Capturing bash execution errors

Question

I want my bash scripts to detect execution errors and exit.

For the longest time I've been using the try/yell/die approach as seen here.

yell() { echo "$0: $*" >&2; }
die() { echo -e "\e[31m$*\e[0m"; exit 1; }
try() { "$@" || die "cannot $*"; }

However this requires me to wrap my commands line this;

try curl https://blah.com | try bash

It seems the better approach would be to use;

set -e
set -o pipefail
set -E

Are there any downsides to using the set approach, opposed to try/yell/die?

See [Why doesn't set -e (or set -o errexit, or trap ERR) do what I expected?](http://mywiki.wooledge.org/BashFAQ/105) for some related discussion. — Tom Fenech, Jun 20 '16 at 09:38
Thanks for the link @TomFenech, this should be an answer imho! — SleepyCal, Jun 20 '16 at 09:57

score 1 · Answer 1 · edited Jun 20 '20 at 09:12

1

edited Jun 20 '20 at 09:12

Community

answered Jul 14 '16 at 08:15

Armali

score 1 · Answer 2 · answered Jul 26 '19 at 17:16

FYI:

I use ...

#!/usr/bin/env bash

shout() { echo "$0: $*" >&2; }
die() { shout "${@:2} ($1)"; exit $1; }
try() { "$@" || die $? "cannot $*"; }

2 minor differences:

The exit code we eventually receive is the one that was generated by the command being tried.
The shout includes the exit code after the command in (, ).

2 Answers2