Usage

$ pip install -g jss

$ cat file.json
{
  "foo": [
    "bar",
    {
      "baz": "quux"
    }
  ],
  "wut": {
    "name": "foo",
    "metadata": {
      "owner": "danvk",
      "blah": "whatever"
    }
  },
  "name": "dan"
}

# Pull out all values with key "name", from anywhere in the JSON.
$ jss .name file.json
"foo"
"dan"

# Remove fields named "metadata", wherever they occur (JSON→JSON transform):
$ jss -v .metadata file.json
{
  "foo": [
    "bar",
    {
      "baz": "quux"
    }
  ],
  "wut": {
    "name": "foo"
  },
  "name": "dan"
}

# Keep only fields named "name", plus their ancestors (JSON→JSON transform):
$ jss -k .name file.json
{
  "wut": {
    "name": "foo"
  },
  "name": "dan"
}

# Keep only top-level entries with "whatever" in some value underneath them:
# (JSON→JSON transform using jQuery-style selectors):
$ jss -k ':root>*:has(:contains("whatever"))' file.json
{
  "wut": {
    "name": "foo",
    "metadata": {
      "owner": "danvk",
      "blah": "whatever"
    }
  }
}