External commands from Scilab

We are going to begin with a directory called physionet where we have installed a file called `example' whose content is (this section assumes that we are running Linux)

echo "this is an example"

Then we assign permissions and the result is

[j@localhost physionet]$ chmod 777 example
[j@localhost physionet]$ ls -l
total 4
-rwxrwxrwx    1 j        j              25 Jul 31 20:59 example*

Now we are able to run the file from the prompt

[j@localhost physionet]$ ./example
this is an example

We can see that it functions properly. Now we are going to try to run this small script with Scilab. We begin by calling Scilab from any other window

                           ===========
                           S c i l a b
                           ===========
 
 
                          scilab-2.6
                  Copyright (C) 1989-2001 INRIA 
 
 
Startup execution:
  loading initial environment
 
-->

We have to go to the directory where we are going to store our data files. We can go there with the next command

-->chdir('/home/j/physionet')
 ans  =
 
    0.

Now we want to repeat the commands from the Scilab prompt

-->unix_g('ls -l')
 ans  =
 
!total 4                                                          !
!                                                                 !
!-rwxrwxrwx    1 j        j              25 Jul 31 20:59 example  !

Notice that we can assign the result to a variable

-->str = unix_g('ls -l')
 str  =
 
!total 4                                                          !
!                                                                 !
!-rwxrwxrwx    1 j        j              25 Jul 31 20:59 example  !

-->str
 str  =
 
!total 4                                                          !
!                                                                 !
!-rwxrwxrwx    1 j        j              25 Jul 31 20:59 example  !

Pay attention to this result because it is very important. We have run a command and the result has been stored in a variable. The command that we have used (`ls') does not belong to Scilab. Now we are going to run the script that we created before

 

-->unix_g('./example')        
 ans  =
 
 this is an example   
 
-->str2  = unix_g('./example')
 str2  =
 
 this is an example   
 
-->[str2;str]
 ans  =
 
!this is an example                                               !
!                                                                 !
!total 4                                                          !
!                                                                 !
!-rwxrwxrwx    1 j        j              25 Jul 31 20:59 example  !

Notice that we can store the results of the commands as a matrix of strings. Now we can manipulate the strings like any other Scilab object. Although this matrix is a Scilab variable, it was not produced by Scilab but from the prompt and some very simple script that we created.

The core of this method is the command `unix': from inside Scilab you run a command as if you were in the prompt of the operating system. You have four variants of this command (these commands are identical if you are running Scilab on Windows systems)

unix_g :

shell (sh) command execution, output redirected to a variable

unix_w :

shell (sh) command execution, output redirected to scilab window

unix_x :

shell (sh) command execution, output redirected to a window

unix_s :

shell (sh) command execution, no output

In the next chapters we are going to make an extensive use of these commands. Sometimes we need that `unix' assigns the result to a variable (for instance to introduce a result without creating any file), sometimes we do not want to store the variable but we want to see the result (for instance we want to see the content of an ASCII file) and sometimes we do not need to store the result of the command.

Of the different variants of `unix', one of them (`unix_x') creates a fast and powerful visualization window that is especially suited to explore the content of ASCII files.

The next chapters explore the possible use of these tools to make Scilab work in common with other programs.