Update app.py

app.py

@@ -20,15 +20,17 @@ import re
 from streamlit_paste_button import paste_image_button
 import pytz
 import shutil
+import wave
+import audioop
 
-# Patch for nested async - sneaky fix! 🐍✨
+# Patch for nested async
 nest_asyncio.apply()
 
-# Static config - constants rule! 📏👑
+# Static config
 icons = '🤖🧠🔬📝'
 START_ROOM = "Sector 🌌"
 
-# Page setup - dressing up the window! 🖼️🎀
+# Page setup
 st.set_page_config(
     page_title="🤖🧠MMO Chat Brain📝🔬",
     page_icon=icons,
@@ -36,7 +38,7 @@ st.set_page_config(
     initial_sidebar_state="auto"
 )
 
-# Funky usernames - who’s who in the zoo with unique voices! 🎭🐾🎙️
+# Funky usernames with voices
 FUN_USERNAMES = {
     "CosmicJester 🌌": "en-US-AriaNeural",
     "PixelPanda 🐼": "en-US-JennyNeural",
@@ -60,7 +62,7 @@ FUN_USERNAMES = {
     "ChronoChimp 🐒": "en-GB-LibbyNeural"
 }
 
-# Folders galore - organizing chaos! 📂🌀
+# Folders
 CHAT_DIR = "chat_logs"
 VOTE_DIR = "vote_logs"
 STATE_FILE = "user_state.txt"
@@ -78,10 +80,10 @@ QUOTE_VOTES_FILE = os.path.join(VOTE_DIR, "quote_votes.md")
 MEDIA_VOTES_FILE = os.path.join(VOTE_DIR, "media_votes.md")
 HISTORY_FILE = os.path.join(HISTORY_DIR, "chat_history.md")
 
-# Fancy digits - numbers got style! 🔢💃
+# Unicode digits
 UNICODE_DIGITS = {i: f"{i}\uFE0F⃣" for i in range(10)}
 
-# Massive font collection - typography bonanza! 🖋️🎨
+# Font collection
 UNICODE_FONTS = [
     ("Normal", lambda x: x),
     ("Bold", lambda x: "".join(chr(ord(c) + 0x1D400 - 0x41) if 'A' <= c <= 'Z' else chr(ord(c) + 0x1D41A - 0x61) if 'a' <= c <= 'z' else c for c in x)),
@@ -104,7 +106,7 @@ UNICODE_FONTS = [
     ("Regional Indicator", lambda x: "".join(chr(ord(c) - 0x41 + 0x1F1E6) if 'A' <= c <= 'Z' else c for c in x)),
 ]
 
-# Global state - keeping tabs! 🌍📋
+# Global state
 if 'server_running' not in st.session_state:
     st.session_state.server_running = False
 if 'server_task' not in st.session_state:
@@ -140,13 +142,12 @@ if 'last_transcript' not in st.session_state:
 if 'image_hashes' not in st.session_state:
     st.session_state.image_hashes = set()
 
-# Timestamp wizardry - clock ticks with flair! ⏰🎩
+# Utility functions
 def format_timestamp_prefix(username):
     central = pytz.timezone('US/Central')
     now = datetime.now(central)
     return f"{now.strftime('%I-%M-%p-ct-%m-%d-%Y')}-by-{username}"
 
-# Compute image hash from binary data
 def compute_image_hash(image_data):
     if isinstance(image_data, Image.Image):
         img_byte_arr = io.BytesIO()
@@ -156,7 +157,6 @@ def compute_image_hash(image_data):
         img_bytes = image_data
     return hashlib.md5(img_bytes).hexdigest()[:8]
 
-# Node naming - christening the beast! 🌐🍼
 def get_node_name():
     parser = argparse.ArgumentParser(description='Start a chat node with a specific name')
     parser.add_argument('--node-name', type=str, default=None)
@@ -166,7 +166,6 @@ def get_node_name():
     log_action(username, "🌐🍼 - Node naming - christening the beast!")
     return args.node_name or f"node-{uuid.uuid4().hex[:8]}", args.port
 
-# Action logger - spying on deeds! 🕵️📜
 def log_action(username, action):
     if 'action_log' not in st.session_state:
         st.session_state.action_log = {}
@@ -180,13 +179,11 @@ def log_action(username, action):
             f.write(f"[{datetime.now(central).strftime('%Y-%m-%d %H:%M:%S')}] {username}: {action}\n")
         user_log[action] = current_time
 
-# Clean text - strip the fancy stuff! 🧹📝
 def clean_text_for_tts(text):
     cleaned = re.sub(r'[#*!\[\]]+', '', text)
     cleaned = ' '.join(cleaned.split())
     return cleaned[:200] if cleaned else "No text to speak"
 
-# Chat saver - words locked tight! 💬🔒
 async def save_chat_entry(username, message, is_markdown=False):
     await asyncio.to_thread(log_action, username, "💬🔒 - Chat saver - words locked tight!")
     central = pytz.timezone('US/Central')
@@ -206,7 +203,6 @@ async def save_chat_entry(username, message, is_markdown=False):
     st.session_state.last_chat_update = time.time()
     return audio_file
 
-# Chat loader - history unleashed! 📜🚀
 async def load_chat():
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "📜🚀 - Chat loader - history unleashed!")
@@ -216,7 +212,6 @@ async def load_chat():
         content = await asyncio.to_thread(f.read)
     return content
 
-# User lister - who’s in the gang! 👥🎉
 async def get_user_list(chat_content):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "👥🎉 - User lister - who’s in the gang!")
@@ -227,13 +222,11 @@ async def get_user_list(chat_content):
             users.add(user)
     return sorted(list(users))
 
-# Join checker - been here before? 🚪🔍
 async def has_joined_before(client_id, chat_content):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🚪🔍 - Join checker - been here before?")
     return any(f"Client-{client_id}" in line for line in chat_content.split('\n'))
 
-# Suggestion maker - old quips resurface! 💡📝
 async def get_message_suggestions(chat_content, prefix):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "💡📝 - Suggestion maker - old quips resurface!")
@@ -241,7 +234,6 @@ async def get_message_suggestions(chat_content, prefix):
     messages = [line.split(': ', 1)[1] for line in lines if ': ' in line and line.strip()]
     return [msg for msg in messages if msg.lower().startswith(prefix.lower())][:5]
 
-# Vote saver - cheers recorded! 👍📊
 async def save_vote(file, item, user_hash, username, comment=""):
     await asyncio.to_thread(log_action, username, "👍📊 - Vote saver - cheers recorded!")
     central = pytz.timezone('US/Central')
@@ -254,7 +246,6 @@ async def save_vote(file, item, user_hash, username, comment=""):
         chat_message += f" - {comment}"
     await save_chat_entry(username, chat_message)
 
-# Vote counter - tallying the love! 🏆📈
 async def load_votes(file):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🏆📈 - Vote counter - tallying the love!")
@@ -275,7 +266,6 @@ async def load_votes(file):
                 user_votes.add(vote_key)
     return votes
 
-# Hash generator - secret codes ahoy! 🔑🕵️
 async def generate_user_hash():
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🔑🕵️ - Hash generator - secret codes ahoy!")
@@ -283,7 +273,6 @@ async def generate_user_hash():
         st.session_state.user_hash = hashlib.md5(str(random.getrandbits(128)).encode()).hexdigest()[:8]
     return st.session_state.user_hash
 
-# Audio maker - voices come alive! 🎶🌟
 async def async_edge_tts_generate(text, voice, rate=0, pitch=0, file_format="mp3"):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🎶🌟 - Audio maker - voices come alive!")
@@ -300,7 +289,6 @@ async def async_edge_tts_generate(text, voice, rate=0, pitch=0, file_format="mp3
             f.write(f"[{datetime.now(central).strftime('%Y-%m-%d %H:%M:%S')}] {username}: Audio failed - No audio received for '{text}'\n")
         return None
 
-# Audio player - tunes blast off! 🔊🚀
 def play_and_download_audio(file_path):
     if file_path and os.path.exists(file_path):
         st.audio(file_path)
@@ -312,7 +300,6 @@ def play_and_download_audio(file_path):
         dl_link = f'<a href="data:audio/mpeg;base64,{b64}" download="{os.path.basename(file_path)}">🎵 Download {os.path.basename(file_path)}</a>'
         st.markdown(dl_link, unsafe_allow_html=True)
 
-# Image saver - pics preserved with naming! 📸💾
 async def save_pasted_image(image, username):
     await asyncio.to_thread(log_action, username, "📸💾 - Image saver - pics preserved!")
     img_hash = compute_image_hash(image)
@@ -325,7 +312,6 @@ async def save_pasted_image(image, username):
     st.session_state.image_hashes.add(img_hash)
     return filepath
 
-# Video renderer - movies roll with autoplay! 🎥🎬
 async def get_video_html(video_path, width="100%"):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🎥🎬 - Video renderer - movies roll!")
@@ -334,14 +320,12 @@ async def get_video_html(video_path, width="100%"):
     video_url = f"data:video/mp4;base64,{base64.b64encode(video_data).decode()}"
     return f'<video width="{width}" controls autoplay><source src="{video_url}" type="video/mp4">Your browser does not support the video tag.</video>'
 
-# Audio renderer - sounds soar! 🎶✈️
 async def get_audio_html(audio_path, width="100%"):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🎶✈️ - Audio renderer - sounds soar!")
     audio_url = f"data:audio/mpeg;base64,{base64.b64encode(await asyncio.to_thread(open, audio_path, 'rb').read()).decode()}"
     return f'<audio controls style="width: {width};"><source src="{audio_url}" type="audio/mpeg">Your browser does not support the audio element.</audio>'
 
-# Websocket handler - chat links up! 🌐🔗
 async def websocket_handler(websocket, path):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🌐🔗 - Websocket handler - chat links up!")
@@ -364,7 +348,6 @@ async def websocket_handler(websocket, path):
         if room_id in st.session_state.active_connections and client_id in st.session_state.active_connections[room_id]:
             del st.session_state.active_connections[room_id][client_id]
 
-# Message broadcaster - words fly far! 📢✈️
 async def broadcast_message(message, room_id):
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "📢✈️ - Message broadcaster - words fly far!")
@@ -378,7 +361,6 @@ async def broadcast_message(message, room_id):
         for client_id in disconnected:
             del st.session_state.active_connections[room_id][client_id]
 
-# Server starter - web spins up! 🖥️🌀
 async def run_websocket_server():
     username = st.session_state.get('username', 'System 🌟')
     await asyncio.to_thread(log_action, username, "🖥️🌀 - Server starter - web spins up!")
@@ -387,15 +369,31 @@ async def run_websocket_server():
         st.session_state.server_running = True
         await server.wait_closed()
 
-# Voice processor - speech to text! 🎤📝
-async def process_voice_input(audio_bytes):
-    username = st.session_state.get('username', 'System 🌟')
+async def save_audio_recording(audio_bytes, username, voice):
+    timestamp = format_timestamp_prefix(username)
+    voice_id = voice.split('-')[-1].lower()
+    filename = f"rec_{username}_{voice_id}_{timestamp}.mp3"
+    filepath = os.path.join(AUDIO_DIR, filename)
+    
+    with open(filepath, 'wb') as f:
+        f.write(audio_bytes)
+    
+    cleaned_text = "Audio recording"
+    audio_file = await async_edge_tts_generate(cleaned_text, voice, file_format="mp3")
+    if audio_file:
+        os.rename(audio_file, filepath)
+    
+    return filepath
+
+async def process_voice_input(audio_bytes, username, voice):
     await asyncio.to_thread(log_action, username, "🎤📝 - Voice processor - speech to text!")
     if audio_bytes:
-        text = "Voice input simulation"
+        audio_file = await save_audio_recording(audio_bytes, username, voice)
+        text = f"Voice recording saved: {os.path.basename(audio_file)}"
         await save_chat_entry(username, text)
+        return audio_file
+    return None
 
-# Dummy AI lookup function (replace with actual implementation)
 async def perform_ai_lookup(query, vocal_summary=True, extended_refs=False, titles_summary=True, full_audio=False, useArxiv=True, useArxivAudio=False):
     username = st.session_state.get('username', 'System 🌟')
     result = f"AI Lookup Result for '{query}' (Arxiv: {useArxiv}, Audio: {useArxivAudio})"
@@ -405,12 +403,9 @@ async def perform_ai_lookup(query, vocal_summary=True, extended_refs=False, titl
         if audio_file:
             st.audio(audio_file)
 
-# Delete all user files function
 def delete_user_files():
     protected_files = {'app.py', 'requirements.txt', 'README.md'}
     deleted_files = []
-    
-    # Directories to clear
     directories = [MEDIA_DIR, AUDIO_DIR, CHAT_DIR, VOTE_DIR, HISTORY_DIR]
     
     for directory in directories:
@@ -424,14 +419,12 @@ def delete_user_files():
                             deleted_files.append(file_path)
                         except Exception as e:
                             st.error(f"Failed to delete {file_path}: {e}")
-            # Remove empty directories
             try:
                 shutil.rmtree(directory, ignore_errors=True)
-                os.makedirs(directory, exist_ok=True)  # Recreate empty directory
+                os.makedirs(directory, exist_ok=True)
             except Exception as e:
                 st.error(f"Failed to remove directory {directory}: {e}")
     
-    # Clear session state caches
     st.session_state.image_hashes.clear()
     st.session_state.audio_cache.clear()
     st.session_state.base64_cache.clear()
@@ -439,7 +432,6 @@ def delete_user_files():
     
     return deleted_files
 
-# ASR Component HTML
 ASR_HTML = """
 <html>
 <head>
@@ -525,6 +517,13 @@ ASR_HTML = """
                 status.textContent = 'Stopped';
                 startButton.disabled = false;
                 stopButton.disabled = true;
+                if (fullTranscript) {
+                    sendDataToPython({
+                        value: fullTranscript,
+                        dataType: "json",
+                        stopped: true
+                    });
+                }
             };
 
             clearButton.onclick = () => {
@@ -553,7 +552,11 @@ ASR_HTML = """
                     lastUpdateTime = Date.now();
                     output.textContent = fullTranscript + (interimTranscript ? '... ' + interimTranscript : '');
                     output.scrollTop = output.scrollHeight;
-                    sendDataToPython({value: fullTranscript, dataType: "json"});
+                    sendDataToPython({
+                        value: fullTranscript,
+                        dataType: "json",
+                        stopped: false
+                    });
                 }
             };
 
@@ -603,10 +606,8 @@ ASR_HTML = """
 </html>
 """
 
-# Main execution - let’s roll! 🎲🚀
 def main():
     NODE_NAME, port = get_node_name()
-
     loop = asyncio.new_event_loop()
     asyncio.set_event_loop(loop)
 
@@ -626,21 +627,42 @@ def main():
 
         audio_bytes = audio_recorder()
         if audio_bytes:
-            await process_voice_input(audio_bytes)
-            st.rerun()
+            audio_file = await process_voice_input(audio_bytes, st.session_state.username, st.session_state.voice)
+            if audio_file:
+                st.audio(audio_file)
+                st.rerun()
 
-        # Continuous Speech Input (ASR)
         st.subheader("🎤 Continuous Speech Input")
         asr_component = components.html(ASR_HTML, height=400)
         if asr_component and isinstance(asr_component, dict) and 'value' in asr_component:
             transcript = asr_component['value'].strip()
+            stopped = asr_component.get('stopped', False)
             if transcript and transcript != st.session_state.last_transcript:
                 st.session_state.transcript_history.append(transcript)
-                await save_chat_entry(st.session_state.username, transcript, is_markdown=True)
                 st.session_state.last_transcript = transcript
-                st.rerun()
+                if stopped:
+                    audio_file = await save_audio_recording(audio_bytes, st.session_state.username, st.session_state.voice)
+                    await save_chat_entry(st.session_state.username, f"Voice message: {transcript}\nAudio file: {os.path.basename(audio_file)}", is_markdown=True)
+                    st.rerun()
+
+        st.subheader("🎵 Recorded Audio Files")
+        audio_files = glob.glob(f"{AUDIO_DIR}/rec_{st.session_state.username}_*.mp3")
+        if audio_files:
+            st.write(f"Found {len(audio_files)} recordings for {st.session_state.username}")
+            for audio_file in sorted(audio_files, key=os.path.getmtime, reverse=True):
+                col1, col2 = st.columns([3, 1])
+                with col1:
+                    st.audio(audio_file)
+                    st.write(f"File: {os.path.basename(audio_file)}")
+                with col2:
+                    if audio_file not in st.session_state.base64_cache:
+                        with open(audio_file, "rb") as f:
+                            b64 = base64.b64encode(f.read()).decode()
+                        st.session_state.base64_cache[audio_file] = b64
+                    b64 = st.session_state.base64_cache[audio_file]
+                    dl_link = f'<a href="data:audio/mpeg;base64,{b64}" download="{os.path.basename(audio_file)}">🎵 Download</a>'
+                    st.markdown(dl_link, unsafe_allow_html=True)
 
-        # Load and display chat
         st.subheader(f"{START_ROOM} Chat 💬")
         chat_content = await load_chat()
         chat_lines = chat_content.split('\n')
@@ -759,85 +781,4 @@ def main():
                 await save_chat_entry(st.session_state.username, f"Pasted image: {st.session_state.pasted_image_data}")
                 st.session_state.pasted_image_data = None
             st.session_state.message_text = ''
-            st.rerun()
-
-        tab_main = st.radio("Action:", ["🎤 Voice", "📸 Media", "🔍 ArXiv", "📝 Editor"], horizontal=True)
-        useArxiv = st.checkbox("Search Arxiv for Research Paper Answers", value=True)
-        useArxivAudio = st.checkbox("Generate Audio File for Research Paper Answers", value=False)
-
-        st.subheader("Upload Media 🎨🎶🎥")
-        uploaded_file = st.file_uploader("Upload Media", type=['png', 'jpg', 'mp4', 'mp3'])
-        if uploaded_file:
-            timestamp = format_timestamp_prefix(st.session_state.username)
-            username = st.session_state.username
-            ext = uploaded_file.name.split('.')[-1]
-            file_hash = hashlib.md5(uploaded_file.getbuffer()).hexdigest()[:8]
-            if file_hash not in st.session_state.image_hashes:
-                filename = f"{timestamp}-{file_hash}.{ext}"
-                file_path = os.path.join(MEDIA_DIR, filename)
-                await asyncio.to_thread(lambda: open(file_path, 'wb').write(uploaded_file.getbuffer()))
-                st.success(f"Uploaded {filename}")
-                await save_chat_entry(username, f"Uploaded media: {file_path}")
-                st.session_state.image_hashes.add(file_hash)
-                if file_path.endswith('.mp4'):
-                    st.session_state.media_notifications.append(file_path)
-
-        # Big Red Delete Button
-        st.subheader("🛑 Danger Zone")
-        if st.button("Try Not To Delete It All On Your First Day", key="delete_all", help="Deletes all user-added files!", type="primary", use_container_width=True):
-            deleted_files = delete_user_files()
-            if deleted_files:
-                st.markdown("### 🗑️ Deleted Files:\n" + "\n".join([f"- `{file}`" for file in deleted_files]))
-            else:
-                st.markdown("### 🗑️ Nothing to Delete!")
-            st.session_state.image_hashes.clear()
-            st.session_state.audio_cache.clear()
-            st.session_state.base64_cache.clear()
-            st.session_state.displayed_chat_lines.clear()
-            st.rerun()
-
-        st.subheader("Media Gallery 🎨🎶🎥")
-        media_files = glob.glob(f"{MEDIA_DIR}/*.png") + glob.glob(f"{MEDIA_DIR}/*.jpg") + glob.glob(f"{MEDIA_DIR}/*.mp4") + glob.glob(f"{MEDIA_DIR}/*.mp3")
-        if media_files:
-            media_votes = await load_votes(MEDIA_VOTES_FILE)
-            st.write("### All Media Uploads")
-            seen_files = set()
-            for media_file in sorted(media_files, key=os.path.getmtime, reverse=True):
-                if media_file not in seen_files:
-                    seen_files.add(media_file)
-                    filename = os.path.basename(media_file)
-                    vote_count = media_votes.get(media_file, 0)
-                    col1, col2 = st.columns([3, 1])
-                    with col1:
-                        st.markdown(f"**{filename}**")
-                        if media_file.endswith(('.png', '.jpg')):
-                            st.image(media_file, use_container_width=True)
-                        elif media_file.endswith('.mp4'):
-                            st.markdown(await get_video_html(media_file), unsafe_allow_html=True)
-                        elif media_file.endswith('.mp3'):
-                            st.markdown(await get_audio_html(media_file), unsafe_allow_html=True)
-                    with col2:
-                        if st.button(f"👍 {vote_count}", key=f"media_vote_{media_file}"):
-                            await save_vote(MEDIA_VOTES_FILE, media_file, await generate_user_hash(), st.session_state.username)
-                            st.rerun()
-
-        st.subheader("Refresh ⏳")
-        refresh_rate = st.slider("Refresh Rate", 1, 300, st.session_state.refresh_rate)
-        st.session_state.refresh_rate = refresh_rate
-        timer_placeholder = st.empty()
-        for i in range(st.session_state.refresh_rate, -1, -1):
-            font_name, font_func = random.choice(UNICODE_FONTS)
-            countdown_str = "".join(UNICODE_DIGITS[int(d)] for d in str(i)) if i < 10 else font_func(str(i))
-            timer_placeholder.markdown(f"<p class='timer'>⏳ {font_func('Refresh in:')} {countdown_str}</p>", unsafe_allow_html=True)
-            time.sleep(1)
-        st.rerun()
-
-        st.sidebar.subheader("Chat History 📜")
-        with open(HISTORY_FILE, 'r') as f:
-            history_content = f.read()
-        st.sidebar.markdown(history_content)
-
-    loop.run_until_complete(async_interface())
-
-if __name__ == "__main__":
-    main()
+            st.rerun()