Convert a comma-separated value CSV file to Gettext PO format (i18n, internationalization, localization, translation)

csv2po.php

#!/usr/bin/env php
<?php
/**
 * Convert a comma-separated value CSV file to Gettext PO format.
 *
 * @copyright Nick Freear, 2 October 2013.
 */

$path = '/Users/Nick';
$csv_file = $path . "/Downloads/iSpot SPANISH v. luis lopez-sangil - Spanish.csv";

$locale = 'es';
$id = 'iSpot';
$source = basename($csv_file);
$po_file = $path . '/workspace2/'. strtolower($id) .'-csv.'. strtolower($locale) .'.po';

$row_min = 1;    # 0-based.

$col_msgid = 1;  # 0-based.
$col_msgstr= 2;
$col_notes = 3;


$duplicates_check = array();


$output = get_header($locale, $source, $id);

$row = 1;
if (($handle = fopen($csv_file, "r")) !== FALSE) {
    while (($data = fgetcsv($handle, 1000, ",")) !== FALSE) {
        $num = count($data);
        $row++;

        if (! isset($data[$col_msgid]) || ''==$data[$col_msgid]) {
          echo "Skipping empty msgid, row: $row" .PHP_EOL;
          continue;
        }

        $msgid = addslashes($data[$col_msgid]);
        if (in_array($msgid, $duplicates_check)) {
          echo "Skipping duplicate: $msgid" .PHP_EOL;
          continue;
        }
        $duplicates_check[] = $msgid;

        if (isset($data[$col_notes])) {
          $output .= PHP_EOL. "#. Context: $row, ". $data[$col_notes] .PHP_EOL;
        }
        $output .= 'msgid "'. $msgid .'"' .PHP_EOL;
        if (isset($data[$col_msgstr])) {
          $output .= 'msgstr "'. addslashes($data[$col_msgstr]) .'"' .PHP_EOL;
        }
        else {
          $output .= 'msgstr ""' .PHP_EOL;
        }
    }
    fclose($handle);
} else {
  die("Error opening CSV file: $csv_file");
}


$bytes = file_put_contents($po_file, $output);




function get_header($locale = NULL, $source = NULL, $id = NULL, $charset = 'UTF-8') {
    return <<<EOH
# $id language/translation.
#
# Source:  $source
#
# Copyright (c) 2013 The Open University.
# This file is distributed under the same license as the PACKAGE package.
# IET-OU <EMAIL@ADDRESS>, YEAR.
#
#, fuzzy
msgid ""
msgstr ""
"Project-Id-Version: $id\\n"
"Report-Msgid-Bugs-To: iet-webmaster+@+open.ac.uk\\n"
"POT-Creation-Date: 2013-10-02 14:00+0100\\n"
"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\\n"
"Last-Translator: FULL NAME <EMAIL@ADDRESS>\\n"
"Language-Team: LANGUAGE <[email protected]>\\n"
"Language: $locale\\n"
"MIME-Version: 1.0\\n"
"Content-Type: text/plain; charset=$charset\\n"
"Content-Transfer-Encoding: 8bit\\n"
"Plural-Forms: nplurals=INTEGER; plural=EXPRESSION;\\n"


EOH;
}


?>