Upload app.py with huggingface_hub

app.py

@@ -1,148 +1,201 @@
-from flask import Flask, render_template, request, jsonify, url_for
-from gtts import gTTS
-import os
-import logging
-from translation_data import (
-    translation_dict_A,
-    translation_dict_B,
-    translation_dict_C,
-    translation_dict_D,
-    translation_dict_F
-    # Include translation_dict_E, translation_dict_G if they exist
-)
-
-# Initialize Flask app
-app = Flask(__name__)
-
-# Configure logging
-logging.basicConfig(level=logging.DEBUG)
-
-# Directory to save audio files
-AUDIO_DIR = os.path.join('static', 'audio')
-
-# Ensure the audio directory exists
-os.makedirs(AUDIO_DIR, exist_ok=True)
-
-# Flashcards data organized by categories
-flashcards = {
-    'A': {
-        'chinese_sentences': list(translation_dict_A.keys()),
-        'japanese_translations': list(translation_dict_A.values())
-    },
-    'B': {
-        'chinese_sentences': list(translation_dict_B.keys()),
-        'japanese_translations': list(translation_dict_B.values())
-    },
-    'C': {
-        'chinese_sentences': list(translation_dict_C.keys()),
-        'japanese_translations': list(translation_dict_C.values())
-    },
-    'D': {
-        'chinese_sentences': list(translation_dict_D.keys()),
-        'japanese_translations': list(translation_dict_D.values())
-    },
-    'F': {
-        'chinese_sentences': list(translation_dict_F.keys()),
-        'japanese_translations': list(translation_dict_F.values())
-    },
-    # Add 'E' and 'G' similarly if needed
-}
-
-# Helper function to generate audio
-def generate_audio(text, set_name, index):
-    """Generate an audio file from Japanese text using gTTS."""
-    filename = f"{set_name}_{index}.mp3"
-    filepath = os.path.join(AUDIO_DIR, filename)
-
-    if not os.path.exists(filepath):
-        logging.info(f"Generating audio file: {filepath}")
-        try:
-            tts = gTTS(text=text, lang='ja')  # Japanese TTS
-            tts.save(filepath)
-        except Exception as e:
-            logging.error(f"Error generating audio: {e}")
-            return None
-    else:
-        logging.info(f"Using existing audio file: {filepath}")
-
-    return filepath
-
-# Route for the portal page
-@app.route('/')
-def portal():
-    """Render the portal page with links to different categories."""
-    return render_template('portal.html')
-
-# Route to render the flashcards page
-@app.route('/flashcards')
-def flashcards_page():
-    """Render the flashcards page for a specific set and index."""
-    set_name = request.args.get('set', 'A')
-    try:
-        index = int(request.args.get('index', 0))
-    except ValueError:
-        return "Invalid index parameter", 400
-
-    if set_name not in flashcards:
-        return "Set not found", 404
-
-    total = len(flashcards[set_name]['chinese_sentences'])
-    if not (0 <= index < total):
-        return "Index out of range", 404
-
-    chinese = flashcards[set_name]['chinese_sentences'][index]
-    japanese = flashcards[set_name]['japanese_translations'][index]
-    audio_url = url_for('static', filename=f"audio/{set_name}_{index}.mp3")
-
-    return render_template(
-        'flashcards.html',
-        set_name=set_name,
-        index=index,
-        total=total,
-        japanese=japanese,      # Changed from 'english' to 'japanese'
-        chinese=chinese,        # Added 'chinese'
-        audio_url=audio_url
-    )
-
-# API endpoint to fetch flashcard data
-@app.route('/api/flashcards')
-def api_flashcards():
-    """API endpoint to fetch flashcard data."""
-    set_name = request.args.get('set', 'A')
-    try:
-        index = int(request.args.get('index', 0))
-    except ValueError:
-        return jsonify({'error': 'Invalid index parameter'}), 400
-
-    if set_name in flashcards:
-        chinese_sentences = flashcards[set_name]['chinese_sentences']
-        japanese_translations = flashcards[set_name]['japanese_translations']
-        total = len(chinese_sentences)
-
-        if 0 <= index < total:
-            chinese = chinese_sentences[index]
-            japanese = japanese_translations[index]
-
-            # Generate audio from Japanese text
-            audio_path = generate_audio(japanese, set_name, index)
-            if audio_path:
-                audio_url = url_for('static', filename=f"audio/{set_name}_{index}.mp3")
-            else:
-                audio_url = None  # Handle audio generation failure
-
-            return jsonify({
-                'set_name': set_name,
-                'index': index,
-                'total': total,
-                'japanese': japanese,    # Changed from 'english' to 'japanese'
-                'chinese': chinese,      # Added 'chinese'
-                'audio_url': audio_url
-            })
-        else:
-            return jsonify({'error': 'Index out of range'}), 404
-    else:
-        return jsonify({'error': 'Set not found'}), 404
-
-if __name__ == '__main__':
-    # Run the Flask app on all available IPs on port 7860
-    app.run(debug=True, host="0.0.0.0", port=7860)
+# app.pyの一部
+
+from flask import Flask, render_template, request, jsonify, url_for
+from gtts import gTTS
+import os
+import logging
+import hashlib
+from translation_data import (
+    translation_dict_A,
+    translation_dict_B,
+    translation_dict_C,
+    translation_dict_D,
+    translation_dict_F
+    # Include translation_dict_E, translation_dict_G if they exist
+)
+
+# Initialize Flask app
+app = Flask(__name__)
+
+# Configure logging
+logging.basicConfig(level=logging.DEBUG)
+
+# Directory to save audio files
+AUDIO_DIR = os.path.join('static', 'audio')
+
+# Ensure the audio directory exists
+os.makedirs(AUDIO_DIR, exist_ok=True)
+
+# Flashcards data organized by categories
+flashcards = {
+    'A': {
+        'chinese_sentences': list(translation_dict_A.keys()),
+        'japanese_translations': list(translation_dict_A.values())
+    },
+    'B': {
+        'chinese_sentences': list(translation_dict_B.keys()),
+        'japanese_translations': list(translation_dict_B.values())
+    },
+    'C': {
+        'chinese_sentences': list(translation_dict_C.keys()),
+        'japanese_translations': list(translation_dict_C.values())
+    },
+    'D': {
+        'chinese_sentences': list(translation_dict_D.keys()),
+        'japanese_translations': list(translation_dict_D.values())
+    },
+    'F': {
+        'chinese_sentences': list(translation_dict_F.keys()),
+        'japanese_translations': list(translation_dict_F.values())
+    },
+    # Add 'E' and 'G' similarly if needed
+}
+
+# Helper function to generate a truncated hash
+def generate_truncated_hash(text, length=8):
+    """
+    Generate a truncated SHA-256 hash for the given text.
+
+    Args:
+        text (str): The input text to hash.
+        length (int): The length of the truncated hash.
+
+    Returns:
+        str: Truncated hash string.
+    """
+    hash_object = hashlib.sha256(text.encode('utf-8'))
+    hex_dig = hash_object.hexdigest()
+    return hex_dig[:length]
+
+# Helper function to generate gTTS audio
+def generate_audio_def(text, set_name, index):
+    """
+    Generate a gTTS audio file for the given text.
+
+    Args:
+        text (str): The text to convert to speech.
+        set_name (str): The set identifier.
+        index (int): The index of the flashcard.
+
+    Returns:
+        str: URL to the saved audio file.
+    """
+    sentence_hash = generate_truncated_hash(text)
+    folder_path = os.path.join(AUDIO_DIR, sentence_hash)
+    os.makedirs(folder_path, exist_ok=True)
+
+    filename = f"{set_name}_{index}_def.mp3"
+    filepath = os.path.join(folder_path, filename)
+
+    if not os.path.exists(filepath):
+        logging.info(f"Generating gTTS audio file: {filepath}")
+        try:
+            tts = gTTS(text=text, lang='ja')  # Japanese TTS
+            tts.save(filepath)
+        except Exception as e:
+            logging.error(f"Error generating gTTS audio: {e}")
+            return None
+    else:
+        logging.info(f"Using existing gTTS audio file: {filepath}")
+
+    return url_for('static', filename=f"audio/{sentence_hash}/{filename}")
+
+# Route for the portal page
+@app.route('/')
+def portal():
+    """Render the portal page with links to different categories."""
+    return render_template('portal.html')
+
+# Route to render the flashcards page
+@app.route('/flashcards')
+def flashcards_page():
+    """Render the flashcards page for a specific set and index."""
+    set_name = request.args.get('set', 'A')
+    try:
+        index = int(request.args.get('index', 0))
+    except ValueError:
+        return "Invalid index parameter", 400
+
+    if set_name not in flashcards:
+        return "Set not found", 404
+
+    total = len(flashcards[set_name]['chinese_sentences'])
+    if not (0 <= index < total):
+        return "Index out of range", 404
+
+    chinese = flashcards[set_name]['chinese_sentences'][index]
+    japanese = flashcards[set_name]['japanese_translations'][index]
+    audio_def_url = generate_audio_def(japanese, set_name, index)
+
+    # Predefined variant filenames (assuming they are pre-generated)
+    sentence_hash = generate_truncated_hash(japanese)
+    variant_urls = {}
+    for variant_num in range(1, 4):  # Variants 1 to 4
+        variant_filename = f"variant_{variant_num}.mp3"
+        variant_filepath = os.path.join(AUDIO_DIR, sentence_hash, variant_filename)
+        if os.path.exists(variant_filepath):
+            variant_urls[f"variant_{variant_num}"] = url_for('static', filename=f"audio/{sentence_hash}/{variant_filename}")
+        else:
+            variant_urls[f"variant_{variant_num}"] = None  # Handle missing files
+
+    return render_template(
+        'flashcards.html',
+        set_name=set_name,
+        index=index,
+        total=total,
+        japanese=japanese,      # Japanese text
+        chinese=chinese,        # Chinese text
+        audio_def_url=audio_def_url,  # gTTS audio URL
+        variant_urls=variant_urls      # Other variant audio URLs
+    )
+
+# API endpoint to fetch flashcard data
+@app.route('/api/flashcards')
+def api_flashcards():
+    """API endpoint to fetch flashcard data with multiple audio variants."""
+    set_name = request.args.get('set', 'A')
+    try:
+        index = int(request.args.get('index', 0))
+    except ValueError:
+        return jsonify({'error': 'Invalid index parameter'}), 400
+
+    if set_name in flashcards:
+        chinese_sentences = flashcards[set_name]['chinese_sentences']
+        japanese_translations = flashcards[set_name]['japanese_translations']
+        total = len(chinese_sentences)
+
+        if 0 <= index < total:
+            chinese = chinese_sentences[index]
+            japanese = japanese_translations[index]
+
+            # gTTSによるお手本音声
+            audio_def_url = generate_audio_def(japanese, set_name, index)
+
+            # 他のバリアント音声のURLを取得
+            sentence_hash = generate_truncated_hash(japanese)
+            variant_urls = {}
+            for variant_num in range(1, 5):  # Variants 1 to 4
+                variant_filename = f"variant_{variant_num}.mp3"
+                variant_filepath = os.path.join(AUDIO_DIR, sentence_hash, variant_filename)
+                if os.path.exists(variant_filepath):
+                    variant_urls[f"variant_{variant_num}"] = url_for('static', filename=f"audio/{sentence_hash}/{variant_filename}")
+                else:
+                    variant_urls[f"variant_{variant_num}"] = None  # Handle missing files
+
+            return jsonify({
+                'set_name': set_name,
+                'index': index,
+                'total': total,
+                'japanese': japanese,
+                'chinese': chinese,
+                'audio_def_url': audio_def_url,      # gTTS音声
+                'variant_urls': variant_urls         # 他のバリアント音声
+            })
+        else:
+            return jsonify({'error': 'Index out of range'}), 404
+    else:
+        return jsonify({'error': 'Set not found'}), 404
+
+if __name__ == '__main__':
+    # Run the Flask app on all available IPs on port 7860
+    app.run(debug=True, host="0.0.0.0", port=7860)