I have a JSONB column with the following values:

{
  "collections": {
    "drafts": [
      {
        "availabilities": [],
        "languageCodes": [],
        "languages": []
      }
    ],
    "items": [
      {
        "languageCodes": [],
        "languages": []
      },
      {
        "languageCodes": [],
        "languages": []
      } 
    ]
  }
}

I want to delete all of the languages elements using a wildcard JSON path. I can delete one using the ‘#-‘ as follows:

select 
'
{
  "collections": {
    "drafts": [
      {
        "availabilities": [],
        "languageCodes": [],
        "languages": []
      }
    ],
    "items": [
      {
        "languageCodes": [],
        "languages": []
      },
      {
        "languageCodes": [],
        "languages": []
      } 
    ]
  }
}
'::jsonb
#-
'{"collections", "items", "0", "languages"}'

I can also find all of the relavent elements with the following using the ** wildcard:

select jsonb_path_query(
'
{
  "collections": {
    "drafts": [
      {
        "availabilities": [],
        "languageCodes": [],
        "languages": []
      },
      {
        "availabilities": [],
        "languageCodes": [],
        "languages": []
      }
    ],
    "items": [
      {
        "languageCodes": [],
        "languages": []
      },
      {
        "languageCodes": [],
        "languages": []
      } 
    ]
  }
}
','$.**.languages')

However, there doesn’t seem to be a way to delete those elements with one statement.

I have tried using wildcards in the first query, but no combination seems to work. I believe there is probably a way of doing this using jsonb_each(), deleting elements and repacking the original JSONB, but I am wondering if there is something less convoluted.