File: /home/xedaptot/ai.naniguide.com/app/Library/StringHelper.php
<?php
/**
* StringHelper class.
*
* Provide string helper methods
*
* LICENSE: This product includes software developed at
* the Acelle Co., Ltd. (http://acellemail.com/).
*
* @category Acelle Library
*
* @copyright Acelle Co., Ltd
* @license Acelle Co., Ltd
*
* @version 1.0
*
* @link http://acellemail.com
*/
namespace App\Library;
use DOMDocument;
use DomXpath;
use Closure;
use Exception;
use function App\Helpers\handlehtmlentities;
use function App\Helpers\is_non_web_link;
class StringHelper
{
/**
* Custom base64 encoding. Replace unsafe url chars.
*
* @param string $val
*
* @return string
*/
public static function base64UrlEncode($string)
{
if (is_null($string)) {
return null;
}
return str_replace(['+','/','='], ['-','_',''], base64_encode($string));
}
/**
* Custom base64 decode. Replace custom url safe values with normal
* base64 characters before decoding.
*
* @param string $val
*
* @return string
*/
public static function base64UrlDecode($string)
{
if (is_null($string)) {
return null;
}
return base64_decode(str_replace(['-','_'], ['+','/'], $string));
}
/**
* Custom base64 decode. Replace custom url safe values with normal
* base64 characters before decoding.
*
* @param string $val
*
* @return string
*/
public static function cleanupMessageId($msgId)
{
return preg_replace('/[<>\s]*/', '', $msgId);
}
/**
* Custom base64 decode. Replace custom url safe values with normal
* base64 characters before decoding.
*
* @param string $val
*
* @return string
*/
public static function getDomainFromEmail($emailAddr)
{
$s = trim($emailAddr);
// Extract address inside <...> if present
if (preg_match('/<\s*([^>]+)\s*>/u', $s, $m)) {
$s = trim($m[1]);
}
// Get substring after last '@'
$at = strrpos($s, '@');
if ($at === false) {
return null;
}
return trim(substr($s, $at + 1));
}
/**
* Generate MessageId from domain name.
*
* @param string $val
*
* @return string
*/
public static function generateMessageId($domain, $customerUid = null)
{
if (is_null($customerUid)) {
$customerUid = '0000000000000'; // 13 digits
}
return time().rand(100000, 999999).'.'.uniqid().'.'.$customerUid.'@'.$domain;
}
public static function extractCustomerUidFromMessageId(string $messageId)
{
preg_match('/\.(?P<CustomerUid>[a-z0-9]+)@/', $messageId, $matched);
return $matched['CustomerUid'] ?? null;
}
/**
* Custom base64 decode. Replace custom url safe values with normal
* base64 characters before decoding.
*
* @param string $val
*
* @return string
*/
public static function joinUrl()
{
$array = array_map(function ($e) {
return preg_replace('/(^\/+|\/+$)/', '', $e);
}, func_get_args());
return implode('/', $array);
}
/**
* Extract SendGrid X-Message-Id from Smtp-Id
* For example, extract "GuUFV1znQTmkQyPXrPLyxA" from "<[email protected]>".
*
* @param string $val
*
* @return string
*/
public static function extractSendGridMessageId($smtpId)
{
$cleaned = self::cleanupMessageId($smtpId);
return substr($cleaned, 0, strpos($cleaned, '@'));
}
/**
* Detect file encoding.
*
* @param string file path
*
* @return string encoding or false if cannot detect one
*/
public static function detectEncoding($file, $max = 100)
{
// @important: sometimes PHP fails to correctly detect UTF-8 encoding (it returns ISO-8859-1 for a UTF-8 file!)
// So, for safety, we simply return false here
return false;
// legacy code:
$file = fopen($file, 'r');
$sample = '';
$count = 0;
while (!feof($file) && $count <= $max) {
$count += 1;
$sample .= fgets($file);
}
fclose($file);
return mb_detect_encoding($sample, ['UTF-8', 'ISO-8859-1'], true);
}
/**
* Convert from one encoding to the other.
*
* @param string file path
*/
public static function toUTF8($file, $from = 'UTF-8')
{
$content = file_get_contents($file);
$content = mb_convert_encoding($content, 'UTF-8', $from);
file_put_contents($file, $content);
}
/**
* Check if a (UTF-8 encoded) file contains BOM
* Fix it (remove BOM chars) if any.
*
* @param string file path
*/
public static function checkAndRemoveUTF8BOM($file)
{
$bom = pack('H*', 'EFBBBF');
$text = file_get_contents($file);
$matched = preg_match("/^$bom/", $text, $result);
if (!$matched) {
return false;
}
$text = preg_replace("/^$bom/", '', $text);
file_put_contents($file, $text);
return true;
}
// Remove from string, use for email addresses
public static function removeUTF8BOM($text)
{
$bom = pack('H*', 'EFBBBF');
// Standard method
$text = preg_replace("/^$bom/", '', $text);
// More destructive method, as the first method may miss the following ones with more than one BOM:
// [email protected]
// [email protected]
$text = str_replace("\xEF\xBB\xBF", '', $text);
return $text;
}
public static function appendHtml($content, $html, $appendTo = 'body')
{
return StringHelper::updateHtmlWithoutWrapping($content, function ($document) use ($html, $appendTo) {
$tmpDoc = new DOMDocument();
$tmpDoc->encoding = 'utf-8';
$tmpDoc->loadHTML(handlehtmlentities($html), LIBXML_NOWARNING | LIBXML_NOERROR);
$whatToAppend = $tmpDoc->getElementsByTagName('body');
foreach ($whatToAppend->item(0)->childNodes as $node) {
$node = $document->importNode($node, true);
$target = $document->getElementsByTagName($appendTo);
if ($target->length == 0) {
throw new \Exception("Cannot find <$appendTo> tag to append to");
}
$target->item(0)->appendChild($node);
}
});
}
public static function replaceBareLineFeed($content)
{
return preg_replace("/(?<=[^\r])\n/", "\r\n", $content);
}
public static function getRandomUSIpAddresses()
{
$ips = [
'204.113.58.0',
'121.54.104.0',
'213.157.236.0',
'31.142.80.0',
'212.95.142.0',
'172.237.145.0',
'202.3.232.0',
'123.50.72.0',
'62.162.72.0',
'195.76.126.0',
'104.235.112.0',
'203.84.183.0',
'79.107.195.0',
'197.210.185.0',
'216.9.154.0',
'173.25.32.0',
'156.157.173.0',
'31.132.224.0',
'23.18.2.0',
'58.181.96.0',
'201.190.47.0',
'208.84.83.0',
'213.101.168.0',
'201.217.11.0',
'212.235.251.0',
'153.215.187.0',
'217.131.28.0',
'203.128.83.0',
'37.218.168.0',
'61.94.163.0',
'188.140.171.0',
'188.135.24.0',
'89.135.185.0',
'84.237.238.0',
'118.97.136.0',
'61.5.40.0',
'87.110.26.0',
'87.110.41.0',
'41.224.103.0',
'125.139.107.0',
'59.98.0.0',
'222.45.112.0',
'85.71.183.0',
'1.9.100.0',
'213.100.103.0',
'201.252.22.0',
'201.240.143.0',
'201.216.155.0',
'212.52.128.0',
'213.154.39.0',
'59.129.59.0',
'213.14.139.0',
'182.1.62.0',
'93.187.176.0',
'89.142.48.0',
'36.37.133.0',
'58.157.102.0',
'41.74.46.0',
'205.242.198.0',
'59.104.0.0',
'212.156.99.0',
'83.239.122.0',
'88.229.68.0',
'94.66.154.0',
'120.89.125.0',
'220.224.146.0',
'84.112.150.0',
'31.206.151.0',
'37.26.98.0',
'94.67.116.0',
'202.188.148.0',
'109.178.2.0',
'115.133.224.0',
'74.128.61.0',
'125.162.76.0',
'109.76.207.0',
'31.209.96.0',
'194.157.4.0',
'94.25.40.0',
'94.66.59.0',
'213.87.251.0',
'37.233.130.0',
'31.144.173.0',
'89.47.78.0',
'43.229.13.0',
'217.16.81.0',
'203.210.235.0',
'88.231.42.0',
'213.179.245.0',
'31.128.41.0',
'197.231.251.0',
'89.143.77.0',
'78.171.254.0',
'59.131.117.0',
'90.139.213.0',
'87.229.46.0',
'217.73.140.0',
'188.245.148.0',
'94.66.206.0',
'36.252.32.0',
'212.26.188.0',
'217.107.126.0',
'196.200.61.0',
'31.152.23.0',
'223.205.207.0',
'217.117.176.0',
'36.37.231.0',
'202.53.146.0',
'202.8.73.0',
'180.54.50.0',
'103.30.138.0',
'41.63.223.0',
'41.73.212.0',
'31.141.120.0',
'31.128.90.0',
'212.90.50.0',
'202.252.244.0',
'203.104.24.0',
'80.249.68.0',
'196.22.6.0',
'205.160.111.0',
'116.96.30.0',
'117.242.81.0'
];
return $ips[array_rand($ips)];
}
public static function purifyHtml($html, $tags = [])
{
$defaults = ['script'];
$defaults = array_merge($defaults, $tags);
$dom = new DOMDocument();
libxml_use_internal_errors(true); // disable warning for invalid tags
$dom->loadHTML($html, LIBXML_HTML_NODEFDTD | LIBXML_HTML_NOIMPLIED);
foreach ($defaults as $tag) {
$itemsToRemove = $dom->getElementsByTagName($tag);
$remove = [];
foreach ($itemsToRemove as $item) {
$remove[] = $item;
}
foreach ($remove as $item) {
$item->parentNode->removeChild($item);
}
}
return $dom->saveHTML();
}
// @Important: use this helper method whenever it comes to update DOM document
// This method already solves the following:
// + UTF8 characters are wrongly encoded
// + Ampersand & is encoded to &, breaking URLs
// + Others?
public static function updateHtml(string $content, Closure $callback)
{
$document = new DOMDocument();
$document->encoding = 'utf-8';
$document->loadHTML(handlehtmlentities($content), LIBXML_NOWARNING | LIBXML_NOERROR);
$callback($document);
// IMPORTANT:
// Do not use urldecode() here, use rawurldecode() instead
// Otherwise, it will encode plus (+) chars to spaces, breaking IMG tag with Base64 src content:
// For example:
// <img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA+kAAA..."
// Would become:
// <img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA kAAA..."
return self::saveHTMLWithUTF8AndDoctype($document);
}
public static function updateHtmlWithoutWrapping(string $content, Closure $callback)
{
$document = new DOMDocument();
$document->encoding = 'utf-8';
$document->loadHTML(handlehtmlentities($content), LIBXML_NOWARNING | LIBXML_NOERROR | LIBXML_HTML_NOIMPLIED);
$callback($document);
// IMPORTANT:
// Do not use urldecode() here, use rawurldecode() instead
// Otherwise, it will encode plus (+) chars to spaces, breaking IMG tag with Base64 src content:
// For example:
// <img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA+kAAA..."
// Would become:
// <img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA kAAA..."
return rawurldecode($document->saveHTML($document->documentElement));
}
public static function saveHTMLWithUTF8AndDoctype($document)
{
// Do not use "htmlspecialchars_decode()" here, which transforms ">" to ">"
// as it will crash with PHP's DOMDocument
// Use it as an explicit & final step before rendering HTML content
$output = rawurldecode($document->saveHTML($document->documentElement));
$output = '<!DOCTYPE html>'.$output; // saveHTML(params) removes DOCTYPE
return $output;
}
public static function sanitizeFilename($filename)
{
return preg_replace('/[^a-zA-Z0-9\.\-]+/', '_', trim($filename));
}
public static function generateUniqueName($directory, $name)
{
$count = 1;
$path = join_paths($directory, $name);
$newName = $name;
while (file_exists($path)) {
$regxp = '/(?<ext>\.[^\/\.]+$)/';
preg_match($regxp, $name, $matched);
if (array_key_exists('ext', $matched)) {
$fileExt = $matched['ext'];
} else {
$fileExt = '';
}
$base = preg_replace($regxp, '', $name);
$newName = $base.'_'.$count.$fileExt;
$path = join_paths($directory, $newName);
$count += 1;
}
return $newName;
}
public static function isTag($string)
{
return preg_match('/{[a-zA-Z0-9_]+}/', $string);
}
public static function containsUnsubscribeTag(string $content): bool
{
// Match canonical Twig forms only: `{{ unsubscribe_url }}` (lowercase)
// and `{{ UNSUBSCRIBE_URL }}` (uppercase). Mixed casing would pass an
// /i regex but fail to render — TagTranslationService::duplicateUppercaseKeys
// only injects the strtoupper() variant, so `{{ Unsubscribe_url }}` resolves
// to empty string in the send pipeline and recipients see the literal tag.
return (bool) preg_match('/\{\{\s*(?:unsubscribe_url|UNSUBSCRIBE_URL)\s*\}\}/', $content);
}
public static function fromHumanIpAddress($ipAddress)
{
$googleIpRanges = config('google');
foreach ($googleIpRanges as $cidr) {
if (\Symfony\Component\HttpFoundation\IpUtils::checkIp($ipAddress, $cidr)) {
return false;
}
}
return true;
}
public static function replaceUrl($html, Closure $process, $throwIfInvalidUrl = true, $throwIfAttrMissing = false, $throwIfAttrEmpty = false)
{
return static::updateHtmlWithoutWrapping($html, function ($document) use ($process, $throwIfInvalidUrl, $throwIfAttrMissing, $throwIfAttrEmpty) {
$searchFor = [
'link' => 'href',
'a' => 'href',
'img' => 'src',
'video' => 'src',
'audio' => 'src',
'source' => 'src',
'track' => 'src',
'embed' => 'src',
];
foreach ($searchFor as $tag => $attr) {
$elements = $document->getElementsByTagName($tag);
foreach ($elements as $element) {
if (!$element->hasAttribute($attr)) {
if ($throwIfAttrMissing) {
throw new \Exception("Cannot find '{$attr}' attribute for a <{$element->tagName}> tag at: ".$document->saveHTML($element), 1);
} else {
continue;
}
}
$url = trim($element->getAttribute($attr));
if (empty($url)) {
if ($throwIfAttrEmpty) {
throw new \Exception("'{$attr}' value of an <{$element->tagName}> tag is empty, at: ".$document->saveHTML($element), 1);
} else {
continue;
}
}
// If the given URL is a non-web link, like <a href="tel:"... (mailto:, ftp: etc.)
if (is_non_web_link($url)) {
continue;
}
// parse_url($url, PHP_URL_HOST) returns null when the URL is relative (e.g. "/abc/xyz" or "abc/xyz")
$host = parse_url($url, PHP_URL_HOST);
if ($host === false) {
if ($throwIfInvalidUrl) {
throw new \Exception("'{$attr}' value of an <{$element->tagName}> tag is an invalid url: {$url}, at".$document->saveHTML($element), 1);
} else {
continue;
}
}
/*
if (is_null($host)) {
$urlWithHost = join_url($this->getAppUrl(), $url);
$element->setAttribute($attr, $urlWithHost);
}
*/
$updatedUrl = $process($element, $url);
$element->setAttribute($attr, $updatedUrl);
}
}
});
}
/**
* Recursively find and modify *all* nodes in a nested associative array.
*
* - Traverses any structure (no assumption about "children" key)
* - Applies closure to every node where [$keyToFind] === $valueToFind
*
* @param array &$data
* @param string $keyToFind
* @param mixed $valueToFind
* @param Closure $closure function (&$element) { ... }
* @return int Number of modified nodes
*/
public static function iterate(array &$data, Closure $closure): int
{
$count = 1;
$closure($data);
foreach ($data as &$value) {
if (is_array($value)) {
$count += static::iterate($value, $closure);
}
}
return $count;
}
}