Python script to parse the output Redis INFO command and convert it into a JSON dictionary.

parse_redis_info.py

#!/usr/bin env python
#coding: utf-8

import re
from json import dumps
from sys import (stdin, exit)


DEBUG_LEVEL = 0x0
PARSER = 0x01
POSTPROCESSER = 0x02
PARSER_COMMAS = 0x04
ALL_DICT = 0x08
def debug(level, message):
    if not (level & DEBUG_LEVEL):
        return
    if level & ALL_DICT:
        from pprint import pprint
        pprint(message)
    else:
        prefix = '> ' if level & PARSER else \
                '% ' if level & POSTPROCESSER else \
                '$ ' if level & PARSER_COMMAS else ''
        print '%s%s' % (prefix, message)


EMPTY_LINE = re.compile(r'^\s*$')
#In [25]: k = re.match(r'^# (\w+(\s+\w+)*).*$', '# Melon'); k.group(1)                                                                               
#Out[25]: 'Melon'
#
#In [26]: k = re.match(r'^# (\w+(\s+\w+)*).*$', '# Melon Andador'); k.group(1)
#Out[26]: 'Melon Andador'
#
#In [27]: k = re.match(r'^# (\w+(\s+\w+)*).*$', '# Melon Andador  '); k.group(1)
#Out[27]: 'Melon Andador'
SECTION_LINE = re.compile(r'^# (\w+)\W*$')
#In [29]: k = re.match(r'^(.*):(.*)$', 'cute_key:120.0'); k.group(2)                                                                                 
#Out[29]: '120.0'
#
#In [30]: k = re.match(r'^(.*):(.*)$', 'cute_key:120.0.6'); k.group(2)                                                                               
#Out[30]: '120.0.6'
KEY_VALUE_LINE = re.compile(r'^(\w+)[=:](\S+)$')


def parse_comma_separated_key_values(string):
    '''Convert something like 'ip=172.31.2.6,port=7011,state=online,offset=490698,lag=1' into something like {"ip": "172.31.2.6", "state": "online", "lag": "1", "port": "7011", "offset": "490698"}
    '''
    new_dict = {}
    for pair in string.split(','):
        result = KEY_VALUE_LINE.match(pair)
        key, value = result.group(1), result.group(2)
        new_dict.update({key: value})
    return new_dict

#print parse_comma_separated_key_values('ip=172.31.2.6,port=7011,state=online,offset=490698,lag=1')
#exit(0)


def postprocess(main_dict, regexp, keys_list):
    this_dict = main_dict
    # Climb the dict's tree till the right level
    for key in keys_list:
        try:
            this_dict = this_dict[key]
            debug(POSTPROCESSER, 'kd: %s' % this_dict)
        except KeyError:
            debug(POSTPROCESSER, 'the key "%s" was not found in the dict "%s"' % (key, this_dict))
            return  #fail silently
    # Find the keys which match the regexp
    for key in this_dict.keys():
        if re.match(regexp, key):
            value = this_dict[key]
            debug(POSTPROCESSER, 'key:"%s", value:"%s"' % (key, value))
            this_dict[key] = parse_comma_separated_key_values(value)


# Read all the Redis INFO data from stdin
# convert it into a dictionary
main_dict = {}
this_dict = main_dict
for line in stdin:
    line = line.rstrip('\n').rstrip('\r')  # remove endliners
    debug(PARSER, 'line: \'%s\'' % line)

    # empty lines
    if EMPTY_LINE.match(line):
        debug(PARSER, '> empty line')
        continue

    # NewSection lines
    result = SECTION_LINE.match(line)
    if result:
        section_name = result.group(1).lower()
        this_dict = {}
        main_dict[section_name] = this_dict
        debug(PARSER, '> new section: %s' % section_name)

    # key:value lines
    result = KEY_VALUE_LINE.match(line)
    if result:
        key = result.group(1)
        value = result.group(2)
        this_dict.update({key: value})
        debug(PARSER, '> key_value: \'%s\' and \'%s\'' % (key, value))

# Do some postprocessing for very specific fields
postprocess(main_dict, r'^db\d+$', ['keyspace'])
postprocess(main_dict, r'^slave\d+$', ['replication'])

debug(ALL_DICT, main_dict)

print dumps(main_dict)