Update app.py

app.py

@@ -6,76 +6,71 @@ import zipfile
 import tempfile
 from urllib.parse import urljoin
 from bs4 import BeautifulSoup
-from mutagen.mp3 import MP3
-from mutagen.id3 import ID3
-
-def get_clean_title(filepath):
-    try:
-        audio = MP3(filepath, ID3=ID3)
-        for tag in ['TIT2', 'TIT3', 'TALB']:
-            if tag in audio:
-                title = audio[tag].text[0]
-                title = re.sub(r'[\\/*?:"<>|]', '', title).strip()
-                return title[:100]
-        return os.path.basename(filepath).split('.')[0]
-    except Exception:
-        return os.path.basename(filepath).split('.')[0]
 
 def process_url(url):
     try:
-        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'}
-        response = requests.get(url, headers=headers, timeout=15)
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
+            'Referer': 'https://www.radiofrance.fr/'
+        }
+        response = requests.get(url, headers=headers)
         response.raise_for_status()
     except Exception as e:
-        return None, f"Erreur : {str(e)}"
+        return None, f"Erreur de connexion : {str(e)}"
 
     soup = BeautifulSoup(response.text, 'html.parser')
     mp3_links = []
-    
-    # Méthode 1 : Balises JSON-LD
-    for script in soup.find_all('script', type='application/ld+json'):
+
+    # Nouvelle méthode de détection
+    scripts = soup.find_all('script', type='application/ld+json')
+    for script in scripts:
         if script.string:
             matches = re.findall(r'"contentUrl"\s*:\s*"([^"]+?\.mp3)', script.string)
             for match in matches:
-                clean_url = urljoin(url, match.split('?')[0])
-                if clean_url not in mp3_links:
-                    mp3_links.append(clean_url)
+                full_url = urljoin(url, match.split('?')[0])
+                if full_url not in mp3_links:
+                    mp3_links.append(full_url)
 
-    # Méthode 2 : Fallback HTML
+    # Fallback pour les URLs dans les attributs data
     if not mp3_links:
-        for a in soup.find_all('a', href=re.compile(r'\.mp3')):
-            mp3_url = urljoin(url, a['href'].split('?')[0])
-            if mp3_url not in mp3_links:
-                mp3_links.append(mp3_url)
+        for tag in soup.find_all(attrs={"data-url": re.compile(r"\.mp3")}):
+            mp3_url = urljoin(url, tag['data-url'].split('?')[0])
+            mp3_links.append(mp3_url)
 
+    # Dernier recours : recherche globale
     if not mp3_links:
-        return None, "Aucun épisode trouvé"
+        matches = re.findall(r'(https?://[^\s"\']+?\.mp3)', response.text)
+        for match in matches:
+            clean_url = urljoin(url, match.split('?')[0])
+            if clean_url not in mp3_links and 'podcast' in clean_url:
+                mp3_links.append(clean_url)
+
+    # Filtrage final
+    mp3_links = list(dict.fromkeys(mp3_links))  # Supprime les doublons
 
+    if not mp3_links:
+        return None, "Aucun épisode trouvé - Structure de page inconnue"
+
+    # Téléchargement et création ZIP
     temp_dir = tempfile.mkdtemp()
     filenames = []
     
     for idx, mp3_url in enumerate(mp3_links, 1):
         try:
-            temp_path = os.path.join(temp_dir, f"temp_{idx}.mp3")
+            filename = f"{idx:02d}_{os.path.basename(mp3_url)}"
+            filepath = os.path.join(temp_dir, filename)
             
-            # Téléchargement
-            with requests.get(mp3_url, headers=headers, stream=True, timeout=20) as r:
+            with requests.get(mp3_url, headers=headers, stream=True) as r:
                 r.raise_for_status()
-                with open(temp_path, 'wb') as f:
+                with open(filepath, 'wb') as f:
                     for chunk in r.iter_content(chunk_size=8192):
                         f.write(chunk)
-            
-            # Renommage
-            title = get_clean_title(temp_path)
-            final_name = f"{idx:02d} - {title}.mp3"
-            final_path = os.path.join(temp_dir, final_name)
-            os.rename(temp_path, final_path)
-            filenames.append(final_path)
+            filenames.append(filepath)
         except Exception:
             continue
 
     if not filenames:
-        return None, "Échec des téléchargements"
+        return None, "Échec du téléchargement"
 
     zip_path = os.path.join(temp_dir, 'podcast.zip')
     with zipfile.ZipFile(zip_path, 'w') as zipf:
@@ -91,19 +86,28 @@ def download_podcast(url):
     return zip_path
 
 with gr.Blocks() as app:
-    gr.Markdown("## 🎙️ Téléchargeur Radio France")
+    gr.Markdown("## 🎧 Téléchargeur Radio France")
     with gr.Row():
-        url = gr.Textbox(label="URL de la série", placeholder="Collez l'URL ici...")
-        btn = gr.Button("Télécharger", variant="primary")
-    output = gr.File(label="Épisodes")
+        url_input = gr.Textbox(
+            label="URL du podcast",
+            placeholder="Ex: https://www.radiofrance.fr/...",
+            max_lines=1
+        )
+    btn = gr.Button("Télécharger les épisodes", variant="primary")
+    output = gr.File(label="Fichier ZIP")
     
     examples = gr.Examples(
         examples=[[
             "https://www.radiofrance.fr/franceculture/podcasts/serie-le-secret-de-la-licorne-les-aventures-de-tintin"
         ]],
-        inputs=[url]
+        inputs=[url_input]
     )
 
-    btn.click(download_podcast, inputs=url, outputs=output)
+    btn.click(
+        fn=download_podcast,
+        inputs=url_input,
+        outputs=output,
+        api_name="download"
+    )
 
 app.launch()