Ytukyg

I am new to nodejs and try to write a program which read a csv file and can select data from csv_file by columns names and convert them into arrays.

I figured how to read the csv file and convert the columns into arrays(Thanks internet), but the columns are selected by the index number rather than the names.

code:

var csv = require('csv');

var csv_obj = csv();



function Column_data(signal_name, signal_type, initial_value, minimum, maximum  ) {

    this.signal_name = signal_name;

    this.signal_type = signal_type;

    this.initial_value = initial_value;

    this.minimum = minimum;

    this.maximum = maximum;

};



var csv_data = ;



csv_obj.from.path('../data_files/Signals_Info.csv).to.array(function (data) {

    for (var row = 0; row < data.length; index++) {

        csv_data.push(new Column_data(data[row][0], data[row][1], data[row][2], data[row][3], data[row][4]));

    }

    console.log(csv_data);

});

Here I'm using index values 1, 2, 3 to access columns 1, 2 and 3 in my csv. I want to be able to access columns by name rather than by index number because index can change.

The following code gives you two options. One is using the 4.0.0 version of the csv library (npm install csv), the other is doing it by hand. Both return the same results given the sample input included below:

const csv = require('csv')



const data = `One,Two,Three

1,2,3

4,5,6`



// Using the csv library:

csv.parse(

  data,

  { columns: true },

  (err, result) => console.log(result)

)





// Doing it manually:

const rowToObject = (headers, cells) =>

  headers.reduce(

    (acc, header, i) => {

      acc[header] = cells[i]

      return acc

    },

    {}

  )



const csvToObjects = file => {

  const [headerRow, ...dataRows] = file.split('n')

  const headers = headerRow.split(',')

  return dataRows.map(

    row => rowToObject(headers, row.split(','))

  )

}



console.log(csvToObjects(data))



// both options output [{One:1,Two:2,Three:3},{One:4,Two:5,Three:6}]

You can see both of these running in this runkit - apologies, I'm not able to run StackOverflow snippets in my browser at this time.

Here I'll mention a third option: It seems you're using an older version of the csv package. In addition to csv().from.path(path).to.array(options, callback), the module also offers a to.object(options, callback) method, but I'm having trouble finding the documentation on the older version (and don't even know exactly which version you're currently using, which makes things even more difficult).

There is no way to access the columns by the column_names. When the csv is read, it's just all comma-separated. Instead, you can change the csv output to be read by the column names, like -

csv_obj.from.path('../data_files/Signals_Info.csv).to.array(function (data) {

    csv_data['signal_name'] = ;

    csv_data['signal_type'] = ;

    csv_data['initial_value'] = ;

    for (var index = 1; index < data.length; index++) {

        csv_data['signal_name'].push(data[index][0]);

        csv_data['signal_type'].push(data[index][1]);

        csv_data['initial_value'].push(data[index][2]);

        ...

    }

    console.log(csv_data);

});

Try using Papaparse

and making a reader function like this first -

function reader (csv) {

        return new Promise((resolve, reject) => {

            if (csv) {

                parser(csv, {

                ...config,  // . this could be as you want it to - read the docs

                complete:(response) => resolve(response.data),

                error:(error) => reject(error),

             })

          }

        })

     }

contents = reader(csv)

csv_obj = contents.map((row) => ({

signal_name: row[0],

signal_type: row[1], 

initial_value: row[2],

minimum: row[3],

maximum: row[4],



        }))