Download Github archives

github-archive.php

#!/usr/bin/env php
<?php

// URL => https://www.githubarchive.org/
//   wget http://data.githubarchive.org/2015-01-{01..30}-{0..23}.json.gz

$dayMonth = [
    1  => 31,
    2  => 30,
    3  => 31,
    4  => 30,
    5  => 31,
    6  => 30,
    7  => 31,
    8  => 31,
    9  => 30,
    10 => 31,
    11 => 30,
    12 => 31,
];

// 2011
for ($day = 2; $day <= 31; $day++) {
    for ($hour = 0; $hour <= 23; $hour++) {
        $command = sprintf('wget http://data.githubarchive.org/2011-12-%1$02d-%2$d.json.gz -O 2011/2011-12-%1$02d-%2$02d.json.gz', $day, $hour);
        
        shell_exec($command);
        echo $command, PHP_EOL;
    }
}

// 2012 -> 2015
for ($year = 2012; $year <= 2015; $year++) {
    for ($month = 1; $month <= 12; $month++){
        $folder = sprintf('%d/%02d', $year, $month);
        if (!file_exists($folder)) {
            mkdir($folder, 0777, true);
        }

        for ($day = 1; $day <= $dayMonth[$month]; $day++) {
            for ($hour = 0; $hour <= 23; $hour++) {
                $command = sprintf('wget http://data.githubarchive.org/%1$04d-%2$02d-%3$02d-%4$d.json.gz -O %1$04d/%2$02d/%1$04d-%2$02d-%3$02d-%4$02d.json.gz', $year, $month, $day, $hour);

                shell_exec($command);
                echo $command, PHP_EOL;
            }
        }
    }
}

// 2016
$year = 2016;
for ($month = 1; $month <= date('m'); $month++){
    $folder = sprintf('%d/%02d', $year, $month);
    if (!file_exists($folder)) {
        mkdir($folder, 0777, true);
    }

    for ($day = 1; $day <= 31; $day++) {
        for ($hour = 0; $hour <= 23; $hour++) {
            $command = sprintf('wget http://data.githubarchive.org/%1$04d-%2$02d-%3$02d-%4$d.json.gz -O %1$04d/%2$02d/%1$04d-%2$02d-%3$02d-%4$02d.json.gz', $year, $month, $day, $hour);

            shell_exec($command);
            echo $command, PHP_EOL;
        }
    }
}