try to guess chat language

[alttp.git] / app / Models / ChatLog.php

diff --git a/app/Models/ChatLog.php b/app/Models/ChatLog.php
index 6f72c352b42ff356123ad3e1bdb29b8d8407e487..8a936a82804adee37cc93213fa42cc5e0fde6b47 100644 (file)
--- a/app/Models/ChatLog.php
+++ b/app/Models/ChatLog.php
@@ -4,6 +4,7 @@ namespace App\Models;
 
 use Illuminate\Database\Eloquent\Factories\HasFactory;
 use Illuminate\Database\Eloquent\Model;
+use LanguageDetector\LanguageDetector;
 
 class ChatLog extends Model {
 
@@ -39,6 +40,7 @@ class ChatLog extends Model {
                                $this->type = 'dm';
                        }
                        $this->text_content = $this->params[1];
+                       $this->detectLanguage();
                        if ($this->scanForSpam()) {
                                $this->banned = true;
                        }
@@ -75,6 +77,23 @@ class ChatLog extends Model {
                $this->channel()->associate($channel);
        }
 
+       protected function detectLanguage() {
+               $languages = ['de', 'en', 'es', 'fr'];
+               if (!is_null($this->channel)) {
+                       $languages = array_values($this->channel->languages);
+                       if (!in_array('en', $languages)) {
+                               $languages[] = 'en';
+                       }
+               }
+               $detector = LanguageDetector::detect($this->text_content, $languages);
+               $scores = $detector->getScores();
+               $lang = strval($detector->getLanguage());
+               //var_dump($scores, $lang, $this->text_content);
+               if (is_array($scores) && isset($scores[$lang]) && $scores[$lang] > 0.35) {
+                       $this->detected_language = $lang;
+               }
+       }
+
        protected function scanForSpam() {
                if (substr($this->text_content, 0, 1) == '!') {
                        return true;