Frank

It is very important to catch errors from shell commands. We have basically
no error checking within the scripts, as the "set -e" is guaranteed to
catch anything unexpected. These scripts are difficult to debug since
errors are usually encountered by novices, so this kind of thorough
checking is important.

If you temporarily need to disable error checking, then this can be done
locally, by disabling and then re-enabling error checking.

The alternative would be to add explicit error checking to essentially all
commands, in the form of adding "|| exit 1" after most commands.

-erik


On Tue, Jun 9, 2015 at 11:13 AM, <[email protected]> wrote:

> User: knarf
> Date: 2015/06/09 10:13 AM
>
> Modified:
>  /trunk/src/
>   detect.sh
>
> Log:
>  don't abort if shell function returns with a value
>
> File Changes:
>
> Directory: /trunk/src/
> ======================
>
> File [modified]: detect.sh
> Delta lines: +1 -1
> ===================================================================
> --- trunk/src/detect.sh 2015-06-09 14:18:59 UTC (rev 111)
> +++ trunk/src/detect.sh 2015-06-09 15:13:03 UTC (rev 112)
> @@ -8,7 +8,7 @@
>  if [ "$(echo ${VERBOSE} | tr '[:upper:]' '[:lower:]')" = 'yes' ]; then
>      set -x                      # Output commands
>  fi
> -set -e                          # Abort on errors
> +#set -e                          # Abort on errors
>
>  . $CCTK_HOME/lib/make/bash_utils.sh
>
>
> _______________________________________________
> Commits mailing list
> [email protected]
> http://cactuscode.org/mailman/listinfo/commits
>



-- 
Erik Schnetter <[email protected]>
http://www.perimeterinstitute.ca/personal/eschnetter/
_______________________________________________
Users mailing list
[email protected]
http://cactuscode.org/mailman/listinfo/users

Reply via email to