Welcome to the session on combining commands
and files. Linux is a modular operating system. One can
combine different commands to achieve a complex task. One can also use multiple files for both input
and output. In this session, we will see how we can combine both commands and files so that a single command line could actually
achieve a fairly complex task. So, here are a few examples of how commands
can be combined on the command line. The very first way is something that we have already
seen. You have the commands that are executed one after the other, irrespective of whether the
commands have been successful or not. Now the second means is where you have got
the two ampersand symbols, which would lead to an effect where the second command is executed
only if the first command is successful. There is one more way of combining the commands
where you have got the two pipe symbols between the two commands. The second command would
be executed only if the first command were to be false.

The written codes for each command, something that we have discussed in the previous session, would be very useful. Because if the written code was 0, then it would be true. And if it were to be any other value greater than 0, then it would be false. We could use the Boolean values indicated by the written code to combine the commands appropriately. We could see that the double ampersand is used for the and logical operator. And the double pipe that is used as our logical operator could be then interpreted similar to the Boolean variables combined in the same manner. We have seen in a previous session that we can combine multiple commands on the command line. So, let us combine them. For example, ls and then date command and then counting the number of lines in a file called etc profile. So, these three commands are then executed irrespective of whether they are successful or not. So, let us make one of them fail. So, we would type ls /blah which we know that does not exist and then edit.

command and then wc -l etc profile. And you see that the first command has failed,
nevertheless, the second and the third commands have gone on to execute. Now this is a behavior that we could actually
change where the execution of the second or third commands can depend on whether the first
command has been successful or not. So, we will come to that in a moment, but before
that I want to show you that there is one more way by which you can execute three commands
on the command line together as a list of tasks. So, I would start them with a parenthesis
the bracket you could see that the same action has been repeated. So, there is no difference. However, when you
use the parenthesis what actually happens is that the command is executed in a subshell
which is returned back to the shell that we are using. We can inspect the variable shell
variable called bash subshell. So, this bash subshell is a variable that
contains an integer which tells you at what

level of execution we are. So, let us try that out by passing it on to
the parenthesis to see whether the number remains same or not. So, I would now say echo. Now you see that the number has changed, which
means that this particular command has been executed in a subshell and then is returned
back. So, we could then use this to check how the
execution is happening when we pass multiple commands, and you could see that now the ls
and date were executed in a sub shell and the value of a bash sub shell is then 1. So, you could actually use multiple parentheses
to have subshells within subshells, and we can try that out right away. Now you could actually create a sub shell
within this by editing it like this. Now you see that the bash subshell is reporting
a value of 2, which means that from the bash that we are actually in a subshell has been
created in which the ls command has been launched.

and then another subshell has been created within which the date command has been launched. So, we could verify that as follows. So, you would see that the first command ls is executed here and then the bash subshell shows that it is number one. And then another subshell has been spawned and the date command has been executed. And then the bash subshell value is two, which means that the date command is executed in a shell launched two levels below what we are actually running. Now, why is it useful? Sometimes this is useful to ensure that the environment within which the execution is happening is closed. It is not a good idea to launch subshells indefinitely because each time a subshell is launched, the environment has to be prepared for that and therefore it could be expensive computationally. However, it has certain usages where you could provide a certain environment, namely certain values of the shell variables that are created for a particular executable and so on, to be provided so that the execution can happen in a subshell without being interfered by

other variables that may be present in the
parent shell. Now let us see if we can run the commands
where the execution depends upon the exit status of one of the commands. So, we have run two commands one after other
and we will. Now make the first command fail and then we
will actually have ampersand twice to indicate and operator and then we have the next command
date that is being executed. Now you will see that only the error message
for the ls command has been displayed but the date command did not execute. What this
means is that the error code returned by the ls command is false, which means it is more
than 0, and therefore the second command, namely the date command, did not execute. Why is this useful? Because it helps us in ensuring that the second
task, which depends upon the first task, will execute only if the first one is successful. Now let us look at a possibility that is exactly
the opposite. So, here we are combining the commands where
we are using the double pipe, which is for

the r operator and you see that the date command
has executed, though the first command has been a failure. So, this is a situation where the second command
is like a plan b, where if the first command failed, only then the second command would
execute. If the first command were to be successful,
the second one will not be executed, and that is something that we can see right away. And
you see that the ls command was successful in displaying the list of files and directories
in the current directory, and because it was successful, the second command date did not
execute. So, here is an illustration where the three
symbols, the semicolon, the double ampersand, and the double pipe, could be used to combine
commands in such a way that they can be executed in three different means: that is independent of the execution status,
dependent on the execution status to be true, and dependent on the execution status to be
false. So, now it is up to our imagination to combine
these in very many ways by which we can actually

achieve a complex task that depends upon the
commands that are executed before that. Let us try that out once. So, here I am running three commands and I
am using the and operator for all the three. So, if any of them fail then the next
set of commands on this command line will not get executed. So, we will make the first one fail that is
quite straightforward. We just need to try to list commands of a non-existing directory.
Therefore, nothing after the ls command will get executed. Now we could actually make the first command
successful but the second one failed by giving some arbitrary option which we know that will
not exist. And you see that the date command has been given an option -Q. There is no such
option. So, the date command has failed and therefore
after the date command the wc command did not execute. However, the ls command went on
to execute successfully and that is why the date command was tried out. So, like this you can actually combine multiple
commands using both ampersands as well as

pipes to achieve the Boolean logic that you would desire for the total command that you want to execute, which could be achieving a fairly complex task. Here we are using silly examples like ls or date, but it could actually be a bash script that you would have written which actually takes an action on your data or processing of some information that you have or perform network operations and so on. So, each of these commands can as well be fairly complicated bash commands, and the logic of combining them on the command line would remain the same. Now we come to the concept of file descriptors. So, in C language when you learnt how to read and write content with the files, then you would have come across the concept of file pointers where you have a file pointer pointing to a file which you could read from or write to. So, every command in Linux has three file descriptors. They are called as the standard in, standard out, standard err. So, the standard in is a pointer to the stream which is coming from the keyboard or the user.

input and the standard out, and standard err
are usually pointing to the screen where the display of the output is made. Numerical values
like 0, 1, 2 are assigned to the file descriptors standard in, standard out, and standard err
respectively. So that you could actually use these numerical
values on the command line for redirecting some of those outputs or inputs to either
a file or a command. Now in Linux, we should imagine that each of
these three file pointers are actually looking at only the stream of characters. Now that we could understand that there are
these three file pointers. The next thing that Linux provides us is that
we can redirect these three streams to either a file or a command, or the default behaviour
can also be left as it is, and that opens up a lot of possibilities for us to imagine how
the commands can be combined. And there are certain operators that are used
to combine the commands, and we will explore them shortly. So, here is the first operator that we are
going to use to combine a command and a file.

that is the greater than symbol or the right
angle bracket. So, what does it mean when we say when we
write a command like command greater than file1, what happens is that the standard output
is redirected to a file called file1. And the rest of the pointers will remain the
same, which means that the input would continue to come from the keyboard, and if there is
any error that has come out of running the command, that would come onto the screen;
only the output is actually redirected to the file. Now the file will be created afresh in the
directory where the command is being executed, which means that you must have write permissions
to the directory, and you should not have any file by the same name already, because it would
be overwritten when you run this command. So, let us see this command in execution and
understand how does it work. When you have a command that has a lot of
output and we would like to look at that output.

slowly this feature of greater than symbol
to redirect the output to a file would be very useful. Now I am in the home directory, and we would
actually look at the listing of the user bin directory. So, ls user bin, and you see that a number
of files have come out of the screen. I can actually do the same thing by typing -1
user bin. It would give one command per line of the output, and we would like to then write
this output to a file. So, that we could look at that file slowly.
So, that is where the greater than symbol comes, and then I have to give a file name. So, I would give my file name as file1, and
then the file1 has been created, and you can see that there is no output on the screen. The reason there is no output on the screen
is because there is no error that has actually come when I executed this command. Now I look at the contents of my home directory,
and sure enough, there is a file called file1.

that has been created, and it has 26 kilobytes.
Let us say quite a bit of content because there are a lot of files in the user bin directory
of my computer, and we can now inspect the contents of file1 using the less command. So, you can see that the list of file names
are being displayed on the screen. And I can scroll through at my leisure. I could
also go back with the up arrow key and look at them, and I can press the space bar to scroll page
by page, and when I am done, I can press q and come out. So, you can see that we have the ability to
pass on the output of a command onto the file. Now, if I happen to make an any error and there
is no output, the file will be empty. So, now you can see that the file1 is 26 kilobytes. Now what I would do is I would actually intentionally
make an error. So, I would try to list the contents of a
file called slash blah, which does not exist.

and I rewrite the output to file1. Now, what happens is that there is an error
and the error has actually come out to the screen, and we can inspect the contents of
the file1. You see that it is zero, that means that there was no output naturally because
there is nothing to report as a list of contents for this slash directory, which does not exist. So, you can also see that a file which was
actually having 26 kilobytes has been overwritten, and it is now an empty file, which means that
when you use the greater than symbol, you must be aware that any existing file will be overwritten,
and therefore you should use fairly unique file names when you are in the beginning stages. So, that the danger of overwriting does not
hurt you in losing any of the files that may be valuable to you. So, this is what we have seen: the command
is ls, and the file1 is a text file. When we ran it, what happened is the output was
actually put in the file, and any error was

actually put in the screen for us to view,
and the command did not have any input, and therefore the standard in was not very relevant. Now we will put the greater than symbol usage
to some use where we can actually look at the output of a command which scrolls a bit
too fast. So, let us do that with a command called hwinfo,
and when you type that, it is inspecting the various hardware, and it is now dumping the
output onto the screen. So, we then write the same command, greater
than, and I need to give a file name because this is useful information. I would like to
type it as hwinfo dot txt to indicate that it is a text file, and it has the information
coming from the command hwinfo, and you could see that there is nothing on the screen. Now, because everything has been put onto the
file, now let me look at the file by using the less command, and you see that it is giving
me information about the system, and you see

that it is telling me about the kernel that was booted and initially how the probing of the hardware is happening. As you scroll through, you will see information about your computer hardware, and we will look at that in the next session to inspect the computer hardware. But for now, you could actually see that the content can be made available for you to read at your leisure. Now, if you try this command in a folder where you do not have write permission, you will get an error. Let us just check that out also. So, we would go to let us say slash user, and we would actually look at the permissions, whether we have write permissions or not. You see that the current directory has write permission only for the root, and I am not root right now. Who am I? I am another user, and therefore I do not have write permission because the file is owned by root, and only the owner has write permission. I belong to the other category, and I do

not have the right permission. So, here if I try to run the command, then
I will get an error. So, let us say I type ls greater than and
then file1, and you see that permission denied is the error that I get. It is good because
I do not unintentionally make any changes to the system folders. And this is a security measure that is built
in, and that is one of the reasons why we should not be logged in as a root user with administrative
privileges when we are a novice user. Unknowingly, sometimes we may make some errors. So, before you run the command, you always
ensure where you are, and when you are in your home directory, then you can do these commands
without any error. Now, the greater than symbol can also be used
to actually create files. So, let us look at the help for the cat command,
and it says concatenate files and print on the standard output, which actually can
be interpreted with respect to the information.

that we have about a command. So, we have cat and then greater than and
then a file. Okay. So, what happens when we type this command
is that the cat command is supposed to receive the input from a file that is listed on the
command line. But we did not specify any file. If you did
not specify a file on the command line for the cat command to read the content, then it
would read that content from the keyboard, which means it is reading from the standard
in, and we can then see how we can type some command, which will then go into the file called
file1 because the greater than symbol indicates that whatever cat is supposed to display will
be put into the file1. So, you see that the command prompt is now
empty, and it is expecting us to type some. We can type some content now. So, that is my first line. Second line goes
like this, and how do I come out? You could come out using
control plus d option. Okay.

So, I do that. Now, and I have come out now. What happened is that the three lines I have typed have been read by the command cat, and they are supposed to be displayed onto the screen. But because of the greater than symbol, they are being redirected to file1. So, let us look at the contents of the file1. You see that the file1 has 170 bytes. It is a short file, and as we have already mentioned earlier, it has been overwritten by this command because earlier it had some other content. So, we have overwritten it. So, that is not a problem because these are all only meant for practice. But remember always that whenever you use greater than symbol, that file will be created afresh, and if a file exists by that name, it will be overwritten. Let us look at the contents of the file file1, and you see that what we have typed is there now on the screen for us to see. Now you can see that the difference between cat greater than file1 and a cat file1.

So, what it actually does is that greater than actually would take input from the keyboard and create a new file by the name file1. Whereas without the greater than symbol, the content of the file 1 will be displayed onto the screen. So, it is very important for us to understand how the file pointers are actually being redirected. So, here you can see that by default the cad command was taking input from the standard in keyboard and it was writing it out onto the screen. But we have put a greater than symbol. So, it was writing on to the file1. Now let us test our understanding of the input output redirection by seeing what happens when we just say type cat. What happens is that there is no file specified on the command line, which means that the input for the cat command should come from the keyboard and the output should also be on the screen. So, which means that the command would now take input from the keyboard and place the

output on to the screen and let us see how
that happens. So, as soon as I press enter, you see that
that particular line has been read and it is displayed onto the screen. So, you can see that this command is actually
processing the text line by line, and you can actually finish up this task by actually pressing
ctrl + D, so that you can come out of the command. So, finish up by end of the file using control
+ D to close. So, now I press ctrl + D, and what happens is
that it has already finished writing the content on the screen. So, there is nothing that has happened. So, this actually helps you understand how
the file pointers are being interpreted. So, there are three file pointers: standard
in, standard out, and standard error, and we have to be very conscious which file pointer
is pointing to which of the aspects, namely the keyboard or the screen or the file. You
could actually use two symbols of greater than or the right angle bracket.

So, if you use these two symbols side by side without any space between them, the interpretation is that the output of the command should be appended to the file1. By appended, we mean if the file1 exists, then the content will be written to the bottom of the file. If the file1 does not exist, then it will be created, and it will be written from the top. So, it means that it is a bit of a slightly safer option to use. So, that if there was any file, then it would not be overwritten; it will only be appended to, okay. So, that is why here in the graphic, I have shown the arrow pointing to the bottom of the file1, not to the top, and that is a hint that the content will be appended to the file1 and not replaced. The warning of the file1 to be created remains, and you will have errors if you run this command in a directory where you do not have permissions to create the

file1 because you do not own that particular
directory. Now let us try out the double greater than
symbol to see how to append the content of any file by using the commands and the output
would be written to that file in an appending mode. So, let us try that out. So, I have this file1 and file two. I will
just erase all of those, okay. Now I do not have any files. So, I now use the command to also show you
how it will be created when it does not exist. So, what we do is we type date and double
greater than and we'll say file1, and you see that a file has been created. Now afresh, though we have used the double
greater than symbol because if the file was not existing, it will be created, and if it
was available, then it will be appended to. So, I run the command again. Now and once more and once more. I have run
totally four times.

So, let us look at the contents of that particular file, and you see that there are four lines. From the seconds, you can make out that it is actually appending one after the other. So, this way, we have not lost the content that was appended or inserted earlier, and therefore, this mode is preferable when you want to preserve the contents of the file while redirecting the output of a command to a particular file that you want to inspect later. So, when you want to write multiple commands where the output would be redirected to the file and appended later on with another command, then you could combine all of them also in the same line using the semicolon as follows. So, you would type date, and let us say file two, and then we have another command, let us say wc -l, and then a third command perhaps to inspect the file user bin znew, and that output also I would like to write to file2. So, you can see that I am combining three commands, and the output of each of those is

being written to the file, and then I can actually look at the content of the file too, and you could see that the three outputs have been appended, and we are seeing all of them. Now what I would do is I would make an intentional mistake here, the second one I would actually make with only one greater than symbol, and you. Now see what would happen, you see that only the second and third outputs are there. The output of the first date command is gone because the second command has actually overwritten it because it has created the file to affect, okay. So, this is how you should interpret the greater than symbols. So, double greater than and the greater than both will actually create the file if it was not existing, but double greater than would actually append if the file was already existing, and therefore the safer option for you to explore. Now we will use the same double greater than symbol to append text to a file that we are

creating. So, we would remove these files. Now we would actually try to append to a file
and the file does not exist initially. It does not matter. So, I will have file1 and I want to write
something here. So, I would write like this. Okay, first attempt
to add text to the file file1. Now I come out using control + D. Now I have
come out. So, I would run this command once more. Now I say this is my second attempt to append
text to the file called file1. We can inspect the content using cat or less
or more commands. Okay, I come out using control + D. So, we have done two attempts. So, we would do one more. This is the third
attempt to append content to the file called file1. So, we have now done three attempts and let us see what
is the content of the file1 and you see that all the lines are there. All the six lines
are there. The first attempt had these three

lines which are now reproduced here. The second attempt had these three lines, and
they are here now, and the third attempt had only one line, and that line is here. So, you can see that you could use the cat
command to append text to your file. So, you can actually create files with the
text that you want to create in it by using the cat command, the double greater than symbol,
to append, and this is one very simple way by which you can create text files. We will learn how to edit these files in a
visual environment within the command line in one of the sessions later on, but this is
a very quick and dirty way by which you can actually create text files and append stuff
to that one by one.

Lecture 7 - Combining commands and files

Lecture 7 - Combining commands and files

Scripts