mirror of
https://github.com/manualdousuario/marreta.git
synced 2026-10-19 07:23:34 +00:00
Revert "Fix Invalid URL format error"
This commit is contained in:
Renan Altendorf
GitHub
parent
a8df04e4c2
commit
cee85c3ef8
8 changed files with 133 additions and 73 deletions
|
|
@ -327,7 +327,7 @@ main {
|
|||
|
||||
input {
|
||||
background-color: var(--input-bg);
|
||||
padding: 16px 70px 16px 44px;
|
||||
padding: 16px 0 16px 44px;
|
||||
border: 0;
|
||||
border-radius: 8px;
|
||||
width: 100%;
|
||||
|
|
|
|||
2
app/dist/css/style.css
vendored
2
app/dist/css/style.css
vendored
File diff suppressed because one or more lines are too long
2
app/dist/css/style.css.map
vendored
2
app/dist/css/style.css.map
vendored
File diff suppressed because one or more lines are too long
|
|
@ -1,2 +1,52 @@
|
|||
User-agent: *
|
||||
Disallow: /
|
||||
Allow: /
|
||||
|
||||
User-agent: *
|
||||
Disallow: /p/
|
||||
Disallow: /api/
|
||||
|
||||
User-agent: AI2Bot
|
||||
User-agent: Ai2Bot-Dolma
|
||||
User-agent: Amazonbot
|
||||
User-agent: anthropic-ai
|
||||
User-agent: Applebot
|
||||
User-agent: Applebot-Extended
|
||||
User-agent: Bytespider
|
||||
User-agent: CCBot
|
||||
User-agent: ChatGPT-User
|
||||
User-agent: Claude-Web
|
||||
User-agent: ClaudeBot
|
||||
User-agent: cohere-ai
|
||||
User-agent: cohere-training-data-crawler
|
||||
User-agent: Crawlspace
|
||||
User-agent: Diffbot
|
||||
User-agent: DuckAssistBot
|
||||
User-agent: FacebookBot
|
||||
User-agent: FriendlyCrawler
|
||||
User-agent: Google-Extended
|
||||
User-agent: GoogleOther
|
||||
User-agent: GoogleOther-Image
|
||||
User-agent: GoogleOther-Video
|
||||
User-agent: GPTBot
|
||||
User-agent: iaskspider/2.0
|
||||
User-agent: ICC-Crawler
|
||||
User-agent: ImagesiftBot
|
||||
User-agent: img2dataset
|
||||
User-agent: ISSCyberRiskCrawler
|
||||
User-agent: Kangaroo Bot
|
||||
User-agent: Meta-ExternalAgent
|
||||
User-agent: Meta-ExternalFetcher
|
||||
User-agent: OAI-SearchBot
|
||||
User-agent: omgili
|
||||
User-agent: omgilibot
|
||||
User-agent: PanguBot
|
||||
User-agent: PerplexityBot
|
||||
User-agent: PetalBot
|
||||
User-agent: Scrapy
|
||||
User-agent: SemrushBot
|
||||
User-agent: Sidetrade indexer bot
|
||||
User-agent: Timpibot
|
||||
User-agent: VelenPublicWebCrawler
|
||||
User-agent: Webzio-Extended
|
||||
User-agent: YouBot
|
||||
Disallow: /
|
||||
|
|
@ -17,6 +17,7 @@ self.addEventListener('share_target', (event) => {
|
|||
event.respondWith((async () => {
|
||||
const formData = await event.request.formData();
|
||||
const url = formData.get('url') || '';
|
||||
return Response.redirect('/?url=' + encodeURIComponent(url), 303);
|
||||
const redirectUrl = `/p/${encodeURIComponent(url)}`;
|
||||
return Response.redirect(redirectUrl, 303);
|
||||
})());
|
||||
});
|
||||
|
|
|
|||
|
|
@ -24,29 +24,17 @@ class Router
|
|||
{
|
||||
$this->dispatcher = FastRoute\simpleDispatcher(function(FastRoute\RouteCollector $r) {
|
||||
// Main route - home page
|
||||
$r->addRoute(['GET'], '/', function() {
|
||||
$r->addRoute(['GET','POST'], '/', function() {
|
||||
require_once __DIR__ . '/../config.php';
|
||||
require_once __DIR__ . '/../inc/Cache.php';
|
||||
require_once __DIR__ . '/../inc/Language.php';
|
||||
|
||||
\Inc\Language::init(LANGUAGE);
|
||||
|
||||
// Check GET param
|
||||
$url = isset($_GET['url']) ? trim($_GET['url']) : '';
|
||||
if (!empty($url)) {
|
||||
$sanitizedUrl = $this->processUrl($url);
|
||||
if (empty($sanitizedUrl)) {
|
||||
header('Location: /?message=INVALID_URL');
|
||||
exit;
|
||||
}
|
||||
|
||||
header('Location: ' . SITE_URL . '/p/' . urlencode($sanitizedUrl));
|
||||
exit;
|
||||
}
|
||||
|
||||
|
||||
$message = '';
|
||||
$message_type = '';
|
||||
|
||||
$url = '';
|
||||
|
||||
// Sanitize and process query string messages
|
||||
if (isset($_GET['message'])) {
|
||||
$message_key = htmlspecialchars(trim($_GET['message']), ENT_QUOTES | ENT_HTML5, 'UTF-8');
|
||||
|
|
@ -54,26 +42,45 @@ class Router
|
|||
$message = htmlspecialchars($messageData['message'], ENT_QUOTES | ENT_HTML5, 'UTF-8');
|
||||
$message_type = htmlspecialchars($messageData['type'], ENT_QUOTES | ENT_HTML5, 'UTF-8');
|
||||
}
|
||||
|
||||
|
||||
// Process form submission
|
||||
if ($_SERVER['REQUEST_METHOD'] === 'POST' && isset($_POST['url'])) {
|
||||
$originalUrl = trim($_POST['url']);
|
||||
if (filter_var($originalUrl, FILTER_VALIDATE_URL)) {
|
||||
$sanitizedUrl = $this->sanitizeUrl($originalUrl);
|
||||
if (!empty($sanitizedUrl)) {
|
||||
header('Location: ' . SITE_URL . '/p/' . $sanitizedUrl);
|
||||
exit;
|
||||
}
|
||||
}
|
||||
$messageData = \Inc\Language::getMessage('INVALID_URL');
|
||||
$message = $messageData['message'];
|
||||
$message_type = $messageData['type'];
|
||||
}
|
||||
|
||||
// Initialize cache for counting
|
||||
$cache = new \Inc\Cache();
|
||||
$cache_folder = $cache->getCacheFileCount();
|
||||
|
||||
|
||||
require __DIR__ . '/views/home.php';
|
||||
});
|
||||
|
||||
// API route - uses URLProcessor in API mode
|
||||
$r->addRoute('GET', '/api/{url:.+}', function($vars) {
|
||||
$url = urldecode($vars['url']);
|
||||
|
||||
$sanitizedUrl = $this->processUrl($url);
|
||||
if (empty($sanitizedUrl)) {
|
||||
|
||||
if (!preg_match('#^https?://#', $url)) {
|
||||
$url = 'https://' . $url;
|
||||
}
|
||||
|
||||
if (filter_var($url, FILTER_VALIDATE_URL)) {
|
||||
$sanitizedUrl = $this->sanitizeUrl($url);
|
||||
$processor = new URLProcessor($sanitizedUrl, true);
|
||||
$processor->process();
|
||||
} else {
|
||||
header('Location: /?message=INVALID_URL');
|
||||
exit;
|
||||
}
|
||||
|
||||
$processor = new URLProcessor($sanitizedUrl, true);
|
||||
$processor->process();
|
||||
});
|
||||
|
||||
// API route without parameters - redirects to root
|
||||
|
|
@ -85,41 +92,58 @@ class Router
|
|||
// Processing route - uses URLProcessor in web mode
|
||||
$r->addRoute('GET', '/p/{url:.+}', function($vars) {
|
||||
require_once __DIR__ . '/../config.php';
|
||||
|
||||
|
||||
$url = urldecode($vars['url']);
|
||||
|
||||
$sanitizedUrl = $this->processUrl($url);
|
||||
if (empty($sanitizedUrl)) {
|
||||
|
||||
$originalUrl = $vars['url'];
|
||||
$needsRedirect = false;
|
||||
|
||||
if ($originalUrl !== $url || preg_match('#^https?://#', $url)) {
|
||||
$needsRedirect = true;
|
||||
}
|
||||
|
||||
if (!preg_match('#^https?://#', $url)) {
|
||||
$url = 'https://' . $url;
|
||||
}
|
||||
|
||||
if (filter_var($url, FILTER_VALIDATE_URL)) {
|
||||
$sanitizedUrl = $this->sanitizeUrl($url);
|
||||
|
||||
if ($needsRedirect && !empty($sanitizedUrl)) {
|
||||
header('Location: ' . SITE_URL . '/p/' . $sanitizedUrl);
|
||||
exit;
|
||||
}
|
||||
|
||||
$processor = new URLProcessor($sanitizedUrl, false);
|
||||
$processor->process();
|
||||
} else {
|
||||
header('Location: /?message=INVALID_URL');
|
||||
exit;
|
||||
}
|
||||
|
||||
$processor = new URLProcessor($sanitizedUrl, false);
|
||||
$processor->process();
|
||||
});
|
||||
|
||||
|
||||
// Processing route with query parameter or without parameters
|
||||
$r->addRoute('GET', '/p[/]', function() {
|
||||
if (isset($_GET['url']) || isset($_GET['text'])) {
|
||||
$originalUrl = isset($_GET['url']) ? trim($_GET['url']) : '';
|
||||
$originalText = isset($_GET['text']) ? trim($_GET['text']) : '';
|
||||
|
||||
if (!empty($originalUrl)) {
|
||||
$sanitizedUrl = $this->processUrl($originalUrl);
|
||||
|
||||
if (!empty($originalUrl) && filter_var($originalUrl, FILTER_VALIDATE_URL)) {
|
||||
$sanitizedUrl = $this->sanitizeUrl($originalUrl);
|
||||
if (!empty($sanitizedUrl)) {
|
||||
header('Location: /p/' . urlencode($sanitizedUrl));
|
||||
header('Location: /p/' . $sanitizedUrl);
|
||||
exit;
|
||||
}
|
||||
} else if (!empty($originalText)) {
|
||||
$sanitizedUrl = $this->processUrl($originalText);
|
||||
if (!empty($sanitizedUrl)) {
|
||||
header('Location: /p/' . urlencode($sanitizedUrl));
|
||||
} elseif (!empty($originalText) && filter_var($originalText, FILTER_VALIDATE_URL)) {
|
||||
$sanitizedText = $this->sanitizeUrl($originalText);
|
||||
if (!empty($sanitizedText)) {
|
||||
header('Location: /p/' . $sanitizedText);
|
||||
exit;
|
||||
}
|
||||
} else {
|
||||
header('Location: /?message=INVALID_URL');
|
||||
exit;
|
||||
}
|
||||
|
||||
header('Location: /?message=INVALID_URL');
|
||||
exit;
|
||||
}
|
||||
header('Location: /');
|
||||
exit;
|
||||
|
|
@ -132,24 +156,6 @@ class Router
|
|||
});
|
||||
}
|
||||
|
||||
/**
|
||||
* Process and sanitize a valid URL
|
||||
* @param string $url
|
||||
* @return string
|
||||
*/
|
||||
private function processUrl(string $url): string
|
||||
{
|
||||
if (!preg_match('#^https?://#', $url)) {
|
||||
$url = 'https://' . $url;
|
||||
}
|
||||
|
||||
if (!filter_var($url, FILTER_VALIDATE_URL)) {
|
||||
return '';
|
||||
}
|
||||
|
||||
return $this->sanitizeUrl($url);
|
||||
}
|
||||
|
||||
/**
|
||||
* Sanitizes and normalizes URLs
|
||||
* @param string $url The URL to sanitize and normalize
|
||||
|
|
@ -193,7 +199,8 @@ class Router
|
|||
$cleanedUrl = preg_replace('/[\x00-\x1F\x7F]/', '', $cleanedUrl);
|
||||
$cleanedUrl = filter_var($cleanedUrl, FILTER_SANITIZE_URL);
|
||||
|
||||
return $cleanedUrl;
|
||||
// Convert special characters to HTML entities
|
||||
return htmlspecialchars($cleanedUrl, ENT_QUOTES | ENT_HTML5, 'UTF-8');
|
||||
}
|
||||
|
||||
/**
|
||||
|
|
|
|||
|
|
@ -2,10 +2,10 @@
|
|||
|
||||
namespace App;
|
||||
|
||||
use Inc\Cache;
|
||||
use Inc\Language;
|
||||
use Inc\URLAnalyzer;
|
||||
use Inc\URLAnalyzer\URLAnalyzerException;
|
||||
use Inc\Cache;
|
||||
|
||||
/**
|
||||
* URL Processor
|
||||
|
|
@ -81,7 +81,8 @@ class URLProcessor
|
|||
if (!$this->isApi) {
|
||||
$redirectInfo = $this->analyzer->checkStatus($this->url);
|
||||
if ($redirectInfo['hasRedirect'] && $redirectInfo['finalUrl'] !== $this->url) {
|
||||
$this->redirect(SITE_URL . '/p/' . urlencode($redirectInfo['finalUrl']));
|
||||
$finalUrl = preg_replace('#^https?://#', '', $redirectInfo['finalUrl']);
|
||||
$this->redirect(SITE_URL . '/p/' . urlencode($finalUrl));
|
||||
}
|
||||
}
|
||||
|
||||
|
|
@ -89,8 +90,9 @@ class URLProcessor
|
|||
$content = $this->analyzer->analyze($this->url);
|
||||
|
||||
if ($this->isApi) {
|
||||
$displayUrl = preg_replace('#^https?://#', '', $this->url);
|
||||
$this->sendApiResponse([
|
||||
'url' => SITE_URL . '/p/' . urlencode($this->url),
|
||||
'url' => SITE_URL . '/p/' . $displayUrl
|
||||
]);
|
||||
} else {
|
||||
echo $content;
|
||||
|
|
@ -123,7 +125,7 @@ class URLProcessor
|
|||
$url = ''; // Initialize url variable for the view
|
||||
|
||||
// Initialize cache for counting
|
||||
$cache = new Cache();
|
||||
$cache = new \Inc\Cache();
|
||||
$cache_folder = $cache->getCacheFileCount();
|
||||
|
||||
require __DIR__ . '/views/home.php';
|
||||
|
|
|
|||
|
|
@ -68,7 +68,7 @@
|
|||
<p class="walls_destroyed">
|
||||
<strong><?php echo number_format($cache_folder, 0, ',', '.'); ?></strong> <span><?php echo \Inc\Language::get('walls_destroyed'); ?></span>
|
||||
</p>
|
||||
<form id="urlForm" method="GET" action="/" class="space-y-6">
|
||||
<form id="urlForm" method="POST" action="/" class="space-y-6">
|
||||
<div class="fields">
|
||||
<div class="input">
|
||||
<span class="icon icon--link"></span>
|
||||
|
|
@ -115,7 +115,7 @@
|
|||
<?php echo str_replace('{site_name}', SITE_NAME, \Inc\Language::get('bookmarklet_description')); ?>
|
||||
</p>
|
||||
<div>
|
||||
<a href="javascript:(function(){window.location.href='<?php echo SITE_URL; ?>/p/'+encodeURIComponent(window.location.href);})()"
|
||||
<a href="javascript:(function(){let currentUrl=window.location.href.replace(/^https?:\/\//, '');window.location.href='<?php echo SITE_URL; ?>/p/'+encodeURIComponent(currentUrl);})()"
|
||||
onclick="return false;">
|
||||
<?php echo str_replace('{site_name}', SITE_NAME, \Inc\Language::get('open_in')); ?>
|
||||
</a>
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue