Delete text from files using regex with sed

Question

I'm attempting to use sed to remove some lines from json files. As you can probably tell, I'm very much an amateur but think I have the regex correct. However sed throws various errors including unterminated address and unterminated 's' command.

The text I'm trying to remove is:

    {
  "trait_type": "Accessories",
  "value": "None"
}

And the regex I have to remove this (trait_type is different throughout different files and so I need to remove this entire block based on the value being "none")

(\{)([\r\n].+)(\r|\n|)(.*(?:None).*)([\r\n].*)

Using online regex testers, the above seems to work perfectly.

The command (and a couple of variations of) that I'm using is:

sed -e 's/\{([\r\n].+)(\r|\n|)(.*(?:None).*)([\r\n].*)//g' 2.json

Could anyone assist at all?

Many thanks

Answer 1

Assuming this is part of a JSON file, the following jq command would find all objects with a trait_type key and a value key and delete all of those objects where the value key has the value None.

jq 'del( .. |
    select(
      type == "object" and
      has("trait_type") and has("value") and 
      .value == "None"
    ) )' file

The command above would write the result of the operation to standard output. Redirect the output to a new file to save it.