see all the things listed that will be deleted
    with the command.


mkdir
    Creates one or more directories.
        Syntax:                 mkdir directory ...
    Using the -p option and specifying a whole path will make directories in the path that
    don't exist yet.


type
    Displays information about command type.
    A shell builtin command that displays the kind of command the shell will execute, given
    a particular command name.
    Syntax:
                type command
    Examples of using type:
                type type               output: type is a shell builtin
                type ls                 output: ls is aliased to ls --color=tty
                type cp                 output: cp is /bin/cp


which
    Locates a command.
    Syntax:
                which command
    Example of using which:
                which ls                output: /bin/ls


help
    Displays reference page for shell builtin commands.
    The help command is bash's built-in help facility that is available for each of the shell
    builtins. You can add the -m option to change the format of the output (I think -m doesn't
    work on Macs).

    In the description of a command, when it shows syntax for a command, square brackets refer
    to optional items, and a vertical bar "|" indicuates mutually exclusive items.

    Many executable programs support a "--help" joption that displays a description of the
    command's supported syntax and options (This maybe doesn't work on Macs, but when you type
    incorrect options for a command it will print out an error message and then that command's
    usage syntax, so that works basically the same way).


man
    Displays an on-line command reference.
    Most executable program intended for command line use provide a formal documentation called
    a manual or man page. On most systems the "man" command uses the "less" program to display
    the man page.
    Syntax:                 man program             <- program is the name of the command

    Other documentation:
    Many software packages installed on a system have documentation files residing in the
    /usr/share/doc directory. Most of them are stored in plain text, some in html. If the
    files have a .gz file extension that means they have been compressed with the gzip
    compression program. The gzip package includes a special version of less called zless
    that will display the contents of gzip-compressed text files.

df
    View the partitions on the system. "df" stands for "disk filesystems".
    Partitions slice up the harddrive into pieces.
    Using the -h option specifies "human readable" format, which will display sizes in megabytes
    or gigabytes instead of kilobytes.
        Syntax:     df -h

groups
    Outputs what groups the current user belongs to.

cal
    Displays a calendar in the console window.

apropos
    Used to find commands when you aren't sure of the spelling.
    Takes a word as an argument and will output all available commands related to that word or
    with that word in their spelling. Also outputs a short description of what each command does.
        Syntax:     apropos word

grep
    Used to search for a particular string in a file and output all instances it finds to the
    console window. Searches standard input if no files are given as arguments.
    "grep" stands for Global Regular Expression Print.
        Syntax:     grep string file

wc
    Outputs newline, word, and byte counts for each file listed.
        Syntax:     wc file
        Result:      lineCount wordCount charCount filename
        Result ex:   5 28 107 filename
    Various options to just print out a single count.
        -c prints byte count
        -m prints character count
        -l prints line count
        -w prints word count

find
    Search for files in a directory hierarchy (including the specified directory and all its
    subdirectories).
    Can search by name or type of item in the directory. Probably in some other ways too, not sure.
        Syntax:     find directory searchType searchItem options
        Example:    find . -name "*.txt"
        Result:     Will display all files in the working directory that end in .txt
        Example:    find /home -type -f | wc -l
        Result:     Will output the number of files in the /home directory and all its
                    sub-directories.

export
    Change value of environment variables.
        Syntax:     export envVar="new value"
    Note that you don't include the $ before the envVar when using this command.

unset
    Remove an environment variable from the environment.
        Sytnax:     unset envVar

xargs
    Build and execute command lines from standard input.
        NEED TO READ ABOUT THIS.

passwd
    Allows a user to change their password.
        Syntax:     passwd




I/O Redirection
    Most command line programs that display their results do so by sending them to standard
    output, which displays on screen, but you can redirect standard output to a file using
    the ">" character to overwrite the file, or ">>" to append to the file. If the file
    doesn't exist it will be created.

    >
        Redirects stdout to a file, overwrites the file.
        Syntax:         command > filename
        i.e.            ls > myfile.txt

    >>
        Redirects stdout to a file, appends to the file.
        Syntax:         command >> filename
        i.e.            ls >> myfile.txt


    To redirect standard input from a file instead of from the keyboard use the "<" character.

    <
        Redirects stdin from a file.
        Syntax:         command < myfile.txt
        i.e.            sort < myfile.txt
                        sort < filelist.txt > sorted_filelist.txt

        In the last example, sort takes the contents of filelist.txt as its arguments by
        redirecting stdin to come from that file, and then it redirects stdout to print to
        the second file instead of printing to the screen. The order of the redirection does
        not matter, but the redirection operators must appear after any other options and
        arguments in the commnand.


File Descriptors
    Unix-like systems can support up to 10 file descriptors, numbered 0 to 9. But generally you
    just use the first three:
            Standard Input      STDIN       0
            Standard Output     STDOUT      1
            Standard Error      STDERR      2
    When you use a redirect like > there is an implied 1 there like so: 1>.
    Since > has an implied file descriptor of 1 (STDOUT), in order to redirect STDERR you need
    to make it explicit that you want to redirect it by putting in the file descriptor 2, like so:
            command argument 2> filename
    In the above syntax example if the command causes an error it will redirect the error message
    to the specified file.

    Anytime a file receives more information (from adding text to a file, reading command line
    arguments, or having information piped to a command) STDIN is used.
    Any command that can take a file as an argument can also take its input by STDIN. This is
    because all information received by a file/command is STDIN; if a command takes a file as an
    argument then that file is sent to the command as STDIN, so instead of using a file you can
    just use straight up STDIN, like echo or cat something into the command by piping it in. Piping
    something takes STDOUT from the left operand of the pipe and sends it through STDIN to the
    right operand of the pipe.
        i.e.    wc file                 // this works
                echo "yo man" | wc      // this works too
                cat file | wc           // so does this

    Combining File Descriptors:
        Would do this if you wanted to save boththe input and output to a file.
            i.e.    cat file.txt > file2.txt 2>&1           // combine STDERR and STDOUT
        As shown in the example above you put the combined file redirect at the end of the
        command and use a & on the final one I guess. This takes the output of file.txt, redirects
        it into file2.txt, but the 2>&1 means that all STDERR will be redirected to STDOUT as well.
        So if any errors pop up they will be saved to the file. So if there are no errors with
        "cat file.txt" then that file will be saved to file2.txt, but if there are errors then the
        error message(s) will be saved to file2.txt.
        Note that using 'cat' as shown in the above example is known as a Useless Use of Cat
        because using the redirct > will automatically take the contents of file.txt as STDOUT
        and put them in file2.txt as STDIN, so no need for a cat.

    Disabling all Output:
        Sometimes when you run a command you don't want to see any output (no STDERR or STDOUT).
        To do this you can redirect everything to a special device known as /dev/null.
                Syntax:         file.txt > /dev/null 2>&1
        /dev/null is like a black hole, anything that goes into it vanishes from the system.


Pipelines
    The most useful and powerful thing you can do with I/O redirection is to connect multiple
    commands together with pipelines, a pipeline is the "|" pipe bar character. With pipelines
    the standard output of one command is fed into the standard input of another.

    |
        Connects multiple commands together so the standard output of a command is fed as
        standard input into the next command.
        Syntax:         command | command | command
        i.e.        ls -l | less                      // this lets any command have scrolling
                                                      // output using the less program.
                    ls -lt | head                   // displays the 10 newest files in the
                                                    // current directory.
                    du | sort -nr                   // displays a list of directories and how
                                                    // much space they consume, sorted from
                                                    // largest to smallest
                    find . -type f -print | wc -l   // displays the total number of files in
                                                    // the current working directory and all
                                                    // of its subdirectories
                    ls -l | grep b                  // will take the output of "ls -l" and send
                                                    // it to "grep b" which will only output
                                                    // lines from the "ls -l" output that contain
                                                    // the letter "b"


Filters
    Filters are a kind of program frequently used in pipelines. Filters take standard input
    and perform an operation upon it and send the results to standard output. The can be
    combined to process information in powerful ways. Some common programs that can act as
    filters:
        sort        Sorts lines of a text file then outputs the result on stdout.
        uniq        Given a sorted stream of data from stdin, it removes duplicate lines of data.
        grep        Examines each line of data it receives from stdin and outputs every line that
                    contains a specified pattern of characters. Basically a search.
        fmt         Reads text from stdin, then outputs formatted text (neat paragraphs) to stdout.
        pr          Takes text input from stdin and splits the data into pages with page breaks,
                    headers, and footers in preparation for printing.
        head        Outputs the first few lines of its input. Useful for getting the header of a file.
        tail        Outputs the last few lines of its input. Useful for things like getting the most
                    recent entries from a log file.
        tr          Translates characters. Can be used to perform tasks such as upper/lowercase
                    conversions or changing line termination characters from one type to another (for
                    example, converting DOS text files into Unix style text files).
        sed         Stream editor. Can perform more sophisticated text translations than tr.
        awk         An entire programming langauge designed for constructing filters. Extremely powerful.

        lpr         Accepts stdin and sends it to the printer.
                    i.e.
                        cat text.txt | fmt | pr | lpr       // Formats, paginates and prints text.txt
                        cat unsorted.txt | sort | uniq | pr | lpr   // Sorts, removes duplicates,
                                                                    // paginates, and prints unsorted.txt

        tar         A gzipped tar file is a traditional Unix style tape archive file (created with tar)
                    that has been compressed with gzip, it will have file extnesions like ".tar.gz" or
                    ".tgz". Use the tar command in this example to iew the directory of such a file on
                    a Linux system.
                    i.e.            tar tzvf file.tar.gz | less


Expansion
    Expansion is the process of the shell performing several processes upon the text before it carries
    out the command, like for instance using the * wildcard. Expansion is when you type something and
    it is expanded into something else before the shell acts upon it.

    An example:
    The "echo" command outputs whatever is sent as an argument to it. So "echo my name is todd" outputs
    "my name is todd" to the terminal. But for "echo *" the shell expands the * character to the names
    of all the files in the current working directory, so echo never sees a "*", it only sees the
    expanded content. Likewise "echo c*" would list all the files/directories in the current working
    directory that start with "c".
    echo *s         <-- would print out all files/directories in current directory ending with an "s".

    Pathname Expansion is the mechanism by which wildcards work.
    
    Tilde Expansion is when you use a tile "~" to refer to (and gets expanded to) the name of the
    home directory.
        i.e.        echo ~tkrone *          // this outputs all directories/files in my home directory

    Arithmetic Expansion is how the shell allows arithmetic to be performed on the shell prompt, in the
    form $((expression)).
        i.e.        echo $((3+4))           // this outputs:  7
    Arithmetic expansion only supports integers. Its operators are:    +   -   *   /   %   **

    Brace Expansion is a way to create multiple text strings from a pattern containing braces. Patterns
    to be brace expanded may contains a leading portion called a preamble and a trailing portion called
    a postscript. The brace expansion itself may contain either a comma-separated list of strings, or
    a range of integers or single characters, but it can't contain any embedded whitespace. When using
    echo for a brace expansion each string will be separated by a space.
    Note in the examples, ".." is used to set a range (inclusive). Notice in the last example you can
    also nest brace expansions.
        i.e.    
            echo Preamble-{A,B,C}-Postscript
                    // outputs:     Preamble-A-Postscript Preamble-B-Postscript Preamble-C-Postscript
            echo Number_{1..3}
                    // outputs:     Number_1 Number_2 Number_3
            echo {Z..T}
                    // outputs:     Z Y X W V U T
            echo a{A{1,2},B{3,4}}b
                    // outputs:     aA1b aA2b aB3b aB4b
    The most common application of brace expansions is to make lists of files or directories to be
    created. For example, if you wanted to make a directory of photos split up into sub-directories by
    month and year you could do something like this:
            mkdir 0{1..9}/{2010..2013} {10..12}/{2010..2013}

    Parameter Expansion is basically the use of variables. So it is much more useful in shell scripts
    than just straight up on the Command Line. Parameter expansion expands a variable name into
    whatever value is held by the variable. To use it you just put a $ in front of a variable name.
    Syntax:         $varName
    There are some environment variables built into the system/shell, like $USER refers to the user
    name. So for me the command:
            echo $USER          // outputs:    tkrone
    The environment variable $HOSTNAME contains the hostname for the system.
    To see a list of all environment variables use either of these commands:
                printenv | less         or          env | less
    If you use a variable that doesn't exist with the echo command it will just print a blank line.

    Command Substitution:
    Command substitution allows us to use the output of a command as an expansion.
    Syntax:         $(command)
    So this allows you to package the output of some command as the argument for some other command.
        i.e.    echo $(ls)              // expands the output of ls
                ls -l $(which cp)       // will print the long form of the /bin/cp command
    You can do any command like this, for example you could do some complex command with pipes in it.
    An alternate syntax for command substitution in older shell programs, which is also supported in
    Bash, is to use back-quotes instead of $().
    Alternate Syntax:           `command`                   // instead of $(command)


    Quoting:
    Without using any quotes the echo command removes and extra whitespace and will automatically
    expand anything with a $ before it into a variable, and if that variable doesn't exist it doesn't
    print anything for it.
    For example:       echo This is a $100.00 dollar bill.
                                                            outputs: This is a 00.00 dollar bill.
            The Fix:   echo 'This is a $100.00 dollar bill.'
                                                            outputs: This is a $100.00 dollar bill.
    The above case is true because the shell treats $1 as a variable, which doesn't exist, and so it
    prints nothing for it.

    Using quotes is the fix to the above mentioned problems. Using quotes will make it so any extra
    whitespace is output, and it will make the shell see anything following a $ as part of a string
    instead of as a variable. Note that for the variable example double-quotes will not work, only
    single-quotes will fix it.

    Double Quotes cause most special characters used inside the shell to lose their special meaning and
    are treating as normal characters.
    The exceptions are:     $       \       `           <-- dollar sign, backslash, back-quote
    This means word-splitting, pathname expansion, tilde expansion, and brace expansion are suppresed,
    but parameter expansion, arithmetic expansion, and command substitution are still carried out.
        echo $(cal)     // will print out the whole calendar without any newlines because echo would
                        // treat each space (including newlines_) in the command substitution as
                        // separate arguments.
        echo "$(cal)"   // this will print out the calendar as the command cal normally does because
                        // echo only gets sent one argument, which is the output of $(cal).

    Single Quotes suppress all expansions so that exactly what is written is output in an echo
    statement.

    Escape characters, using the backslash \, can be used when you only want to suppress a single
    special character. Often this is done inside double quotes.
        i.e.    echo "This is \$100.00 dollars."
    The above example would print the $1 instead of thinking it is a variable, due to escaping the $
    with the backslash.
    Escape characters can also be used to suppress the special meaning of a character in a file name.
    Because it's possible to use characters in a file name that has a special meaning in the shell,
    you would want to escape any such characters when typing out that file on the command line.
    Such characters include "$", "!", "&", " ", etc.

    The Backslash characters can also be used to make the shell ignore a newline character, by that I
    mean it allows a single command to flow onto multiple lines. For this to work the newline must
    occur immediately after the backslash, you can't have any spaces after the backslash. The reason
    for this is because the backslash escapes whatever character immediately follows it, so if there
    is a space after it then it will just escape the space instead of the following newline
    character.
        i.e.    echo "Hey Man \
                      whats happening."              // output:      Hey Man whats happening.

    To use the standard escape sequences with the echo command you need to include the option -e.
    Common escape characters are:      \n      \t      \a      \\      \f
        i.e.    echo -e "Print an extra newline\n"

    Side note:  the -n option for echo gets rid of the automatic newline.
        i.e.    echo -n "No automatic newline at the end of this."


Permissions
    Unix-like systems are different from other operating systems in that they are multi-user, meaning
    that more than one person can be operating the computer at the same time, like remoting in via
    ssh (secure shell).
    Commands dealing with permissions are chmod, su, sudo, chown, chgrp.

    On a *nix system each file and directory is assigned access rights for the owner of the file, the
    members of a group of related users, and everybody else. Right can be assigned to read, write, and
    execute a file. The permissions for a file/directory are shown in the long form of the ls command.

    Commands:

        chmod
            Used to change the permissions of a file or directory.
                Syntax:     chmod 754 filename
            There are two ways to specify the permissions, one of them is called octal notation.
            Octal notation:     1 = execute, 2 = write, 4 = read
            The 3 different permission sets, for owner, group, and other are each given in 3 binary
            digits representing all combinations of execute, write, and read. Read, being 4, is
            controlled by the third (left-most) bit. Write, being 2, is controlled by the second (middle)
            bit. Execute, being 1, is controlled by the least-significant bit.
                    1 = 001 = execute
                    2 = 010 = write
                    3 = 011 = write,execute
                    4 = 100 = read
                    5 = 101 = read, execute
                    6 = 110 = read, write
                    7 = 111 = read, write, execute
            Permissions in the ls -l command are in the following format:  -rwx rwx rwx
            The first "-" will show the type of item it is, "d" for directory, "l" for link, etc.
            Each "r", "w", "x" will be shown if it has that permission, or a "-" will be shown wherever
            it doesn't have that permission.
            To use chmod you write the command, followed by the decimal number representing the permissions
            for each set, and then the filename.

            For directories the chmod command is used to set permissions, but the meaning of r,w,x is
            different than for files.
                r - allows contents of the directory to be listed if the "x" attribute is also set.
                w - allows files within the directory to be created, deleted, or renamed if the "x"
                    attribute is also set.
                x - allows a directory to be entered.
            Basically the options for directories each set of users is basically:
                5 = r-x     can enter directory and list the files
                7 = rwx     no restrictions on the directory
                0 = ---     no access to the directory
            chmod Examles:
                chmod u+wx filename
                chmod 755 filename
                chmod g-x filename
                chmod o+rwx filename
                chmod u+x, g-x, o-wx filename
                chmod a+x filename              // "a" here stands for all (i.e. ugo,  i.e. user,group,owner)

        su/sudo
            Allows the user to temporarily become the super user (the Root), in order to perform some
            important system administration tasks. You can also become any other user by sending a dash and
            then a username as the argument to 'su', however to do this you must first be the superuser/Root.
            su stands for substitute user.
            Putting a dash '-' after 'su' means you'll inherit all of Root's specific environment
            variables (including the Path), but with no dash you just keep your own variables.
            When using the su command you will be prompted for the super user's password.
            After executing the su command a new shell session will start as the super user. To exit
            as super user type "exit" in the command shell.
            Syntax for su:      su              // then enter the superuser's password
                                su -            // become superuser and inherit superuser's env variables
                                su - username   // become another user (when you are already the superuser)

            In some distributions, like Linux, you use the sudo command instead of the su command.
            With sudo, one or more users are granted superuser priviledges on an as needed basis. To
            execute a command as superuser, the desired command is simply preceeded with the sudo
            command. After the command is entered, the user is prompted for the user's password rather
            than the superusers password.
            The key difference between sudo and su is that using sudo never starts a new shell as the
            superuser, it merely executes a single command as the superuser, therefore the user is
            less likely to screw stuff up as the superuser and so sudo is a safer command.
            Syntax for sudo:        sudo command        // then enter user's password

        chown
            Changes the owner of a file or directory.
            Most be the superuser to change ownership of a file.
                Syntax (using sudo):     sudo chown newOwner filename
            The chown command works the same way on directories as it does on files.

        chgrp
            Changes the group ownership of a file or directory.
            Don't need to be superuser to change group ownership, but you must be the owner of the
            file or directory to change its group ownership.
                Syntax:     chgrp newGroup filename


Job control
    *nix systems are also multi-tasking, and you can manipulate the multi-tasking element of the
    operating system with the command line interface.
    Multi-tasking is achieved by switching what processes are being executed at intervals so that it
    appears that several tasks are being performed at once.
    The commands that are used to control processes are:    ps, kill, jobs, bg, fg
            ps      list the processes running on the system
            kill    send a signal to one or more processes to kill it
            jobs    an alternate way of listing your own processes
            bg      put a process in the background
            fg      put a process in the foreground
            top     displays in real time the system stats associated with each process

    Daemons
        Daemons are programs that run on the system on their own. They do a single job and just wait
        on the system until something requires them to do their job, and then they do it.

    Running a program in the background:
        You can open up graphical programs from the CLI, however, if you do then you can't use the
        command prompt until the program that was started closes, at which point you can access the
        command prompt again.
        In order to open a graphical program but not lose use of the command prompt you can either
        put "&" after the command to open the program:          command &
        or if you forget to put the "&" when opening the program on the CLI, you can press Ctrl-z when
        the terminal has focus and this will suspend the process - it still exists but is idle - at
        which point you have access to the command prompt, and simply by typing the "bg" command you
        can then move that process to the background.
            So do either:       command &
            or:     command
                    [press Ctrl-z]
                    bg
        To put the process back in the foreground (and thus giving the command prompt back over to it)
        just type the fg command into the command prompt.

    Listing the processes:
        jobs
            Use the jobs command to list all processes running in the background.
        ps
            Use the ps command to list all processes.
            The ps command by itself only lists processes launched from the current terminal. Use
            the -x option to list all processes, even those not launched from the current terminal.
                -a option shows all processes running for all users
                -x option shows even processes run by the system (daemons)
                -u option shows the user who is running the process
            Syntax:    ps -aux
            Some columns show in the ps output are the ProcessID (PID), TT which jdisplays a location
            for your terminal, STAT which stands for "state" and gives the current state of the process,
            TIME is how much processing time the system has used on the process, and COMMAND is the command
            that it is.
        top
            Shows a bunch of real-time (as in it updates itself on the screen) stats associated with
            all the processes running on the system.

    Killing a process:
        You can terminate a process (if it for example becomes unresponsive) using the kill command.
        For background processes, use the jobs command to find the job number of the process, which is
        shown between square brackets, or the process ID, which is shown after the job number. Or
        use the ps command to find the process ID (PID) for the process. Use the kill command in the
        following way:
                            kill PID                i.e. kill 25923
                    or      kill %jobNumber         i.e. kill %1

        While the kill command is used to terminate processes, its real purpose is to send signals to
        processes. Most of the time the signal is to tell the process to go away, but if can be used
        for other signals. Programs that are properly written listen for signals from the OS and respond
        to them, most often to allow some graceful method of terminating. For example, a text editor
        may listen for signals that indicate the user is logging off or the computer is shutting down.
        When it receives such a signal it save the work in progress before exiting. The kill command
        is used to send different signals to processes.

        Typing "kill -l" will give a list of all the signals the kill command supports.
        Use the different kill signals like so,
            Syntax:kill -signal           // -signal can be the signal name or number (list in kill -l)
                    i.e. kill -SIGKILL PID   or     kill -9 PID
        Some useful signals are:
            Signal 1: SIGHUP
                Hang up signal. Programs can listen for this signal and act upon it. This signal is
                sent to processes running in a terminal when you close the terminal.
            Signal 2: SIGINT
                Interrupt signal. This signal is given to processes to interrupt them. Programs can
                process this signal and act upon it. You can also issue this signal directly by typing
                Ctrl-C in the terminal window where the program is running.
            Signal 15: SIGTERM
                Termination signal. This signal is given to processes to terminate them. Programs can
                process this signal and act upon it. This is the default signal sent by the kill
                command when no signal is specified.
            Signal 9: SIGKILL
                Kill signal. This signal causes the immediate termination of the process by the kernel.
                Program cannot listen for this signal, it just automatically kills the process.

        For example, if there is a program that is hopelessly hung up and you want to kill it. First,
        use the ps command to get the PID for that process ("ps -x | grep processName"), use the
        command in the parenthesis to output only the process you are looking for. Then, use the
        SIGTERM, the default kill command, to try to kill the process ("kill -SIGTERM PID" or just
        "kill PID"), where "PID" is the processes PID. If the process doesn't terminate (it is
        ignoring the signal) then send a harsher kill signal like SIGKILL ("kill -SIGKILL PID").





-------------------- Shell Programming --------------------


Shell scripts are written to automate things. A shell script is a file with a series of shell commands.
The shell reads the file and carries out the commands as though they have been entered directly into
the command line.

Beyond basic commands, the shell provides a set of programming language features to be used in shell
scripts to allow those scripts to have the behavior of programs, like branching and looping, etc.

To write a shell script you need to do three things: write the script, give the shell permission to
execute it, and put it somewhere the shell can find it.
So first write the script, then use chmod to make it executable, then you must add the program to a
directory that is in your path, preferably by adding it to the bin directory in your home folder, as
mentioned at the beginning of these notes in the "A note on paths" section.


Basics of Shell Scripting
    Don't need to save an extension on the filename you use for the shell script.
    Do need to include a shebang for the shell, in the case a Bash, use:
                        #!/bin/bash
    The shebang is given to the shell indicating what program is used to interpret the script.
    All comments use the # character.

    To let the shell be able to see the program it must be put in a directory that is in your path.


The Environment
    There are some scripts located in your home directory that are used to configure the behavior of
    your sessions on the computer.
    During your session, the system holds a number of different facts about the world in its memory,
    this information is called the environment, which contains things like your path, user name, name
    of the file where your mail is delivered, etc.
    The "set" command lets you see what is in your environment.
    Two types of commands are often contained in the environment: aliases and shell functions.

    The environment is established when you log on to the system, the bash program starts and reas a
    series of configuration scripts called startup files. These startup files define the default
    environment shared by all users. This is followed by more startup files in your home directory
    that define your personal environment. The exact sequence depends on the type of shell session
    being started. There are two kinds of shell sessions: a login shell session and a non-login shell
    session. A login shell session is one in which we are prompted for our user name and password, 
    when you start a virtual console session, for example. A non-login shell seession typically occurs
    when you launch a terminal session in the GUI.

    Login shells read one or more startup files like:
        /etc/prorile    -   a global config script that applies to all users
        ~/.bash_profile -   a user's personal startup file, can be used to extend or override settings
                            in the global config script.
        ~/.bash_login   -   if ~/.bash_profile is not found, bash attempts to read this script.
        ~/.profile      -   if neither of the lsat two files are found, bash attempts to read this file.
                            This is the default in Debian-based distros, like Ubuntu.

    Non-login shell sessions read the following startup files:
        /etc/bash.bashrc -  a global config script that applies to all users.
        ~/.bashrc        -  a user's personal statup file. Can be used to extend of override settings
                            in the global config scripts.

    In addition to reading the startup files above, non-login shells also inherit the environment from
    their parent process, usually a login shell.


    Aliases:
        An alias is an easy way to craete a new command which acts as an abbreviation for a longer one.
        To make an alias, you put a line with the following syntax in a config file like .bash_profile:
                    alias name='value'
        Where name is the name of the new command and value is the text to be executed whenever name
        is entered on the command line. The value must be in single quotes. so name is the alias command
        while value is the original command that the alias is equivalent to.
                i.e.    alias lsl='ls -l'           // now "lsl" will call the "ls -l" command
        The alias command is just another shell builtin. You can create aliases directly at the
        command prompt, but they will only be in effect for that session.


    Shell Functions:
        Aliases are good for very simle commands, but if you want to create something more complex, you
        should use shell functions. Shell functions are scripts within scripts, or sub-scripts.
        Syntax for a shell function:
                                        functionName() {
                                            # commands...
                                        }
        The () in the shell functioon is actually a shell builtin too.
        Shell functions can be defined in a script or at the command prompt, but when defined at the
        command prompt they will only last for the current session.


Some Output stuff
    Using quotation, it is possible to embed carriage returns in text and have the echo command's
    argument span multiple lines:
        i.e.    echo "blah blah
                blah blah
                blah blah blah"
    
    To get around using quotation marks with echo, in case what is being output by echo uses quotation
    marks and you don't want to have to escape every quotation mark used, you can use a here script.

    Here Documents:
    A here script, also sometimes called a here document, is an additional form of I/O redirection.
    It provides a way to include content that will be given to the STDIN of a command.
        Syntax for a here script:
                                        command << token
                                        // content to be used as command's stdin
                                        token
        i.e. cat << _EOF_
                // content...
             _EOF_

    Using <<- instead of << will causes bash to ignore leading tabs (but not spaces) in here script.


Variables
    To declare a variable you just put the variable named, an equal sign, and its value.
    NOTE: NO SPACES ARE ALLOWED BETWEEN THE NAME, =, AND THE VALUE.
        Syntax:     varName=value
    To use a variable you have to use the "$" with the variable name.
        Syntax:     $varName
    Using the "$" character means you are telling the shell to perform parameter expansion (which is
    how you expand a parameter to gets its value).

        i.e.    yourName="Todd"
                echo $yourName

    Rules for variable naming:
        Variables names in bash must start with a letter, cannot contain and embedded spaces (use
        underscores instead), and cannot have punctuation marks.


Command Substitution and Constants
    Can use command substitution in a shell script just as you would on the command line.
            Syntax:     $(command)
    The $(command) will be expanded to the output of that command.
    You can also assign the results of a command substitution to a variable like so:
            varName=$(command)

    Constants by convention should be in all uppercase. Apparently there is some way to make a constant
    in bash programming but generally programmers just make a variable in all uppercase letters to
    signify that it should be treated as a constant.
            Constants Syntax:       CONST_NAME=value






-------------------- REFERENCE OF SHELL COMMANDS --------------------


$SHELL          shows you what shell you're in
$HISTSIZE       shows the number of commands stored in the shell's history
$HISTSIZE=number        change the max size of the history. Can't go higher than 32,767
history         shows the history of shell commands used