Unix fundamentals and_shell scripting

Editor's Notes

• #5: What is UNIX? The name was intended as a pun on an earlier system called ‘Multics’ (Multiplexed Information and Computing Service). This pun is the key to understand the acronym, since the word ‘Uniplexed’ does not mean anything in particular. (‘Multiplexed’ refers to a communications system able to carry many messages simultaneously.)
• #6: Features Multi-user :- More than one user can use the machine at a time supported via terminals (serial or network connection). Multi-tasking :- More than one program can be run at a time. Hierarchical directory structure :- To support the organization and maintenance of files. Portability :- Only the kernel ( <10%) written in assembler tools for program development a wide range of support tools (debuggers, compilers). But only drawback of UNIX is it is lacking of GUI (Graphical User Interface) and we have to type commands for each every thing.
• #7: Unix operating system originally developed in 1969 by a group of AT&T employees at Bell Labs including Ken Thomson and Dennis Ritchie. University of California enhanced the features of UNIX. Subsequently, AT&T relinquish its ownership and control over UNIX. Various companies sell their own version of UNIX.
• #8: Unix is available in market in various flavors. Different vendors provides different operating system. Solaris is given by Sun Microsystems, AIX is given by IBM. Most popular version of Unix is Linux.
• #14: What above command does? Displays calendar of current month. Displays calendar for specified month (i.e. month 7-July ) and year (i.e. year 2008). Displays calendar of 1752. Note: Valid year values are 1 to 9999
• #15: %aDisplays abbreviated weekday name. %ADisplays full weekday name. %bDisplays abbreviated month name. %BDisplays full month name. %dDisplays day of month. %DDisplays date in the format of mm/dd/yy. %g Displays year in 2 digit. %GDisplays year in 4 digit.
• #18: passwd, when invoked by ordinary user, asks for old password, after which it demands the new password twice. When u enter a password, string is encrypted by the system, and encryption is stored in file /etc/shadow.
• #19: Unix provides two calculators – a graphical object ( xcalc command) that looks like one, and bc command. The former is available in X-window System, and is quite easy to use. The other one is less friendly, extremely powerful. In fact, there are situations when working without bc becomes quite painful. $ bc scale=2 17/7 2.42 <ctrl-d>
• #20: Script command lets you “record” your login session in a file typescript in current directory. All the commands, their output and the error messages are stored in the file for later viewing. Hence, this can be used to maintain a log of your activities. script –a # append activities to existing file typescript script logfile#logs activities to file logfile Note: Some activities won’t be recorded properly, for example, commands used in vi.
• #21: Command 1 :- Clears the screen. Command 2 :- Tells you filename of the terminal you are using (tty stands for teletype). Command 3 :- know your machine name uname –r: shows operating system’s version no. Command 4 :- Prints user’s login name. Command 5 :- sign off
• #44: On Unix, the collection of data that makes up the contents of a directory or a file isn't stored under a name; the data is stored as part of a data structure called an ‘inode’. In the inode, Unix stores information about which disk blocks belong to the contents of the directory or file, as well as information about who owns the directory or file and what access permissions it has. Every directory or file on Unix has its own inode. Unix doesn't store the name of the directory or file in the inode. Every inode in the file system has a unique number, and the file system locates the contents of a directory or file strictly by its inode number.
• #46: Shell acts as the mediator which interprets the command that user gives and then conveys them to the kernel which ultimately executes them. As soon as you login, sh command becomes active. Major time spent by the shell is in waiting for input from user.
• #47: Bourne shell: Original Unix system came with Bourne shell. Created by Steve Bourne C Shell: faster than Bourne. Created by Bill Joy. Allows aliasing of commands. It also has command history feature. Korn Shell: Superset of Bourne shell. Created by David Korn of AT&T’s Bell Labs.
• #48: Two special files: /dev/null - its size always remains zero /dev/tty - refers to your own terminal
• #49: In a pipeline, command on the left of | must use standard output, and the one on right must use standard input
• #50: Unix provides a feature by which you can save the standard output in a file, as well as display it on the terminal (or pipe it to another process). It is made possible by tee command.
• #51: Unix system is controlled by a number of shell variables that are separately set by the system, some during boot sequence, and some after logging in. These variables are called system variables or environment variables. You must know significance of many of them because they determine the environment in which you work. use $ set to display complete list of these variables.
• #52: .profile is the AUTOEXEC.BAT file of Unix. .profile file must be located in your home directory, and it is executed after /etc/profile, the universal profile for all users. Universal environmental settings are kept by the administrator in /etc/profile so that they are available to all users.
• #65: If any of the special characters mentioned above themselves occur in the pattern to be searched then they must be preceeded by a backslash(\) if they are to be treated as ordinary characters. If you want to make searching irrespective of their case, <Esc>:set ignorecaseor <Esc>: set ic
• #66: :$s/director/member/g Replace director with member in last line :1,$s/director/member/iReplace director with member in whole file Sometimes you may like to selectively replace a string. In that case, add the ‘c’ (confirmatory) parameter as the flag at the end. Example: :1,5s/director/member/c - replace director with member in line 1 to 5(y/n/a/q/l/^E/^Y)? y (yes)make this changes n(no)skip the match a(all)make this change and all remaining ones without further confirmation q(quit)don't make any more changes l(last)make this change & quit ^EScoll the text one line up ^YScoll the text one line down
• #67: Use :set all to view the complete list of options available
• #70: Multiple file editors became available under DOS platform in early 90’s, whereas vi offered this facility from its early days. It permits you to load and edit several files simultaneously.
• #79: The cmp command does a character-by-character comparison and indicates whether or not two files are identical. To use cmp files should be sorted. If the two files differ, cmp generates a message indicating the first point where the two files differ. E.g., file1 file2 differ: char 4, line 10 If the two files are identical, cmp simply outputs nothing ("silence is golden"), but simply returns $ prompt.
• #80: The comm command outputs three columns: the first column identifies only those lines that are present in the first file, the second column identifies only those lines that are present in the second file, and the third column identifies those lines that are present in both files. To use comm files should be sorted. Comm can also produce selective output, using options -1, -2 or -3. $ comm -12 file1 file2 # selects only common lines
• #81: The problem with ‘cmp’ is that it is unable to distinguish whether the difference between two files is profound or superficial. A useful alternative to the ‘cmp’ command is the UNIX ‘diff’ command. The ‘diff’ command attempts to determine the minimum set of changes needed to convert one file to another file. Maintaining several versions of a file: To use diff command files should be sorted. Take previous example for file1 and file2
• #83: Process is an instance of a program in execution. Each process is uniquely identified by a number called PID ( Process Identifier ) that is alloted by the kernel when it is born. When you log in to a UNIX system, a process is immediately set up by the kernel: sh (for bourne shell), ksh(korn shell), bash (bourne again shell) Any command that you type in at the prompt is actually the standard input to the shell process. This process remains alive until you logout To know PID of current shell, echo $$
• #85: Fork: A process in Unix is created with fork system call, which creates a copy of the process that invokes it. For example, when you enter a command at the prompt, the shell first creates a copy of itself. The image is practically identical to the calling process, except for few parameters like PID. When a process is forked in this way, child gets a new PID. Forking mechanism is responsible for multiplication of processes in system. Exec: Parent then overwrites the image that it has just created with the copy of the program that has to be executed. This is done by exec system call, and parent is said to exec this process. No new process is created here; existing program is simply replaced with new program. This process has same PID as the child that was just forked. Wait: Parent then executes wait system call to keep waiting for the child process to complete. When child process has completed execution, it sends a termination signal to the parent. The parent is then free to continue with other functions.
• #86: A multitasking system lets a user do more than one job at a time. Since there can be only one job in the foreground, rest of the jobs have to run in the background.
• #87: Background jobs cease to run, however, when a user logs out. That happens because shell dies. And, when the parent dies, its children also die. Unix system permits variation in this default behaviour, using nohup command.
• #88: If you are using Korn or bash shell, you can use its job control facility to manipulate jobs. You can send a job to background, bring it back to foreground, suspend or even kill it. For this purpose, following commands are used: bg, fg, suspend, and kill fg and bg optionally use a job identifier prefixed by a % symbol. If you specify fg %1, the first job is brought to foreground. If you use fg %sort, sort job comes to foreground.
• #89: Kill, by default uses signal number 15 to terminate the process. Signal 9 is also known as sure kill signal You can also kill all processes in your own system except login shell, by using special argument 0 $ kill 0# terminate all processes with signal 15 Kill $! - kills last background job
• #91: cron lets you schedule jobs so that they can run repeatedly. It takes input from user’s crontab file. Cron process is mostly dormant, but every minute it wakes up and looks in a control file in /usr/spool/cron/crontabs) for instructions to be performed at that instant. cron is generally used by superuser to perform tasks like removing outdated files or collecting data on system performance, creating backups
• #103: grep scans a file for occurrence of a pattern, and can display the selected pattern, line numbers in which they were found, or filenames where pattern was found. Grep can also select lines not containing the pattern
• #105: Suppress the prefixing of filenames on output when multiple files are searched $grep -h gupta *
• #107: Egrep command, authored by Alfred Aho, extends grep’s matching capabilities. It offers all options of grep, but its most useful feature is the facility to specify more than one pattern for search. Each pattern is separated from the other by a | (pipe) It has –f, which allows to fetch patterns from some file. egrep -f
• #110: Designed by Lee McMohan. It is used for performing non-interactive applications
• #111: Everything in sed is an instruction. An instruction combines an address for selectiong lines with an action to be taken on them. sed options ‘address action’ file(s) sed has two ways of addressing lines: By line number By specifying a pattern which occurs in a line Option –n is used to suppress duplicate line printing
• #116: Sed’s strongest feature is substitution.
• #117: This command made a late entry into the UNIX system in 1977 to augment the tool kit with suitable report formatting capabilities. Named after its authors, Aho, Weinberger and Kernighnan, awk is unquestionable the most powerful utility for text manipulation. Unlike the other filters, it can access, transform and format individual fields in a record with the greatest of ease. It also accepts regular expressions for pattern matching, has C language-type programming constructs, variables and several built-in functions, which more than suffice to meet the daily needs of report writing. By using utility programs, advanced patterns, field separators, arithmetic statements, and other selection criteria, you can produce much more complex output. The awk language is very useful for producing reports from large amounts of raw data, such as summarizing information from the output of other utility programs .
• #118: In the line number 1 line specifier section is represented by the context address /director/. This section selects the lines that are to be processed in the action section. Moreover the print is the default action of awk there is no need to specify it if you want to print the entire line. For the purpose of the identification of fields, awk uses a contiguous string of space as the default field delimiter. This is done with –F option. Using $ prefixed variable viz. $2 $3 you can specify the field. e,g, $ awk –F “|” ‘/sales/ { print $2,$3,$5 }’ emp.lst The comma has used to delimit the field specification. These ensure that each field is separated from the other by a space. Example shown in line number2. awk also accept the line address to select lines. To select lines 3 to 6 from file filename we use following command $ awk –F”|” ‘ NR == 3, NR == 6 { print NR, $2,$3,$6}’ emp.lst
• #119: The output in the last slide is unformatted, with the width of each field determined solely by the space it occupies in the file awk is a ‘stream formatter’, and uses the C-like printf statement to beautify the output. This statement is used with suitable format specifier to control the width and justification each field should use, irrespective of the width it has in the input. %s format will be used for string data, and %d for numeric. You can now produce a list of all the agarwals: As shown in line number 1, the name and designation have been printed in space 20 and 12 characters wide, respectively, using the ‘–’ symbol to left-justify the output. You can use following characters for formatting. CharacterDescription cASCII character dDecimal integer IDecimal integer (added in POSIX) eFloating-point format ([-]d.precisione[+-]dd) E Floating-point format ([-]d.precisione[+-]dd) fFloating-point format ([-]ddd.precision) ge or f conversion, whichever is shortest, with trailing zeroes removed GE or f conversion, whichever is shortest, with trailing zeroes removed oUnsigned octal values String xUnsigned hexadecimal number. Uses a-f for 10 to 15 XUnsigned hexadecimal number. Uses A-F for 10 to 15 %Literal %
• #120: Line number 1 shows, how do you print the three fields for the directors or chairman? You can always write each awk program separate line and if you want to print only those lines for persons who are neither director nor chairman, you should use the != and && operators instead. awk offer the ~ and !~ operators to match and negate a match, respectively. You can use this feature to print out all the chowdhury’s (spell Choudhury and chowdury) and saxena’s (spelt as saxena and saksena)as shown in line number 2. awk -F "|" '$2 ~ /[cC]ho[wu]dh?ury/ || $2 ~ /sa[xk]s?ena/ { print }' emp.lst awk can also handle numbers, both integer and floating type, all the relational tests can be handled by awk with numbers, as is done in any other language. Using operators from the set shown in the following table, as shown in line number 3. OperatorSignificance < Less than <=Less than or equal to = =Equal to !=Not equal to >=Greater than equal to >Greater than ~Match a regular expression !~Does not match a regular expression
• #121: Like the shell, awk can perform computations on numbers. It uses the arithmetic operators +, -, *, /, and %(modulus) same as used by C or other programming languages. It also overcomes one of the major limitations of the shell; the inability to handle decimal numbers, example shown in line number 1. A user-defined variable used by awk has a special feature. No type declaration is needed, and it is initialized to zero or a null string, by default, depending on the type. awk has a mechanism of identifying the type of variable used from its context. We can now print a serial number, using the variable kount and apply it to those directors drawing a salary exceeding 6700. By default value of kount is 0.
• #122: The programs are getting larger, you an store into file. So if the previous program was stored in the file empawk.awk, then you need to specify the program filename with the –f option : To run the empaawk.awk file type the following command at the shell prompt. $ awk –F”|” –f empawk.awk emp.lst
• #123: awk statement are usually applied to all lines selected by the line specifier or address, and there are no addresses, then they are applied to every line of a file. If you are to print a heading, then the BEGIN section can be used quite gainfully. Similarly, if you want to print some totals after the processing is over then you should do it in the END section. The BEGIN and END sections are optional, and take the form BEGIN { action } END { action } These two sections, when present, are delimited by the body of the awk program. They also use a pair of curly braces to enclose the program. You can use these two sections to print suitable heading at the beginning, and the average salary at the end. Store the awk program in a separate file empawk2.awk, and then use the –f option. You can execute this program using command given below $ awk –F”|” –f empawk2.awk emp.lst
• #124: The previous program can take a more generalized form if the number 7500 is replaced by a variable. To do that, the entire awk command (not just the program) should be stored in shell script, and the parameter supplied as an argument to the script. The parameter is compared with the variable. These variables are properly known positional parameters, and are identified by the shell as $1, $2, $3, etc. in the order they are presented in the line. awk also accepts these parameters without conflicting with the identifiers which also use the same notation. The only difference is that the positional parameters used by awk should be enclosed within single quotes. This will enable awk to distinguish between a positional parameter and a field identifier. So, instead of having line number 1 as the line specifier, you should generalize it as line number 2. There is another method of generalizing the previous program, without having to store the entire command sequence in a separate script file. Since both the shell and awk use variables, you can expect awk to accept values passed by the shell. If you use a variable mpay in the shell command line, then you need to make the following change in the script empawk.awk $6 > mpay { Run the awk program in a similar manner, but this time don’t forget to make the assignment of the shell variable mpay in the command line: shown in line 3. The assignment (mpay=7800) must be made immediately before specifying the filename. This is an unusual feature of this command, which lets you pass shell variables to an variables to an awk program.
• #125: awk has several built-in variables. All these variables are assigned automatically by awk, though it is also possible for a user to reassign them. The major variables you can use profitably with awk are shown below. VariableFunction NRCumulative number of records read FSThe input field separator OFSThe output field separator NFNumber of fields in current record FILENAMEThe current input file FS defines the input field separator, which in this case happens to be the “|”. This is alternative to the –F option of the command, which does the same thing. When used at all it must occur in the BEGIN section so that the body of the program knows its value before it starts processing: BEGIN { FS=”|” } When neither the –F option is specified, nor is this variable defined in the BEGIN section, awk assumes the space and the tab characters as the default input field separator. Likewise, when you used the print statement with comma-separated arguments, each argument was separated from the other by a space. This is awk default output field separator, and can be reassigned using the variable OFS in the BEGIN section: BEGIN { OFS=“~” } see Line number 1. This, however, applies only to the print statement, and not printf, which formats the output strictly in accordance with the format specifier. To locate the particulars of those born in 1952 or 1955, observe that the regular expression 5[25]$ was used to anchor the pattern at the end of the fifth field.
• #126: getline reads from the standard input, it is possible to take a line of information from the keyboard. Using the device name /dev/tty to represent the terminal, you can replace the previous sequence, by placing the getline statement in the BEGIN section. When you execute this script, it will ask you for the value which has to be used as the cut-off point : $ awk –F”|” –f empawk4.awk emp.lst
• #127: awk handles single-dimensional arrays. The index for any array can be virtually anything; it can even be a string. No array declarations are necessary; an array is considered to be declared the moment it is used, and is automatically initialized to zero, unless initialized explicitly. You can use array to store the totals of the basic pay, da, house rent and gross pay of the sales and marketing people. Assume that the da is 25%, and the house rent 50% of the basic pay. Use the array to store the totals of each element of pay and also the gross pay. When you run the program, it outputs the average of the elements of the pay : $ awk –f empawk.awk emp.lst
• #128: awk has several built-in functions, performing both arithmetic and string operations. A list of the major functions is presented in the table below: You may like to clear the screen before displaying an awk output on the terminal or you may want to print the system date at the beginning of the report. You can achieve all this, with the system() function. It accepts any UNIX command (or even a pipeline) as an argument, but the entire command sequence has to be enclosed within double quotes. Simply place the following statements in the BEGIN section of a awk program : BEGIN { system (“tput clear”) system(“date”) } FunctionsDescription int(x)Return the integer value of x sqrt(x)Returns the square root of x index(s1,s2)Returns the position of the string s2 in the string s1 length( )Returns the length of the argument (the complete record in case of none) substr(s1, s2, s3)Returns portion of string of length s3, starting from position s2 in string s1 split(str,arr)Splits string str into array arr Sytem(“cmd”)Executes Unix command cmd, and returns its exit status.
• #129: awk has conditional structures with the if statement, and loops using while and for. They all execute a body of statements depending on the success or failure of the control command. The if statement can be used when the && and || are found to be inadequate for certain tasks. It tests for the condition that is specified on its ‘command line’ (i.e. the line containing the if statement), and then performs the actions that follow it. Optionally, it can also perform a different set of actions if the condition is not found to be true. This statement takes the form if (condition is true) <statement> else <statements> With if you can use any relational operator, as well as the special symbols ~ and !~, to match a regular expression. They all work very well with it, and when combined with the logical operators || and &&, make awk programming quite easy and powerful. Note: Ther is no endif or fi terminator for the if statement used in awk.
• #130: awk suppors two loops, for and while and they both execute the loop body as long as the control command returns a true value. for has two forms. The easier one resembles its C counterpart. A simple example illustrates the first form : for (K=1; K<=9;K+=2) The second form of the for loop doesn’t have a parallel in any known programming language. It has an unusual syntax : for ( k in array ) <commands> This form uses an array where k is the subscript. The unusual aspect of this form is that the subscript here is not restricted to integers alone. It can be virtually anything – even a string ! This makes it very simple to display a count of the employees, grouped according to designation. Since designation happens to be the third field, you can you $3 as the subscript of the array kount[ ].
• #131: In the above program, user defined function newline() is defined, which takes parameters as character and number. This function can be used to print new line on the screen. The for and while loops also accept the break and continue statements, which have similar functions as their C counterparts. break causes an immediate exit from the enclosing loop, while continue starts the next iteration.
• #135: The shell acts as an interpreter. It interprets the command to the UNIX kernel, which in turn executes on the hardware. To achieve a complex solution, often we write chaining of commands joined with various redirection operators. Sometimes we need a programming logic to be involved it our day to day activities. Hence we write series of statements in a shell scripts. Unlike compilers we neither get intermediate object or re-locatable files nor we get executables or output files. In UNIX the .out files work regardless of any extension or no extension name. Here shell script is straightway interpreted by the shell. Any failure on a particular line will terminate the interpretation.
• #136: As in another programming languages we normally write hello world program as first program, here also we will write first “hello world program” as first.sh. In the above example, we have created a file called first.sh, it contained only 1 line which is simple Unix commands. Echo is used as print statement in Unix. The above script will just print the matter given in double quotes. To execute the script, type following line at the shell prompt and you get the similar kind of output. [redhat@unix scripts] $ bash first.sh Hello World of UNIX shell scripts…. [redhat@unix scripts] $ There are 5 different ways execute a shell script, will be discuss in next slide.
• #137: There are 5 ways to execute a shell script. The above mentioned ways would execute a shell script. There are differences in the internal processing of the system during execution, hence one can use the concerned way depending upon the requirement of the application. Each way is explained in detailed in coming slides.
• #138: The first way will spawn a new shell in which the entire script would be executed. Hence all the variables created through the script becomes the part of that new shell. The control is transferred back to the parent shell on the termination of script destroying all the variables which were created during execution. The script does not requires an execute permission since the invoker shell carries the execute permission. Even we don’t have to set the PATH, since we will be running the script from the source directory.
• #139: On entering just the scriptName will give an error as command not found because the shell will assume the script as a command and will try to search it within the built in command lists, if the command is not found then it will search the command as per the directories listed in the PATH environment variable. Obviously the script is not found in the listed directories of PATH also resulting an errors as – command not found. To rectify this error, we have to add our source directory in the PATH variables by using the following command. $ export PATH=$PATH:/home/scripts $PATH is given in order to preserver the original value of PATH. Here : is a separator between the directory listing. Ensure that the directories are not ended with a /. After setting the PATH environment, definitely the command is found, but it gives a new error as – Permission denied. It causes so because yet we have not given an execute permission on the script. Use the chmod command to add the execute permission over the script. $ chmod u+x scriptName The second way will also spawn a new shell. This way is not recommended since it might not execute the script which is required by you, since it can happened that the script name given by you can be found in the other directories listed in the PATH variable prior to your source directory, E.g., consider the PATH as below. PATH=/usr/bin:/bin:/home/script Your source directory is appeared in the last and if your script name is first.sh and lets assume that already there is a script in /usr/bin which has the same as first.sh containing some other business logic, then it will execute first.sh from /usr/bin since the directory appeared prior to your directory, i.e., /home/script.
• #140: This will not give an error related to PATH, since you have mentioned the path using the . (dot) (i.e., your current directory) and / as a separator to locate the required script. It will give an error on execute permission. Hence we have to add only the execute permission by using the following command. $ chmod u+x scriptname. This way bypasses the PATH environment variable, hence we get the accurate and desired file to be executed. This is one of the trusted way to run a script. It also spawns a new shell. In the above diagram lets assume that if we have first.sh written differently across various directories. Then we can execute the required first.sh by using the following commands. [redhat@unix scripts] $ ./first.sh (will execute first.sh from scripts directory) [redhat@unix scripts] $ ./inbox/first.sh (will execute first.sh from inbox directory) [redhat@unix scripts] $ ./outbox/first.sh (will execute first.sh from outbox directory)
• #141: In the above diagram lets assume that if your working directory is etc, then you cant run the required script. If you just enter the script name i.e., first.sh at the shell prompt, then it will obey the PATH directory listing and the execution will take place as per the earlier mentioned 2nd way. [redhat@unix etc] $ /home/redhat/scripts/first.sh This way does not requires a dot operator in the beginning of command, since 'home' is not the directory inside 'etc'. It requires an execute permission. PATH environment is bypassed using this way. It also spawns a new shell.
• #142: All the earlier mentioned ways would execute the script in a child shell. But this way will not spawn a new shell rather it will execute the script in the current shell environment. Hence all the environment variables are accessible to your script. This way does not requires execute permission. If ./ is not given then the system will search the script as per the PATH.
• #143: The sha−bang ( #!) at the head of a script tells your system that this file is a set of commands to be fed to the command interpreter indicated. The #! is actually a two−byte ‘magic number’, a special marker that designates a file type, or in this case an executable shell script (see man magic for more details on this fascinating topic). Immediately following the sha−bang is a path name. This is the path to the program that interprets the commands in the script, whether it be a shell, a programming language, or a utility. This command interpreter then executes the commands in the script, starting at the top (line 1 of the script), ignoring comments. The sha bang statement should be the first statement.
• #144: As discussed earlier, the #! is actually a two−byte ‘magic number’. Check whether the script is there in the directory or not. Give execute permission to the above script and run as shown below $ ./magicScript.sh Check the output. As expected line number 2 should be printed on the screen. But output is nothing. Check whether the script is there in the directory or not. When we run the script after shabang statement it follows the path which is for deleting the file. The first statement deleted the file. Hence we can not see the output of second line.
• #145: Export statement is used to avail the value of a variable in child shells. The child shell can only refer to this exported variable in a read only fashion. Line 1 will print the current process id, i.e., the current shell's process id. Then we have created and initialized variable A to 10. Line 4 will start a new shell session under the current shell. It means that it will start a child shell process. Here the value of A is unavailable unless to export it. Line number 8 will export the variable to all child shells. Doing A=90 inside the child shell will create a new variable A holding value 90. It has nothing to do with the variable A of parent shell. Use exit statement to go back to the parent shell process. Use export –n A to un-export the variable. The exported variable is available only to the child shell created onwards and not vice-versa. Use the unset command to remove current value of any variables. $ unset A #This statement will remove the current value of A
• #146: We are continuing with script first.sh As discussed in previous slide command ‘bash’ spawns new shell and starts new process . Executing first.sh using first way also spawns new shell and starts new process. Output of line 5 will be value of A is Where value of A is set as 100. the variable values can not be used for child processes. Hence value of A is not printed on the screen. After setting value of A=500 it shows output as value of A is 500. To send value to the child processes or child shell export the values of variables using ‘export’ command $ export A After exporting value of A now output of line number 5 can be seen as expected.
• #147: It is extremely important to put comments in shell scripts. Anything followed by # symbol is a comment and therefore ignored by the interpreter. It is good practice to make a shell script self documenting (i.e. someone with almost no scripting knowledge can read your comments in the script and reasonably understand what the script does). Maintenance of the script becomes easy.
• #149: After executing search.sh, the script will ask user to enter file name and pattern. The variables ‘filename’ and ‘pattern’ are used to accept file name and pattern respectively. ‘read’ statement will read the data entered by the user. Input supplied through the standard input is read into the specified variables. grep then runs with variables pattern and filename as arguments. $ sign will replace the value of specified variable.
• #150: We can give command line arguments also. $ksh serachPattern.sh sales emp.lst In above command, ‘searchPattern.sh’ id program name, ‘sales’ and ‘emp.lst’ are two parameters we are passing to the program. Parameter ‘$0’ is program name i.e. ‘searchPattern.sh’. Special variable ‘#’ gives you, number of command line arguments passed. Special variable ‘*’ gives you, all the command line arguments passed. ‘$1’ and ‘$2’ are first and second parameters passed to the command.
• #151: (Source Code : usageTest.sh) The above program takes only 2 arguments from the command line. If more than two or less than two arguments are passed then, it shows the USAGE message guiding the user to pass only two arguments to a script. This way we can control desired number of arguments passed to a script. Here further validation on $1 and $2 can be done such as only integer values, or values in a specific range etc.
• #153: We can set positional parameters inside the script using ‘set’ command. In above slide parameters ‘123’, ‘456’, ‘789’ are positional parameters same as that of command line arguments.
• #154: For ksh, at a time you can pass 9 command line arguments to the script. If you want to pass more than 9 parameters, then use ‘shift’ command, as shown in slide above. In above script, we want to pass 11 arguments to the script. Shift positional parameters by 2. shift 2 So that you can read 10th and 11th parameter, as 8th and 9th parameter. But in that case, 1st and 2nd parameters gets lost. So better to save those parameters in variables for further use.
• #155: The exit command may be used to terminate a script. Every command returns an exit status (sometimes referred to as a return status ). $? reads the exit status of the last command executed. The last command executed in the script, which is, by convention, 0 on success or an integer in the range 1 − 255 on error.
• #156: You can do computation in scripts using ‘expr’. $ expr $A + $B As shown in above script, value of A is 500 and value of B is 20. $ sign prompts values of the variables. You want to add two values, so use addition operator and prefixed the statement with ‘expr’. You can store the value of commands using back ticks (`). ADD=`expr $A + $B` The above command, stores the value of addition in ADD variable. Another way of computation is, $((A+B)) As ‘*’ is meta character in Unix. ‘*’ to work like multiplication operator, prefix ‘*’ with ‘\’, so that it will navigate the meaning of ‘*’ and it will work as multiplication operator.
• #157: Variable values can be extracted inside double quotes, e.g., see the following statement. $ echo "value of A is $A" $ echo value of A is $A $ echo 'value of A is $A' The last command will not show the value of A because the contents between single quote is considered as hardcore string constant. The $A can only be interpreted within double quotes or no quotes. $ echo "Today's date is date" The above line will be printed as it is. If we want the output of date command to be merged with echo then we should enclosed that respective command in grave accent marks, (``). Hence the modified command would be as below. $ echo "Today's date is `date`"
• #158: Computation can be done with the let statement, as shown in slide above.
• #160: The shell provides two operators that allows conditional execution && operator Syntax: cmd1 && cmd2 The && delimits two commands; the command cmd2 is executed only when cmd1 succeeds. In first example, if pattern ‘director’ found in file emp.lst then only the message ‘Pattern found’ is printed on the screen. || operator Syntax: cmd1 || cmd2 The || delimits two commands; the command cmd2 is executed only when cmd1 fails. It acts reverse that of &&. In second example, if pattern ‘manager’ does not found in file emp.lst then the message ‘Pattern not found’ is printed on the screen.
• #161: if statement uses the above mentioned forms, same as in the other languages. Condition in of if statement is written in square brackets. The ‘if’ statement ends with ‘fi’. There is space between if , [ and ].
• #162: As discussed earlier, $? Gives exit states of last executed command. At line number 4, ‘grep’ command is executed. At line number 4, you can check exit status of last executed command, i.e. ‘grep’ command. We can store the value of exit status in the variable. if statement is checking the value of GREP_STATUS is equal to 1 or not using ‘-eq’. You can combine more than one command using semicolon (;).
• #163: When you use ‘if’ to evaluate expressions, you need the test statement because the true or false values returned by expressions can’t be directly handled by ‘if’. test works in three ways Compares two numbers. Compares two string’s or a single one for a null value. Checks a file’s attribute. Test doesn’t display any output but simply sets the parameter $?. Drawback:- Numeric comparison restrict to integers only. You can use various numeric comparison operators. OperatorsMeaning -eqEqual to -neNot equal to -gtGreater than -geGreater than equal to -ltLess than -leLess than equal to
• #164: You an use various string comparison operators, shown above.
• #165: You an use various file comparison operators, shown above.
• #166: Above script, ‘fileSearch.sh’ is reading the file name from user. At line number 2, it is checking whether the user entered file is exist or not, using ‘-e’ option.
• #167: Above script, is for ‘elif’ test script. If you are using ‘elif’ then there should be only one ‘fi’.
• #168: Instead of using, nested ‘if’, as in previous script, you can use case statement.
• #169: Above program, uses case statement for multiple cases. Line 5, reading the choice from the user. case statement, checking for the choice entered by the user. Comparing the choice with various cases. If non of the condition is matching, then ‘*’ will get called, which is used for default condition. ‘case’ statement ends with ‘esac’.
• #170: You can give multiple choices to check one case. Line 2 is asking the user, whether user want to continue or not. read command will read the users choice. User may enter the choice as ‘y’ or ‘Y’ or ‘yes’ or ‘YES’. Taking all the options in the consideration, you can write the cases as shown in above program.
• #171: You can loops in the script. Syntax shows, after while statement and condition there should be ‘do’ statement. ‘while’ statement should end with ‘done’.
• #172: Above script, whileDemo.sh shows example of while loop. Line 17th, asks user whether he/she want to continue or not. Reads choice from user. if choice is either ‘y’ or ‘Y’ continue loop.
• #173: The until construct works almost exactly the same as while. The only difference is that until executes the body of the loop so long as the conditional expression is false, whereas while executes the body of the loop so long as the conditional expression is true. Think of it as saying, "Execute until this condition becomes true.“.
• #174: Above program, until loop executes the body as long as the conditional expression becomes true.
• #175: for loop execute one for each ‘variable’ in the ‘list’ until the ‘list’ is not finished.
• #176: Three ways you can use for loop. In way 1, there is list of all the planets as “Mercury, Mars, Saturn”. Variable ‘planet’ takes all the items one by one and prints those items on the screen. In 2nd way, instead of giving list to the for loop, the list is stored in the variable ‘PLANETS’. The variable “PLANETS” is passed to the for loop. Another variable “planet” in the for loop, taking value from the variable containing list and printing on the screen. 3rd way is C style for loop, only difference is instead of single round bracket, it uses double round brackets.
• #177: break command causes an exit from inside a for or while loop. In above program prints value from 1,2,3… . When if condition encounters value of ‘a’ is greater than 5, it executes the break statement and exit from the while loop.
• #178: Continue command resumes the next iteration of the enclosing for or while loop at the enclosing loop. Above script prints even numbers from 1 to 20. If condition checks whether remainder is 0, if yes then continue the loop else print the value.
• #180: You can redirect block of data to the file using standard output, as shown above. In above script, output of while loop is stored in the file named outfile.txt.
• #181: You can redirect block of data from the file using standard input, as shown above. In above script, input to of while loop is taken from the file named emp.lst.
• #182: ‘#’ is used for single line commenting. If you want to give multi line comments then you do it as shown in line 2 and line 5. Whatever statements you will write in between the line 2 and line 5 those lines will be ignored by the shell.
• #183: Bash support one-dimensional arrays. Array elements may be initialized with the variable[xx] notation. To dereference (find the contents of) an array element, use curly bracket notation, that is, ${variable[xx]}.
• #184: Alternatively, a script may introduce the entire array by an explicit declare -a variable statement. Line 1 declares variable ‘arr’ as array using option ‘-a’. First for loop stores the value from 0 to 9 into the array ‘arr’. Second for loop displays values from the array.
• #185: Function is series of instruction/commands. Function performs particular activity in shell i.e. it had specific work to do or simply say task. To define function use above syntax.
• #186: In above script ‘calc.sh’, there is one function ‘add’, which is taking values from user and prints the addition value on the screen. To call the function just call it by the function name.
• #187: You can pass parameters to the function like command line arguments. Parameters passed will be positional parameters $1, $2 and so on.
• #189: Its name is an abbreviation of "write to all". it displays the contents of a file or standard input to all logged-in users. Second command displays the contents of a file specified. You can write messages to specific user using command ‘write’. $ write [username] message to that user
• #190: mesg is a Unix command that sets or reports the permission other users have to write to your terminal using the write commands. mesg is invoked as: mesg [y|n] The 'y' and 'n' options respectively allow and disallow write access to your terminal. When invoked with no option, the current permission is printed. I/O redirection may be used to control the permission of another tty. As shown in above slide.
• #191: Say I want to send a short note to the person with the email address [email protected]. Here is an example of how I would do this: You can read the mail using ‘mail’ command.
• #193: Unix system provides aspecial login name for exclusive use of system administrator; it is called root. Prompt of root is #, instead of $
• #194: Creating user’s environment: Users often run to administrator with the complaint that a program has stopped running. It is natural for him to first try running it in a simulated environment. su command lets administrator recreate the user’s environment without taking login-password route: su –user1 This executes user1’s .profile and temporarily creates user1’s environment. This mode is terminated by hitting <Ctrl-d> or using exit command.
• #195: passwd: change own password as well as other user’s password passwd user1 Note: when superuser uses passwd command to change a user’s password, he doesn’t have to enter old password. date : Change System date # date 10240934 Sat Oct 24 09:34:00 IST 2009 Note:argument usually an 8-character string of formatMMDDhhmm
• #197: Ex-1 :shutdown after 2 minutes Ex-2 : shutdown immediately Ex-3: shutdown and reboot (init level 6)
• #198: df (disk free) reports amount of free space available on the disk. Output always reports each file system separately. It has option –t (total) # df –t /dev/oracle du (disk usage) command reports usage by recursive examination of the directory tree. It has option –s (summary) # du –s /home/*# assessing space consumed by users 144208/home/user1 99999/home/user2 56432/home/user3
• #199: This is how find operates: It recursively examines all files in the directories specified in path_list It then matches each file for one or more selection_criteria Finally, it takes some action on those selected files
• #201: Taking action on selected files: # find /usr -mtime +30 -exec rm -f {}\; # find /home -size +2000 -atime +30 -ok rm -f {}\;
• #204: Adding a user involves setting few parameters in /etc/passwd: 1. UID and username 2. GID and groupname 3. home directory 4. login shell 5. mailbox 6. password
• #205: fsck command is used to check and repair a damaged filesystem.