up wording comparaison

src/Controller/WikiController.php

@@ -719,11 +719,68 @@ class WikiController extends AbstractController
         // Construct the URLs for the English page and the French page creation form
         $englishUrl = "https://wiki.openstreetmap.org/wiki/Key:{$key}";
         $frenchEditUrl = "https://wiki.openstreetmap.org/w/index.php?title=FR:{$key}&action=edit";
+        
+        // Fetch the HTML content of the English page using wiki_compare.py
+        $scriptPath = $this->getParameter('kernel.project_dir') . '/wiki_compare/wiki_compare.py';
+        $englishHtml = null;
+        $frenchHtml = null;
+        
+        if (file_exists($scriptPath)) {
+            // Create a temporary Python script to fetch the page content
+            $tempScriptPath = $this->getParameter('kernel.project_dir') . '/wiki_compare/temp_fetch_page.py';
+            $pythonCode = <<<EOT
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import sys
+import json
+from wiki_compare import fetch_wiki_page
+
+# Get the key from command line arguments
+key = sys.argv[1]
+language = sys.argv[2]
+
+# Fetch the page
+page = fetch_wiki_page(key, language)
+
+# Output the HTML content
+if page and 'html_content' in page:
+    print(page['html_content'])
+else:
+    print("")
+EOT;
+            
+            file_put_contents($tempScriptPath, $pythonCode);
+            chmod($tempScriptPath, 0755);
+            
+            // Fetch English page
+            $command = "cd " . $this->getParameter('kernel.project_dir') . "/wiki_compare && python3 {$tempScriptPath} {$key} en";
+            $englishHtml = shell_exec($command);
+            
+            // Extract only the content part from the HTML (remove headers, footers, etc.)
+            if ($englishHtml) {
+                $englishHtml = $this->extractMainContent($englishHtml);
+            }
+            
+            // Fetch French page (might not exist, but we'll try)
+            $command = "cd " . $this->getParameter('kernel.project_dir') . "/wiki_compare && python3 {$tempScriptPath} {$key} fr";
+            $frenchHtml = shell_exec($command);
+            
+            // Extract only the content part from the HTML (remove headers, footers, etc.)
+            if ($frenchHtml) {
+                $frenchHtml = $this->extractMainContent($frenchHtml);
+            }
+            
+            // Clean up the temporary script
+            unlink($tempScriptPath);
+        }
 
         return $this->render('admin/wiki_create_french.html.twig', [
             'key' => $key,
             'english_url' => $englishUrl,
-            'french_edit_url' => $frenchEditUrl
+            'french_edit_url' => $frenchEditUrl,
+            'english_html' => $englishHtml,
+            'french_html' => $frenchHtml
         ]);
     }
 
@@ -1436,4 +1493,57 @@ class WikiController extends AbstractController
             'fr_links' => $frLinks
         ]);
     }
+    
+    /**
+     * Extracts the main content from the HTML, removing headers, footers, and other unnecessary elements
+     * 
+     * @param string $html The full HTML content
+     * @return string The extracted main content
+     */
+    private function extractMainContent(string $html): string
+    {
+        // Use a simple approach to extract the content
+        // This could be improved with a more sophisticated HTML parser if needed
+        
+        // Create a DOMDocument to parse the HTML
+        $dom = new \DOMDocument();
+        
+        // Suppress warnings about malformed HTML
+        libxml_use_internal_errors(true);
+        $dom->loadHTML($html);
+        libxml_clear_errors();
+        
+        // Try to find the main content element
+        $contentElement = null;
+        
+        // First, try to find the element with id "mw-content-text"
+        $contentElement = $dom->getElementById('mw-content-text');
+        
+        // If not found, try to find the element with class "mw-content-ltr"
+        if (!$contentElement) {
+            $xpath = new \DOMXPath($dom);
+            $elements = $xpath->query("//*[contains(@class, 'mw-content-ltr')]");
+            if ($elements->length > 0) {
+                $contentElement = $elements->item(0);
+            }
+        }
+        
+        // If still not found, return the original HTML
+        if (!$contentElement) {
+            return $html;
+        }
+        
+        // Get the HTML of the content element
+        $contentHtml = $dom->saveHTML($contentElement);
+        
+        // Clean up the content HTML
+        // Remove script and style elements
+        $contentHtml = preg_replace('/<script\b[^>]*>(.*?)<\/script>/is', '', $contentHtml);
+        $contentHtml = preg_replace('/<style\b[^>]*>(.*?)<\/style>/is', '', $contentHtml);
+        
+        // Remove edit section links
+        $contentHtml = preg_replace('/<span class="mw-editsection">(.*?)<\/span>/is', '', $contentHtml);
+        
+        return $contentHtml;
+    }
 }