Hi everybody!

I am trying to learn bash scripting and have always been puzzled by the fact that all of the guides say to use this code before executing: . I have always skiped this step and gone on to run my program like so: . What not skip the chmod command and do it a simpler way? I know that just typing bash before the name of the script is much easier for me than changing its permissions!

Because it's sloppy. And if you write a script that gets called by the system instead of you invoking it, it will need to have the executable permission.
And in most GUI file managers you won't be able to run a script by clicking it unless it is set to +x.

lets look at it this way. In linux you have three basic permissions (not getting into owner,group,any)

you have r = READ, if you want to be able to touch this file you must have read permissions.

you have w = WRITE, if you wish to change or modify the file you must have write permissions.

you have x = EXECUTE, well duh if you want to execute, run, something it must be set to have EXECUTE permissions.

as cortman mentioned above if you ever wish the script to be run by the OS or via GUI it must have the proper permissions.

say for example you create a backup script to run nightly at 2am local on your system. In order to do this you use a combination of proper permissions a+x and you set it to either roots or users crontab depending on what level of backup you are performing. If both points are not met, your script will fail.

yes, it can be a sloppy affair.
I use
in my ~/.vimrc so that any interpreter on
line(1) will cause vim to make the file "-rwxr-xr-x"

typing sucks.

Have fun.

1 members found this post helpful.

Yes but the vast majority of system tasks will be done in the terminal. But, I can see what you mean, it is sloppy. Oh well, old habits are hard to break.

True, but an equal amount of system tasks will be done automatically by the system, for which you also need to set the executable permission.

Do whatever works for you (within reason).

Its actually less typing in the long run.
Consider:

1. create file : fixed amt of work
2. chmod file : fixed amt of work
3a. run using '/path/to/script'
3b. run as 'bash /path/to/script'

Note the extra typing for each usage via 3a. (EDIT: 3b darn it, not 3a)
The first line in the script should be the hash-bang line.

Everyone else has pretty much explained it to you. But did you know if you put the number sign followed by the shell you want to execute a script on the first line of the script file it will use that shell to execute it regardless of your command line shell interpreter.

For instance, first line of script is:
#/bin/sh

The OS will run this script with the /bin/sh.

Or:
#/bin/csh

The OS will run the script with /bin/csh

Or:
#/bin/ksh

The OS will use /bin/ksh.

Etc. Otherwise, without this specification, the OS will use your default command line shell. So, to make a script more portable, it is a good habit to specify which shell to use. Just a pointer about scripts.

Enjoy.

@ Nbiser

If preceding bash before the script makes you happy go for it.

However, linux veterans probably do this with their shell scripts.

• They chmod their script(s).
• They test their script(s).
• Put script(s) in a bin folder in their home directory
• Add their bin folder to the $PATH variable

This method allows you to call the script from any directory you're in. No need to type bash. Thus, less tying.

@Bill Melater
you missed the '!' each time eg
etc ...