ChatGPT can optimize code.

Show and Tell
1

I was working on this code and thought, I really think that could be a recursive function.
But also new that converting it would take me a lot of learning and work.

{
  let year = []
  for (const [nameYear, year_value] of so.entries()) {
    let body = []
    for (const [nameBody, body_value] of year_value.entries()) {
      let motor = []
      for (const [nameMotor, motor_value] of body_value.entries()) {
        motor.push({ name:nameMotor, value: motor_value.length})
      }
      body.push({ name:nameBody, children: motor})
    }  
    year.push({name:nameYear,children:body});
  }
  return year
}

So I asked chatGPT if it could be optimized into a recursive functions.
And it say, yes and here is how it would look.

function rollupData(map) {
  let result = [];
  for (const [key, value] of map.entries()) {
    if (value instanceof Map) {
      result.push({ name: key, children: rollupData(value) });
    } else {
      result.push({ name: key, value: value.length });
    }
  }
  return result;
}

I only was working with three levels in my tree but this code means that ‘so’ can have as many levels as I desire or in any order.

so = d3.group(tsco,
                d=> d.Year_of_Registration_date,
                d=> d.Body_type,
                d=> d.Motor_tech,
                d=> d.CO2_band 
             )

  // Year_of_Registration_date
  // Month_of_Registration_date
  // Body_type
  // Import_status
  // Motor_tech
  // Purchase_region
  // Vehicle_class
  // CO2_band
  // Country
  // Engine_vol
  // GVM
  // Purchase_cust_type
  // Veh_Mfr
  // CO2_WLTP

The data it rolls works with this. Zoomable Treemap / D3 | Observable

2 Likes