Make Instruments play nice with unix by returning > 0 status code if build fails

unix_instruments.sh

#!/usr/bin/env bash
#
# Copyright (c) 2013 Jonathan Penn (http://cocoamanifest.net)
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
#


# unix_instruments
#
# A wrapper around `instruments` that returns a proper unix status code
# depending on whether the run failed or not. Alas, Apple's instruments tool
# doesn't care about unix status codes, so I must grep for the "Fail:" string
# and figure it out myself. As long as the command doesn't output that string
# anywhere else inside it, then it should work.
#
# I use a tee pipe to capture the output and deliver it to stdout
#
# Author: Jonathan Penn (jonathan@cocoamanifest.net)
#

set -e  # Bomb on any script errors

run_instruments() {
  # Pipe to `tee` using a temporary file so everything is sent to standard out
  # and we have the output to check for errors.
  output=$(mktemp -t unix-instruments)
  instruments "$@" 2>&1 | tee $output

  # Process the instruments output looking for anything that resembles a fail
  # message
  cat $output | get_error_status
}

get_error_status() {
  # Catch "Instruments Trace Error"
  # Catch "Instruments Usage Error"
  # Catch "00-00-00 00:00:00 +000 Fail:"
  # Catch "00-00-00 00:00:00 +000 Error:"
  # Catch "00-00-00 00:00:00 +000 None: Script threw an uncaught JavaScript error"
  ruby -e 'exit 1 if STDIN.read =~ /Instruments Usage Error|Instruments Trace Error|^\d+-\d+-\d+ \d+:\d+:\d+ [-+]\d+ (Fail:|Error:|None: Script threw an uncaught JavaScript error)/'
}

# Running this file with "----test" will try to parse an error out of whatever
# is handed in to it from stdin. Use this method to double check your work if
# you need a custom "get_error_status" function above.
if [[ $1 == "----test" ]]; then
  get_error_status
else
  run_instruments "$@"
fi

Just an FYI - we are using this to run instruments at my day job, and it works great unless you have non-ASCII characters in whatever it is you're selecting / looking for. To fix this, we added the following to the top of our jenkins job shell command:

export LANG=en_US.UTF-8
export LANGUAGE=en_US.UTF-8
export LC_ALL=en_US.UTF-8

Since it's limited to the jenkins' shell, it shouldn't break the rest of OS X. It's apparently an issue with ruby itself, and could be fixed breaking the ruby line out into it's own script, but that was more of a hassle.

In the occasional circumstance where instruments gives you an error code and exits out, this will capture it:

  instruments "$@" 2>&1 | tee $output
  ACTUAL_EXIT_STATUS=${PIPESTATUS[0]}
  if [ 0 != $ACTUAL_EXIT_STATUS ]
  then
    exit $ACTUAL_EXIT_STATUS
  fi