PHP script to fetch license data from GitHub for your project's dependencies. It outputs to a simple HTML document.

collectlicenses.php

<?php
/*
Set $gits to an array of repository author/name combinations.

Run it with `php collectlicenses.php`.

It saves the generated HTML to licenses.html in the same folder.
*/

$gits = ["twbs/bootstrap", "npm/npm", "FortAwesome/Font-Awesome", "thomaspark/bootswatch", "composer/composer", "catfan/Medoo", "nwjs/nw.js"];
$url1 = "https://raw.githubusercontent.com/";
$url2s = ["/master/", "/gh-pages/"];
$endings = ["LICENSE", "LICENSE.md", "LICENSE.txt", "license", "license.txt", "package.json", "LICENCE"];

define("LICENSE_SEP", "\n\n=================================\n\n");

/* http://stackoverflow.com/a/7684862/2534036 */
function is_url_exist($url) {
    $ch = curl_init($url);
    curl_setopt($ch, CURLOPT_NOBODY, true);
    curl_exec($ch);
    $code = curl_getinfo($ch, CURLINFO_HTTP_CODE);

    if ($code == 200) {
        $status = true;
    } else {
        $status = false;
    }
    curl_close($ch);
    return $status;
}

echo "\nScraping licenses from GitHub...\n";

$licenses = [];
$maxtries = count($endings) * count($url2s);
foreach ($gits as $repo) {
    $tries = 0;
    $found = false;
    foreach ($url2s as $url2) {
        foreach ($endings as $end) {
            $url = $url1 . $repo . $url2 . $end;

            $handle = curl_init($url);
            curl_setopt($handle, CURLOPT_RETURNTRANSFER, TRUE);
            /* Get the HTML or whatever is linked in $url. */
            $response = curl_exec($handle);
            /* Check for 404 (file not found). */
            $httpCode = curl_getinfo($handle, CURLINFO_HTTP_CODE);
            if ($httpCode == 200) {
                $licenses[$repo] = "";
                if ($end == "package.json") {
                    $json = json_decode($response, true);
                    if (isset($json['license'])) {
                        $lic = $json['license'];
                        if (is_array($lic)) {
                            // legacy
                            foreach ($lic as $l) {
                                $lname = $l['type'];
                                $licenses[$repo] .= file_get_contents("https://spdx.org/licenses/$lname.txt") . (count($lic) > 1 ? LICENSE_SEP : "");
                            }
                        } else {
                            if (preg_match("/\(.+ AND .+\)/", $lic)) {
                                $lic = str_replace("(", "", $lic);
                                $lic = str_replace(")", "", $lic);
                                foreach (explode(" AND ", $lic) as $l) {
                                    $licenses[$repo] .= file_get_contents("https://spdx.org/licenses/$l.txt") . LICENSE_SEP;
                                }
                            } else if (preg_match("/SEE LICENSE IN .+/", $lic)) {
                                $licenses[$repo] = file_get_contents($url1 . $repo . $url2 . str_replace("SEE LICENSE IN ", "", $lic));
                            } else if (is_url_exist("https://spdx.org/licenses/$lic.txt")) {
                                $licenses[$repo] = file_get_contents("https://spdx.org/licenses/$lic.txt");
                            } else {
                                $tries++;
                                echo "Warning: problem parsing package.json for repo $repo (attempt $tries/$maxtries)\n";
                                continue;
                            }
                        }
                        if (is_array($json['author'])) {
                            $licenses[$repo] = str_replace("<copyright holders>", $json['author']['name'], $licenses[$repo]);
                        } else if (isset($json['author'])) {
                            $licenses[$repo] = str_replace("<copyright holders>", explode(" <", $json['author'])[0], $licenses[$repo]);
                        }
                        $licenses[$repo] = str_replace("<year>", date('Y'), $licenses[$repo]);
                        echo "Warning: extracted license ($lic) for $repo from a metadata file.  Please hand-edit and correct any mistakes.\n";
                        $found = true;
                        break;
                    } else {
                        $tries++;
                        echo "Warning: no license data found in package.json for repo $repo (attempt $tries/$maxtries)\n";
                        continue;
                    }
                } else {
                    $licenses[$repo] = $response;
                    $found = true;
                    break;
                }
            } else {
                $tries++;
                echo "Warning: got http code $httpCode for repo $repo (attempt $tries/$maxtries)\n";
            }
            if ($tries > $maxtries) {
                echo "Error: no license file found for repo $repo.\n";
            }
            curl_close($handle);
        }
        if ($found) {
            break;
        }
    }
}

$out = "<!DOCTYPE html>
<html>
    <head>
        <meta charset=\"UTF-8\">
        <meta http-equiv=\"Content-Type\" content=\"text/html; charset=UTF-8\">
        <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">
        <title>Licenses</title>
        <style>
            pre, p {
                white-space: pre-wrap;
                word-wrap: break-word;
                max-width: 600px;
            }
        </style>
    </head>
    <body>
        <h1>Licenses</h1>
        <p>This application contains open-source code from some or all of these projects:</p>";

foreach ($licenses as $repo => $license) {
    $out .= "
        <div>
            <h2>$repo</h2>
            <pre>$license</pre>
        </div>";
}

$out .= "    </body>
</html>";

file_put_contents('licenses.html', $out);

echo "\nDone!\n";