Use salt outputters to format pepper output, see https://github.com/saltstack/pepper/issues/3

pepper_format.py

#!/home/bruno/.virtualenvs/salt/bin/python
"""
This script formats pepper output using salt outputters.

It can both format sync and async jobs, and can be used in the following ways:
```
$ pepper \* state.sls core | pepper_format.py -o highstate
$ pepper --client=local_async \* state.highstate | pepper_format.py
$ pepper_format.py 20170504160400750792
```

```
$ pepper_format.py -h
usage: pepper_format.py [-h] [-o OUTPUTTER] [-v] [-p] [-n] [jid]

Format pepper output using any salt outputter

positional arguments:
  jid                   Optional job id

optional arguments:
  -h, --help            show this help message and exit
  -o OUTPUTTER, --outputter OUTPUTTER
                        Outputter to use
  -v, --verbose         Verbose output
  -p, --profiling       Activate profiling
  -n, --no-color        No color in output
```
"""
import subprocess
import sys
import json
import time
import argparse

import salt.config
from salt.output import display_output


def lookup_jid(jid):
    return json.loads(subprocess.check_output(
        ['pepper', '--client=runner', 'jobs.lookup_jid', 'jid=%s' % jid]
        ))['return'][0]

def parse_output(ret):
    outputter = None
    if isinstance(ret, dict) and set(ret) == set(('data', 'outputter')):
        outputter = ret['outputter']
        ret = ret['data']
    return ret, outputter

parser = argparse.ArgumentParser(description='Format pepper output using any salt outputter')
parser.add_argument('jid', nargs='?', help='Optional job id', default=None)
parser.add_argument('-o', '--outputter', help='Outputter to use')
parser.add_argument('-v', '--verbose', action='count', help='Verbose output')
parser.add_argument('-p', '--profiling', action='store_true', help='Activate profiling')
parser.add_argument('-n', '--no-color', action='store_true', help='No color in output')
args = parser.parse_args()

__opts__ = salt.config.minion_config('')
__opts__['color'] = not args.no_color
__opts__['state_output_profile'] = args.profiling
__opts__['state_verbose'] = args.verbose > 0
if args.verbose == 1:
    __opts__['state_output'] = 'changes'

jid = args.jid
if not jid and not sys.stdin.isatty():
    ret = json.load(sys.stdin)['return'][0]
    if 'jid' in ret:
        jid = ret['jid']
    else:
        ret, outputter = parse_output(ret)
        display_output(ret, args.outputter or outputter, __opts__)
        sys.exit(0)

if not jid:
    print('Please specify a job id or provide stdin data to parse')
    sys.exit(1)

print('Looking for job id: %s' % jid)
while True:
    ret, outputter = parse_output(lookup_jid(jid))
    if ret:
        print
        break
    sys.stdout.write('.')
    sys.stdout.flush()
    time.sleep(1)
display_output(ret, args.outputter or outputter, __opts__)

Line 63 should be __opts__['state_verbose'] = args.verbose. I got a None type error. If args.verbose, that would still be falsy. I do not think it could ever be negative.

or __opts__['state_verbose'] = args.verbose AND args.verbose > 0