14

I have written a bash script which creates a series of directories and clones a project to selected directories.

For that, I need to cd to each directory (project 1 and project 2), but the script doesn't cd to the second directory nor executes the command.

Instead, it stops after cd and cloning in theproject2 directory. Why doesn't it call the cd_project1 function in the following code?

#!/bin/bash
#Get the current user name 

function my_user_name() {        
current_user=$USER
echo " Current user is $current_user"
}

#Creating useful directories

function create_useful_directories() {  
  if [[ ! -d "$scratch" ]]; then
  echo "creating relevant directory"
  mkdir -p /home/"$current_user"/Downloads/scratch/"$current_user"/project1/project2
  else
     echo "scratch directory already exists"
     :
  fi
}

#Going to project2 and cloning 

function cd_project2() {

  cd /home/"$current_user"/Downloads/scratch/"$current_user"/project1/project2 &&
  git clone https://username@bitbucket.org/teamsinspace/documentation-tests.git
  exec bash
}

#Going to project1 directory and cloning 
function cd_project1() {

  cd /home/"$current_user"/Downloads/scratch/"$current_user"/project1/ &&
  git clone https://username@bitbucket.org/teamsinspace/documentation-tests.git
  exec bash

}

#Running the functions  
function main() {

  my_user_name
  create_useful_directories
  cd_project2
  cd_project1    
}
main

Terminal output:

~/Downloads$. ./bash_install_script.sh    
Current user is mihi
creating relevant directory
Cloning into 'documentation-tests'...
remote: Counting objects: 125, done.
remote: Compressing objects: 100% (115/115), done.
remote: Total 125 (delta 59), reused 0 (delta 0)
Receiving objects: 100% (125/125), 33.61 KiB | 362.00 KiB/s, done.
Resolving deltas: 100% (59/59), done.
~/Downloads/scratch/mihi/project1/project2$
New contributor
Jenny is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
  • 3
    Consider accepting one of the answers. If more than one answer is a solution to a question - accept the best one and up-vote another. – LeonidMew Mar 14 at 22:04
  • 1
    Hi LeonidMew. Sorry I have no idea how to accept the answers. Both answers are equally good though. – Jenny Mar 14 at 22:08
  • 5
    @Jenny, don't feel rushed. Read What should I do when someone answers my question? instead and act accordingly when you are satisfied. Just take your time, there is no reason to hurry. It's perfectly OK if you decide in a day or in a week or in whatever time it takes. – PerlDuck Mar 14 at 22:15
  • 2
    @LeonidMew it's barely been 45 minutes since the question was asked, waiting longer is A-OK, a better answer might even come along (like PerlDuck's comment says, it just popped up while I was typing) – Xen2050 Mar 14 at 22:16
  • 6
    I'm curious what you intended for the exec bash to do. – Dennis Williamson Mar 14 at 23:29
27

The culprits are your exec bash statements in some of your functions. The exec statement is a bit weird and not easily understood in the first place. It means: execute the following command instead of the currently running command/shell/script from here on. That is: it replaces the current shell script (in your case) with an instance of bash and it never returns.

You can try this out with a shell and issue

exec sleep 5

This will replace your current shell (the bash) with the command sleep 5 and when that command returns (after 5 seconds) your window will close because the shell has been replaced with sleep 5.

Same with your script: If you put exec something into your script, the script gets replaced with something and when that something stops execution, the whole script stops.

Simply dropping the exec bash statements should do.

  • 2
    @Jenny Nice to hear. Anecdote: The Perl language also has an exec statement with the same behaviour and if you put some statements after an exec statement (like exec something; print "This won't run";) then Perl will warn you that the print statement will never get executed. – PerlDuck Mar 14 at 22:09
  • 8
    BTW congrats on using && after cd, (if you don’t use set -e). I have seen Code like cd tmp; rm -rf * fail horrible – eckes Mar 15 at 0:07
14

From help exec:

exec: exec [-cl] [-a name] [command [arguments ...]] [redirection ...]
    Replace the shell with the given command.

    Execute COMMAND, replacing this shell with the specified program.
    ARGUMENTS become the arguments to COMMAND.  If COMMAND is not specified,
    any redirections take effect in the current shell.

The key word here is replace - if you exec bash from inside a script, no further script execution can occur.

0

if you want a return to the directory you started you could use

cd -

But if you are not sure whether a cd command was executed at all it would be better to use the commands for putting working directories onto a stack:

pushd

and return to it (even after multiple directory changes)

popd

be sure to have equaly pushd and popd commands.

  • 2
    I'm not sure you read or understood the problem space. None of these commands will help the user. – pipe Mar 15 at 13:02
  • I nowhere can read the necessity of executing some commands in a shell. So I thought he called the bash to get back into the start directory. If you insist you might say: he asked for the reason of the missing continuation, which is perfectly answered by the other answers. – Bernd Wilke πφ Mar 17 at 9:46

Your Answer

Jenny is a new contributor. Be nice, and check out our Code of Conduct.

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.