'Create json from shell script output

I have a shell script version.sh in which i have code like.

#!/bin/sh
data = `lsb_release -a`
echo $data

it returned me the output like:

Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:    20.04
Codename:   focal

so what I exactly want is want a JSON file to be created .and data should be there like

{'Release' : 20.04 , 'Codename' : focal}

I want to print data in JSON form .can anyone please help me related this ?? I am stuck here.any suggestion any help would be matter.

pythonjsonbashshellsh


Solution 1:[1]

You can use python ttp module to get this data. See the following example:

from ttp import ttp
import json

data_to_parse = """
Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:    20.04
Codename:   focal
"""

ttp_template = """
Release:    {{release}}
Codename:   {{codename}}

"""

parser = ttp(data=data_to_parse, template=ttp_template)
parser.parse()

# print result in JSON format
results = parser.result(format='json')[0]
#print(results)

#converting str to json. 
result = json.loads(results)

print(result)

See the output:

enter image description here

EDITED as requested (without using any package):

data_to_parse = """
Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:    20.04
Codename:   focal
"""

lines = data_to_parse.splitlines()

result = {}

for line in lines:
    if 'Release:' in line:
        line2 = line.split(':')
        result['Release'] = (line2[1].strip())

    elif 'Codename:' in line:
        line2 = line.split(':')
        result['Codename'] = (line2[1].strip())

print(result)

See the result:

enter image description here

EDITED 2. time:

New Data:

data_to_parse = """
Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:    20.04
Codename:   focal

Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:    22.04
Codename:   local

Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:    21.04
Codename:   cocal

"""

The code:

lines = data_to_parse.splitlines()

result_list = []
result2 = {}
result_list3 = []

for line in lines:
    
    if 'Release:' in line:
        line2 = line.split(':')
        result_list.append({'Release': line2[1].strip()})
    
    elif 'Codename:' in line:
        line2 = line.split(':')
        result_list.append({'Codename': line2[1].strip()})
        result2 = {**result_list[0], **result_list[1]}
        result_list3.append(result2)
        result_list = []

print(result_list3)

The output:

enter image description here

Solution 2:[2]

Not tagged or asked but with jq generally being an optimal choice for generating and/or manipulating JSON data in/for the shell (which indeed is tagged, along with bash), here's how a solution using jq could look like:

  • the -R option makes the input lines being interpreted as raw strings
  • the -n option makes the inputs keyword stand for an iterator over all input lines
  • the [inputs | …] array generator creates an array with one item per input line
  • the capture filter applies a regex on an input, and turns the matches into an object
  • the from_entries filter reduces an array of objects with key/value entries into one object
  • the {Release, Codename} object generator creates an object by filtering to only certain fields
lsb_release -a | jq -Rn '
  [ inputs | capture("(?<key>.*):\\s+(?<value>.*)") ]
  | from_entries | {Release, Codename}
'

{
  "Release": "20.04",
  "Codename": "focal"
}

Demo

If desired, adding the -c option would turn the output into a single line:

lsb_release -a | jq -Rnc '…'

{"Release":"20.04","Codename":"focal"}

Demo

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1
Solution 2 pmf

Related Questions