1

Have a json file I am trying to parse from podfox to allow me to rename the downloaded files in a "friendly" way. Here's a snippet of json I am working with:

{
    "episodes": [
        {
             "title": "Hired and Fired by Little Richard and Jimi\u2019s first trip on LSD",
             "url": "https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW2392375869.mp3",
             "downloaded": true,
             "listened": false,
             "published": 1582203660.0
        },
        {
             "title": "Stolen Cars, Broken Taboos, and the Search for Billy Davis",
             "url": "https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW5134475908.mp3",
             "downloaded": true,
             "listened": false,
             "published": 1581598860.0
        },
    ]
    "shortname": "27 Club",
    "title": "27 Club",
    "url": "https://feeds.megaphone.fm/HSW5142951139"
}

I am trying to, based on the url get the title and pass that to a variable in bash. I can (for the most part) use grep but I know that jq is a better method, I just cannot figure out the syntax to make jq work.

This works with grep on the command line: grep -B 1 HSW2392375869.mp3 < feed.json | grep "title" | cut -d"\"" -f4 but seems like it is a potentially error prone solution.

When I try: jq -c '.["episodes"].url' the shell just hangs indefinitely. I do not require to use jq here, so any method to allow me to search for the url and return (ultimately) the value for published and title will do just fine.

Improve this question
0

2 Answers 2

Reset to default
3

Actually first need to filter .episodes then the inner array

 jq ".episodes | .[0]" jsonfile
{
  "title": "Hired and Fired by Little Richard and Jimi’s first trip on LSD",
  "url": "https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW2392375869.mp3",
  "downloaded": true,
  "listened": false,
  "published": 1582203660
}

title:

jq ".episodes| .[0].title" jsonfile
"Hired and Fired by Little Richard and Jimi’s first trip on LSD"

published:

jq ".episodes| .[0].published" jsonfile
1582203660

For query based on url value

jq '.episodes | .[] | select(.url=="https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW2392375869.mp3").title' jsonfile    
"Hired and Fired by Little Richard and Jimi’s first trip on LSD"    

jq '.episodes | .[] | select(.url=="https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW2392375869.mp3").published' jsonfile
1582203660
Improve this answer
3
  • Or even jq '.episodes | map(.title, (.url | sub(".*/"; ""))) ' Commented Jun 16, 2020 at 15:53
  • Aweseome! This is certainly a big step forward, but is there a way, with jq to find the url and then from that dataset get title and published? The entire thing starts with matching a filename to a title. Commented Jun 16, 2020 at 16:14
  • @chow yes, I updated the answer. Commented Jun 16, 2020 at 16:25
0

You want to query your JSON file using some URL and get the date of publication and title corresponding to that URL into shell variables.

url='http://example.com/some/path'

title=unknown
published=unknown

eval "$(
    jq -r --arg url "$url" '.episodes[] | select(.url == $url) |
        @sh "title=\(.title); published=\(.published)"' file.json
)"

printf 'title     = "%s"\n' "$title"
printf 'published = "%s"\n' "$published"

This queries the objects in the episodes array for the one with the correct URL. If found, the jq expression will output shell code containing two variable assignments to the variables title and published. The shell will evaluate these before outputting them with printf.

If no match is found, then the variables would not be changed.

Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.