A script for transforming .jsonl file into file ready for bulk import to ES.

es_ndjson_split.rb

# Reads .jsonl file and transforms it into file ready for bulk import to ES.

require 'json'

File.open(ARGV[1], 'wb') {|f|
  ARGF.each_line { |line|
    a = JSON.parse(line)
    out = {
      index: {
        "_id": a["_id"],
        "_index": a["_index"],
      }
    }
    f.write(out.to_json); f.write("\n")
    f.write(a["_source"].to_json); f.write("\n")
  }
}

# run:
# bundle exec ruby es_ndjson_split.rb dest.ndjson < orig_documents.jsonl