jamiepg18 / fromcsv

FromCsv

This module implements an engine for CSV importing.

Compared to other such importers, this lirary is intended to be far more versataile particularly when it comes to situations encountered by users with the matching of CSV data using a hash.

Common user cases

Missing values

Rows with empty data are competely excluded from the upload. If there are no values within a columns that would otherwise be unmatched then that column is ignored.

Missing columns

Missing columns are not considered a reason to abort processing the file.

Reordered columns

Columns can be uploaded in any order.

Unmatched columns

Columns without a match are recorded and be default a file cannot be uploaded with partial matches. Instead a data structure is returned which the caller must validate.

Usage

The engine is ready for use by creating an instance:

const FromCsv = require("fromcsv");

const importer = new FromCsv({
  dialects: {
    some_format: {
      columpMap: {
        "First Column": null,
        "Second Column": null
      }
    }
  }
});

This creates an instance of the importer that can be used to match CSV files of the "some_format" of CSV files. The same instance can be reused for each attempt to process files with the same columns.

fromStream

A text stream containing CSV data can be processed by passing it directly to the fromStream method:

const fs = require("fs");

const inputStream = fs.createReadStream("/path/to/file.csv");

importer.fromStream(inputStream, (err, output) => {
  // ...
});

fromData

To submit an unsuccessful payload to the server to be processed again, the fromData method is used:

const data = {
  header: ["Title", "Surname"],
  rows: [["Mr.", "Smith"], ["Mrs.", "Smith"]]
};

importer.fromData(data, (err, output) => {
  // ...
});