laravel-csv

This package provides a Laravel Facade for writing/reading Csv files.

Installation

$ composer require jdefez/laravel-csv

Reading a Csv file

This utility class does not take hold of any data. It simply provides an iterator you can use to read your csv files.

Basic usage

use Jdefez\LaravelCsv\Facades\Csv;

$file = new SplFileObject('path-to-my-file.csv', 'r');

if ($file->isReadable()) {
  $reader = Csv::reader($file);

  foreach ($reader->read() as $row) {

    // returns an array with the row's values

  }
}

Reading the first row

By default the first row is skipped. If you need to read the first row use $reader->withHeadings() method.

$reader = Csv::reader(new SplFileObject('path-to-my-file.csv', 'r'))
  ->withHeadings();

Reader::keyByColumnName()

The rows will be returned under the form of an associative arrays with the camel cased columns names as keys.

// Given a file
//
// lastname;firstname;date of birth
// Jacky;Terror;1875-02-12
// Julian;Nightmare;1815-11-11

$reader = Csv::reader(new SplFileObject('path-to-my-file.csv', 'r'))
  ->keyByColumnName()

foreach ($reader->read() as $row) {

  //array(
  //    "firstname" => "Jacky",
  //    "lastname" => "Terror",
  //    "date_of_birth" => "1875-02-12"
  //)

  //...

}

Reader::toObject()

The rows will be casted to object using the kamel cased column names as properties.

// Given a file
//
// lastname;firstname;birthdate
// Jacky;Terror;1875-02-12
// Julian;Nightmare;1815-11-11

$reader = Csv::reader(new SplFileObject('path-to-my-file.csv', 'r'))
  ->toObject()

foreach ($reader->read() as $row) {

  //object(stdClass)#277 (2) {
  //    ["firstname"]=> string(4) "Jacky"
  //    ["lastname"]=> string(5) "Terror"
  //    ["date_of_birth"]=> string(13) "1875-02-12"
  //}

  // ...
}

Mapping data

If you need a more sophisticated way to map your file. You could use Reader::read(?callable $callback = null): Generator

$iterator = Csv::reader(new SplFileObject('path-to-my-file.csv', 'r'))
  ->toObject()
  ->read(fn ($item) => UserDataBuilder::make(
    $item->firstname,
    $item->lastname,
    $item->date_of_birth,
  );

foreach ($iterator as $userData) {
  if ($userData->isValid()) {
    User::create($userData);
  }
}

Fixing enconding

For this feature to work, you need to provide a list of expected encodings. They will be used to detect the current line encoding and if it has to be fixed. By default the Reader uses: ['ISO-8859-15', 'ISO-8859-1']

// Fixing encoding from ISO to UTF-8

$reader = $reader->setToEncoding('UTF-8')
    ->setSearchEncodings(['ISO-8859-15', 'ISO-8859-1'])
    ->toObject();

Writing a Csv file

You can both work with SplFileObject or SplTempFileObject.

This gist demonstrates how you can use SplTempFileObject

Writing an entire collection of data

$collection = collect([
  ['Jacky', 'Terror', '1875-02-12'],
  ['Julian', 'Nightmare', '1815-11-11'],
  // ...
]);

Csv::writer()
  ->setFile(new SplFileObject('path-to-my-file.csv', 'w'))
  ->setColumns(['firstname', 'lastname', 'date of birth'])
  ->setData($collection)
  ->write();

Writing line by line.

$collection = collect([
  ['firstname', 'lastname', 'date_of_birth'],
  ['Jacky', 'Terror', '1875-02-12'],
  ['Julian', 'Nightmare', '1815-11-11'],
  // ...
]);

$writer = Csv::writer()->setFile(new SplFileObject('path-to-my-file.csv', 'w'));

$collection->each(fn ($line) => $writer->put($line));

Mapping data

You can also map data when writing to the file with Writer::write(callable $callback) of Writer::put(array|callable $row).

$models = Users::all();

$writer = Csv::writer(new SplFileObject('path-to-my-file.csv', 'w'));

$writer->setData($models)
  ->write(fn ($item) => [
    $item->firstname,
    $item->lastname,
    $item->birthday->format('Y-m-d')
  ]);

// Or iterate over the collection and append each line to the file.

$models->each(fn ($model) => $writer->put(fn () => [
    $model->firstname,
    $model->lastname,
    $model->birthday->format('Y-m-d')
]);

Todo:

Reader:

• Adding the ability to setup headings names. It could be a convinient way to map data. Especialy if we want to key by columns names or cast rows to stdClass when there are no columns names at all.

Writer:

• Writing to a given encoding