Compatibility with October 3.x and introduction of new features

.gitignore

@@ -0,0 +1,2 @@
+tests/
+phpunit.xml

Plugin.php

@@ -1,10 +1,14 @@
-<?php namespace Bombozama\LinkCheck;
+<?php
+
+namespace Bombozama\LinkCheck;
 
-use Bombozama\LinkCheck\Models\BrokenLink;
-use System\Classes\PluginBase;
-use Bombozama\LinkCheck\Models\Settings;
 use Backend;
+use Flash;
 use Lang;
+use System\Classes\PluginBase;
+use Bombozama\LinkCheck\Models\Context;
+use Bombozama\LinkCheck\Models\Settings;
+use Bombozama\LinkCheck\ReportWidgets\BrokenLinks;
 
 /**
  * LinkCheck Plugin Information File
@@ -23,10 +27,24 @@ public function pluginDetails()
             'description' => 'bombozama.linkcheck::lang.details.description',
             'author'      => 'Gonzalo Henríquez',
             'icon'        => 'icon-chain-broken',
-            'homepage'    => 'https://github.com/bombozama/linkcheck'
+            'homepage'    => 'https://github.com/bombozama/linkcheck',
         ];
     }
 
+    public function register()
+    {
+        Settings::extend(function ($model) {
+            $model->bindEvent('model.beforeSave', function () use ($model) {
+                if (!post('Settings[plugins]')) {
+                    // If no plugin was selected, remove the selected model fields
+                    // by setting modelators to an empty string.
+                    $model->modelators = '';
+                    Flash::info(Lang::get('bombozama.linkcheck::lang.strings.plugins_empty'));
+                }
+            });
+        });
+    }
+
     public function registerPermissions()
     {
         return [
@@ -38,6 +56,10 @@ public function registerPermissions()
                 'tab'   => 'bombozama.linkcheck::lang.plugin.tab',
                 'label' => 'bombozama.linkcheck::lang.plugin.view',
             ],
+            'bombozama.linkcheck.useragent' => [
+                'tab'   => 'bombozama.linkcheck::lang.plugin.tab',
+                'label' => 'bombozama.linkcheck::lang.plugin.useragent',
+            ],
         ];
     }
 
@@ -51,16 +73,25 @@ public function registerSettings()
                 'icon'        => 'icon-chain-broken',
                 'class'       => 'Bombozama\LinkCheck\Models\Settings',
                 'order'       => 410,
-                'permissions' => ['bombozama.linkcheck.manage']
+                'permissions' => ['bombozama.linkcheck.manage'],
             ],
             'brokenlinks' => [
                 'label'       => 'bombozama.linkcheck::lang.menu.brokenlinks.label',
                 'description' => 'bombozama.linkcheck::lang.menu.brokenlinks.description',
                 'category'    => 'bombozama.linkcheck::lang.plugin.category',
                 'icon'        => 'icon-list',
-                'url'         => Backend::url('bombozama/linkcheck/brokenlinks'),
+                'url'         => Backend::url('bombozama/linkcheck/context'),
                 'order'       => 411,
-                'permissions' => ['bombozama.linkcheck.view']
+                'permissions' => ['bombozama.linkcheck.view'],
+            ],
+            'userAgent' => [
+                'label'       => 'bombozama.linkcheck::lang.menu.useragent.label',
+                'description' => 'bombozama.linkcheck::lang.menu.useragent.description',
+                'category'    => 'bombozama.linkcheck::lang.plugin.category',
+                'icon'        => 'icon-user-secret',
+                'url'         => Backend::url('bombozama/linkcheck/useragent'),
+                'order'       => 412,
+                'permissions' => ['bombozama.linkcheck.useragent'],
             ],
         ];
     }
@@ -74,15 +105,30 @@ public function registerListColumnTypes()
 
     public function httpStatus($value, $column, $record)
     {
-        return '<span title="' . Lang::get('bombozama.linkcheck::lang.codes.' . $value ) . '">' . $value . '</span>';
+        return '<span title="' . Lang::get('bombozama.linkcheck::lang.codes.' . $value) . '">' . $value . '</span>';
+    }
+
+    public function registerReportWidgets()
+    {
+        return [
+            BrokenLinks::class => [
+                'label' => 'bombozama.linkcheck::lang.details.name',
+                'context' => 'dashboard',
+                'permissions' => [
+                    'bombozama.linkcheck.view',
+                ],
+            ],
+        ];
     }
 
     // Please do https://octobercms.com/docs/setup/installation#crontab-setup
     public function registerSchedule($schedule)
     {
         $settings = Settings::instance();
-        $schedule->call(function(){
-            BrokenLink::processLinks();
-        })->cron($settings->time);
+        if ($settings->time) {
+            $schedule->call(function() {
+                Context::processLinks();
+            })->cron($settings->time);
+        }
     }
 }

Readme.md

@@ -1,11 +1,20 @@
 # LinkCheck Plugin for OctoberCMS
 Schedules a task to check for broken links in database fields and/or CMS pages.
 
+## Refactoring
+* The plugin has been revised and optimised for October v3.x.
+* DB: Splitting of URL/status and contexts (CMS pages, model fields) in which the URLs are located
+* URL-based approach: URLs are stored in a set so that URLs are only checked once
+* cURL only calls headers
+
 ## Features
 * Checks for broken links within database fields and CMS text files
 * User controls which fields to check
 * User controls which response codes should be logged
 * User controls task scheduling
+* Management and use of user agents (new in v2.x)
+* Dashboard report widget with overview (new in v2.x)
+* Selection of plugins and directories for the check in the settings (new in v2.x)
 
 ## Usage
 A link status report can be found in the backend Settings tab.

classes/Helper.php

@@ -1,12 +1,40 @@
-<?php namespace Bombozama\LinkCheck\Classes;
+<?php
+
+namespace Bombozama\LinkCheck\Classes;
 
 class Helper
 {
-    public static function scanForUrls($string)
+    public function scanForUrls(string $string): array
     {
+        $urls = array();
         $expression = '#\bhttps?://[^\s()<>]+(?:\([\w\d]+\)|([^[:punct:]\s]|/))#';
+
         preg_match_all($expression, $string, $matches);
-        return ($matches[0]);
+
+        foreach($matches[0] as $match) {
+            try {
+                $url = parse_url($match);
+
+                // Use URL without fragment
+                $parsed_url = $url['scheme'] . '://' . $url['host'];
+
+                if (isset($url['path'])) {
+                    $parsed_url .= $url['path'];
+                }
+
+                if (isset($url['query'])) {
+                    $parsed_url .= '?' . $url['query'];
+                }
+
+                $urls[] = $parsed_url;
+
+            } catch (\Exception $ex) {
+                // If URL is invalid / malformed, log an error
+                \Log::info('LinkCheck Error: Could not parse URL: ' . $match . ' due to the following error: ' . $ex);
+            }
+        }
+
+        return $urls;
     }
 
     public static function getFullClassNameFromFile($pathToFile)
@@ -15,47 +43,78 @@ public static function getFullClassNameFromFile($pathToFile)
         $class = $namespace = $buffer = '';
         $i = 0;
         while (!$class) {
-            if (feof($fp))
+            if (feof($fp)) {
                 break;
+            }
 
             $buffer.= fread($fp, 512);
             $tokens = token_get_all($buffer);
 
-            if (strpos($buffer, '{') === false)
+            if (strpos($buffer, '{') === false) {
                 continue;
+            }
 
             for (; $i < count($tokens); $i++) {
-                if($tokens[$i][0] === T_NAMESPACE)
-                    for($j = $i+1; $j < count($tokens); $j++)
-                        if($tokens[$j][0] === T_STRING)
+                if ($tokens[$i][0] === T_NAMESPACE) {
+                    for ($j = $i+1; $j < count($tokens); $j++) {
+                        if ($tokens[$j][0] === 265) { // PHP token id 265 == string
                             $namespace .= '\\' . $tokens[$j][1];
-                        else if($tokens[$j] === '{' || $tokens[$j] === ';')
+                        } elseif ($tokens[$j] === '{' || $tokens[$j] === ';') {
                             break;
-
-                if($tokens[$i][0] === T_CLASS)
-                    for($j = $i+1; $j < count($tokens); $j++)
-                        if($tokens[$j] === '{')
-                            $class = $tokens[$i+2][1];
+                        }
+                    }
+                }
+                if ($tokens[$i][0] === T_CLASS) {
+                    for ($j = $i+1; $j < count($tokens); $j++) {
+                        if ($tokens[$j] === '{') {
+                            if (is_array($tokens[$i+2]) && $tokens[$i+2][0] !== 392) { // PHP token id 392 == whitespace 
+                                $class = $tokens[$i+2][1];
+                            }
+                        }
+                    }
+                }
             }
         }
         return $namespace . '\\' . $class;
     }
 
-    public static function getResponseCode($url)
+    public static function getResponseCode(string $url, mixed $userAgent): int
     {
         $ch = curl_init();
         curl_setopt($ch, CURLOPT_URL, $url);
         curl_setopt($ch, CURLOPT_HEADER, 1);
-        curl_setopt($ch , CURLOPT_RETURNTRANSFER, 1);
+        curl_setopt($ch , CURLOPT_NOBODY, 1);
+        curl_setopt($ch, CURLOPT_TIMEOUT_MS, 30000); # Setting timeout to 30 seconds
+        if ($userAgent) {
+            curl_setopt($ch, CURLOPT_USERAGENT, $userAgent);
+        }
         $data = curl_exec($ch);
         $headers = curl_getinfo($ch);
         curl_close($ch);
-
         # In case of timeouts, let's throw out a 408 error "Request timeout"
-        if($headers['http_code'] == 0)
-            $headers['http_code'] = '408';
-
+        if ($headers['http_code'] == 0) {
+            $headers['http_code'] = 408;
+        }
         return $headers['http_code'];
     }
 
+    /**
+     * Flatten an arbitrarily nested array
+     * 
+     * see https://www.lambda-out-loud.com/posts/flatten-arrays-php/
+     * 
+     * @param array
+     * @return array
+     */
+    public function flattenArray(array $array): array
+    {
+        $recursiveArrayIterator = new \RecursiveArrayIterator(
+            $array,
+            \RecursiveArrayIterator::CHILD_ARRAYS_ONLY
+        );
+        $iterator = new \RecursiveIteratorIterator($recursiveArrayIterator);
+
+        return iterator_to_array($iterator, false);
+    }
+
 }

controllers/Context.php

@@ -0,0 +1,37 @@
+<?php
+
+namespace Bombozama\LinkCheck\Controllers;
+
+use BackendMenu;
+use Backend\Classes\Controller;
+use System\Classes\SettingsManager;
+use Flash;
+use Lang;
+
+/**
+ * Broken Links Back-end Controller
+ */
+class Context extends Controller
+{
+    public $requiredPermissions = ['bombozama.linkcheck.view'];
+    public $implement = [
+        'Backend.Behaviors.ListController',
+    ];
+    public $listConfig = 'config_list.yaml';
+
+    public function __construct()
+    {
+        parent::__construct();
+        BackendMenu::setContext('October.System', 'system', 'settings');
+        SettingsManager::setContext('Bombozama.LinkCheck', 'brokenlinks');
+    }
+
+    public function onRefreshLinkList()
+    {
+        Flash::warning(Lang::get('bombozama.linkcheck::lang.strings.working'));
+        $links = \Bombozama\Linkcheck\Models\Context::processLinks();
+        Flash::success(Lang::get('bombozama.linkcheck::lang.strings.total_links', ['number' => $links['reported'], 'total' => $links['checked']]));
+
+        return $this->listRefresh();
+    }
+}

controllers/UserAgent.php

@@ -0,0 +1,46 @@
+<?php
+
+namespace Bombozama\Linkcheck\Controllers;
+
+use BackendMenu;
+use Backend\Classes\Controller;
+use System\Classes\SettingsManager;
+
+/**
+ * User Agent Backend Controller
+ *
+ * @link https://docs.octobercms.com/3.x/extend/system/controllers.html
+ */
+class UserAgent extends Controller
+{
+    public $implement = [
+        \Backend\Behaviors\FormController::class,
+        \Backend\Behaviors\ListController::class,
+    ];
+
+    /**
+     * @var string formConfig file
+     */
+    public $formConfig = 'config_form.yaml';
+
+    /**
+     * @var string listConfig file
+     */
+    public $listConfig = 'config_list.yaml';
+
+    /**
+     * @var array required permissions
+     */
+    public $requiredPermissions = ['bombozama.linkcheck.useragent'];
+
+    /**
+     * __construct the controller
+     */
+    public function __construct()
+    {
+        parent::__construct();
+
+        BackendMenu::setContext('October.System', 'system', 'settings');
+        SettingsManager::setContext('Bombozama.LinkCheck', 'userAgent');
+    }
+}

controllers/context/_list_toolbar.htm

@@ -0,0 +1,7 @@
+<?php if (BackendAuth::userHasAccess('bombozama.linkcheck.manage')): ?>
+    <div data-control="toolbar">
+        <button type="button" data-request="onRefreshLinkList" class="btn btn-primary oc-icon-plus">
+            <?= e(trans('bombozama.linkcheck::lang.strings.refresh_list')) ?>
+        </button>
+    </div>
+<?php endif ?>