Update app.py

app.py

@@ -1,51 +1,28 @@
-# requirements.txt additions:
-"""
-streamlit-marquee
-"""
-
-# app.py
 import streamlit as st
 import anthropic, openai, base64, cv2, glob, json, math, os, pytz, random, re, requests, time, zipfile
-import plotly.graph_objects as go
-import streamlit.components.v1 as components
 from datetime import datetime
 from audio_recorder_streamlit import audio_recorder
-from collections import defaultdict, deque, Counter
+from collections import defaultdict, Counter
 from dotenv import load_dotenv
 from gradio_client import Client
 from huggingface_hub import InferenceClient
-from io import BytesIO
 from PIL import Image
-from PyPDF2 import PdfReader
-from urllib.parse import quote
-from xml.etree import ElementTree as ET
 from openai import OpenAI
-import extra_streamlit_components as stx
 import asyncio
 import edge_tts
-from streamlit_marquee import st_marquee
+from streamlit_marquee import streamlit_marquee
 
-# Core setup
 st.set_page_config(
     page_title="🚲TalkingAIResearcher🏆",
     page_icon="🚲🏆",
-    layout="wide",
-    initial_sidebar_state="auto",
+    layout="wide"
 )
 
-# Initialize session state
-if 'tts_voice' not in st.session_state:
-    st.session_state['tts_voice'] = "en-US-AriaNeural"
-if 'audio_format' not in st.session_state:
-    st.session_state['audio_format'] = 'mp3'
-if 'scroll_text' not in st.session_state:
-    st.session_state['scroll_text'] = ''
-
 EDGE_TTS_VOICES = [
     "en-US-AriaNeural",
-    "en-US-GuyNeural",
+    "en-US-GuyNeural", 
     "en-US-JennyNeural",
-    "en-GB-SoniaNeural",
+    "en-GB-SoniaNeural"
 ]
 
 FILE_EMOJIS = {
@@ -53,81 +30,113 @@ FILE_EMOJIS = {
     "mp3": "🎵",
     "wav": "🔊",
     "txt": "📄",
-    "pdf": "📑",
-    "json": "📊",
-    "csv": "📈",
-    "zip": "📦"
+    "pdf": "📑"
 }
 
+# Initialize session states
+if 'tts_voice' not in st.session_state:
+    st.session_state['tts_voice'] = EDGE_TTS_VOICES[0]
+if 'audio_format' not in st.session_state:
+    st.session_state['audio_format'] = 'mp3'
+if 'messages' not in st.session_state:
+    st.session_state['messages'] = []
+if 'chat_history' not in st.session_state:
+    st.session_state['chat_history'] = []
+if 'viewing_prefix' not in st.session_state:
+    st.session_state['viewing_prefix'] = None
+if 'should_rerun' not in st.session_state:
+    st.session_state['should_rerun'] = False
+
+# API Setup
+openai_client = OpenAI(api_key=os.getenv('OPENAI_API_KEY'))
+claude_client = anthropic.Anthropic(api_key=os.getenv('ANTHROPIC_API_KEY'))
+
 @st.cache_resource
 def get_cached_audio_b64(file_path):
-    """Cache audio file as base64"""
     with open(file_path, "rb") as f:
         return base64.b64encode(f.read()).decode()
 
 def beautify_filename(filename):
-    """Make filename more readable"""
     name = os.path.splitext(filename)[0]
-    name = name.replace('_', ' ').replace('.', ' ')
-    return name
-
-def load_files_for_sidebar():
-    """Load and group files by timestamp prefix"""
-    md_files = glob.glob("*.md")
-    mp3_files = glob.glob("*.mp3") 
-    wav_files = glob.glob("*.wav")
+    return name.replace('_', ' ').replace('.', ' ')
 
-    md_files = [f for f in md_files if os.path.basename(f).lower() != 'readme.md']
+def create_zip_of_files(md_files, mp3_files, wav_files, query=''):
     all_files = md_files + mp3_files + wav_files
+    if not all_files: return None
+    
+    timestamp = datetime.now().strftime("%y%m_%H%M")
+    zip_name = f"{timestamp}_archive.zip"
+    with zipfile.ZipFile(zip_name, 'w') as z:
+        for f in all_files:
+            z.write(f)
+    return zip_name
 
+def get_download_link(file_path, file_type="zip"):
+    with open(file_path, "rb") as f:
+        b64 = base64.b64encode(f.read()).decode()
+    ext_map = {'zip': '📦', 'mp3': '🎵', 'wav': '🔊', 'md': '📝'}
+    emoji = ext_map.get(file_type, '')
+    return f'<a href="data:application/{file_type};base64,{b64}" download="{os.path.basename(file_path)}">{emoji} Download {os.path.basename(file_path)}</a>'
+
+def load_files_for_sidebar():
+    files = [f for f in glob.glob("*.*") if not f.lower().endswith('readme.md')]
     groups = defaultdict(list)
-    for f in all_files:
+    for f in files:
         basename = os.path.basename(f)
         group_name = basename[:9] if len(basename) >= 9 else 'Other'
         groups[group_name].append(f)
-
     return sorted(groups.items(), 
                  key=lambda x: max(os.path.getmtime(f) for f in x[1]), 
                  reverse=True)
 
+def display_marquee_controls():
+    st.sidebar.markdown("### 🎯 Marquee Settings")
+    cols = st.sidebar.columns(2)
+    with cols[0]:
+        bg_color = st.color_picker("🎨 Background", "#1E1E1E")
+        text_color = st.color_picker("✍️ Text", "#FFFFFF")
+    with cols[1]:
+        font_size = st.slider("📏 Size", 10, 24, 14)
+        duration = st.slider("⏱️ Speed", 1, 20, 10)
+    
+    return {
+        "background": bg_color,
+        "color": text_color,
+        "font-size": f"{font_size}px",
+        "animationDuration": f"{duration}s",
+        "width": "100%",
+        "lineHeight": "35px"
+    }
+
 def display_file_manager_sidebar(groups_sorted):
-    """Enhanced sidebar with audio players and beautified names"""
     st.sidebar.title("📚 File Manager")
-
-    all_md, all_mp3, all_wav = [], [], []
+    all_files = {'md': [], 'mp3': [], 'wav': []}
+    
     for _, files in groups_sorted:
         for f in files:
-            if f.endswith(".md"): all_md.append(f)
-            elif f.endswith(".mp3"): all_mp3.append(f)
-            elif f.endswith(".wav"): all_wav.append(f)
+            ext = os.path.splitext(f)[1].lower().strip('.')
+            if ext in all_files:
+                all_files[ext].append(f)
 
-    # File management buttons
     cols = st.sidebar.columns(4)
-    with cols[0]:
-        if st.button("🗑️ MD"): 
-            [os.remove(f) for f in all_md]
-            st.session_state.should_rerun = True
-    with cols[1]:
-        if st.button("🗑️ MP3"):
-            [os.remove(f) for f in all_mp3]
-            st.session_state.should_rerun = True
-    with cols[2]:
-        if st.button("🗑️ WAV"):
-            [os.remove(f) for f in all_wav]
-            st.session_state.should_rerun = True
-    with cols[3]:
-        if st.button("📦 Zip"):
-            zip_name = create_zip_of_files(all_md, all_mp3, all_wav, 
-                                         st.session_state.get('last_query', ''))
-            if zip_name:
-                st.sidebar.markdown(get_download_link(zip_name), unsafe_allow_html=True)
-
-    # Display file groups
+    for i, (ext, files) in enumerate(all_files.items()):
+        with cols[i]:
+            if st.button(f"🗑️ {ext.upper()}"):
+                [os.remove(f) for f in files]
+                st.session_state.should_rerun = True
+                
+    if st.sidebar.button("📦 Zip All"):
+        zip_name = create_zip_of_files(
+            all_files['md'], all_files['mp3'], all_files['wav']
+        )
+        if zip_name:
+            st.sidebar.markdown(get_download_link(zip_name), unsafe_allow_html=True)
+
     for group_name, files in groups_sorted:
-        timestamp_dt = datetime.strptime(group_name, "%y%m_%H%M") if len(group_name) == 9 else None
-        group_label = timestamp_dt.strftime("%Y-%m-%d %H:%M") if timestamp_dt else group_name
+        timestamp = (datetime.strptime(group_name, "%y%m_%H%M").strftime("%Y-%m-%d %H:%M") 
+                    if len(group_name) == 9 else group_name)
         
-        with st.sidebar.expander(f"📁 {group_label} ({len(files)})", expanded=True):
+        with st.sidebar.expander(f"📁 {timestamp} ({len(files)})", expanded=True):
             c1, c2 = st.columns(2)
             with c1:
                 if st.button("👀", key=f"view_{group_name}"):
@@ -144,53 +153,119 @@ def display_file_manager_sidebar(groups_sorted):
                 st.write(f"{emoji} **{pretty_name}**")
                 
                 if ext in ['mp3', 'wav']:
-                    audio_b64 = get_cached_audio_b64(f)
                     st.audio(f)
-                    cols = st.columns([3,1])
-                    with cols[1]:
-                        if st.button("🔄", key=f"loop_{f}"):
-                            components.html(
-                                f'''
-                                <audio id="player_{f}" loop>
-                                    <source src="data:audio/{ext};base64,{audio_b64}">
-                                </audio>
-                                <script>
-                                    document.getElementById("player_{f}").play();
-                                </script>
-                                ''',
-                                height=0
-                            )
+                    if st.button("🔄", key=f"loop_{f}"):
+                        audio_b64 = get_cached_audio_b64(f)
+                        st.components.v1.html(
+                            f'''
+                            <audio id="player_{f}" loop>
+                                <source src="data:audio/{ext};base64,{audio_b64}">
+                            </audio>
+                            <script>
+                                document.getElementById("player_{f}").play();
+                            </script>
+                            ''',
+                            height=0
+                        )
+
+async def edge_tts_generate(text, voice, file_format="mp3"):
+    text = re.sub(r'\s+', ' ', text).strip()
+    if not text: return None
+    communicate = edge_tts.Communicate(text, voice)
+    filename = f"{datetime.now().strftime('%y%m_%H%M')}_{voice}.{file_format}"
+    await communicate.save(filename)
+    return filename
+
+def parse_arxiv_refs(text):
+    papers = []
+    current_paper = None
+    
+    for line in text.split('\n'):
+        if '|' in line:
+            if current_paper:
+                papers.append(current_paper)
+            parts = line.strip('* ').split('|')
+            current_paper = {
+                'date': parts[0].strip(),
+                'title': parts[1].strip(),
+                'authors': '',
+                'summary': '',
+                'id': re.search(r'(\d{4}\.\d{5})', line).group(1) if re.search(r'(\d{4}\.\d{5})', line) else ''
+            }
+        elif current_paper:
+            if not current_paper['authors']:
+                current_paper['authors'] = line.strip('* ')
+            else:
+                current_paper['summary'] += ' ' + line.strip()
+    
+    if current_paper:
+        papers.append(current_paper)
+    return papers
+
+def perform_ai_lookup(query):
+    client = Client("awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern")
+    response = client.predict(
+        query, 20, "Semantic Search", 
+        "mistralai/Mixtral-8x7B-Instruct-v0.1",
+        api_name="/update_with_rag_md"
+    )
+    
+    papers = parse_arxiv_refs(response[0])
+    marquee_settings = display_marquee_controls()
+    
+    for paper in papers:
+        content = f"📄 {paper['title']} | 👤 {paper['authors']} | 📝 {paper['summary']}"
+        streamlit_marquee(
+            content=content,
+            **marquee_settings,
+            key=f"paper_{paper['id'] or random.randint(1000,9999)}"
+        )
+        st.write("")  # Spacing
+    
+    return papers
 
 def main():
-    # Add scrolling banner
-    st_marquee(
-        text=" | ".join(st.session_state.get('scroll_text', '🚀 Welcome to TalkingAIResearcher').split('\n')),
-        font_size=20,
+    marquee_settings = display_marquee_controls()
+    
+    streamlit_marquee(
+        content="🚀 Welcome to TalkingAIResearcher | 🤖 Your Research Assistant",
+        **marquee_settings,
+        key="welcome"
     )
+    
+    tab = st.radio("Action:", ["🎤 Voice", "🔍 ArXiv", "📝 Editor"], horizontal=True)
+    
+    if tab == "🔍 ArXiv":
+        query = st.text_input("🔍 Search:")
+        if query:
+            papers = perform_ai_lookup(query)
+            st.write(f"Found {len(papers)} papers")
+    
+    groups = load_files_for_sidebar()
+    display_file_manager_sidebar(groups)
+    
+    if st.session_state.should_rerun:
+        st.session_state.should_rerun = False
+        st.rerun()
 
-    # Rest of the main UI code...
-    # (Keep existing main() implementation but with beautified filenames)
+# Condensed sidebar markdown
+sidebar_md = """# 📚 Research Papers
 
-# Compressed sidebar markdown
-sidebar_md = """
-# 🧠 AGI Levels
+## 🧠 AGI Levels
 L0 ❌ No AI
-L1 🌱 Emerging (ChatGPT, Bard)
-L2 💪 Competent (Watson)
-L3 🎯 Expert (DALL·E)
-L4 🏆 Virtuoso (AlphaGo)
-L5 🚀 Superhuman (AlphaFold)
-
-# 🧬 AlphaFold2
-1. 🧬 Input Seq
-2. 🔍 DB Search
-3. 🧩 MSA
-4. 📑 Templates  
-5. 🔄 Evoformer
-6. 🧱 Structure
-7. 🎯 3D Predict
-8. ♻️ Recycle x3
-"""
-
-if __name__=="__main__":
+L1 🌱 ChatGPT/Bard [2303.08774v1](https://arxiv.org/abs/2303.08774) [PDF](https://arxiv.org/pdf/2303.08774.pdf)
+L2 💪 Watson [2201.11903v1](https://arxiv.org/abs/2201.11903) [PDF](https://arxiv.org/pdf/2201.11903.pdf)
+L3 🎯 DALL·E [2204.06125v1](https://arxiv.org/abs/2204.06125) [PDF](https://arxiv.org/pdf/2204.06125.pdf)
+L4 🏆 AlphaGo [1712.01815v1](https://arxiv.org/abs/1712.01815) [PDF](https://arxiv.org/pdf/1712.01815.pdf) 
+L5 🚀 AlphaFold [2203.15556v1](https://arxiv.org/abs/2203.15556) [PDF](https://arxiv.org/pdf/2203.15556.pdf)
+
+## 🧬 AlphaFold2 
+[2203.15556v1](https://arxiv.org/abs/2203.15556) [PDF](https://arxiv.org/pdf/2203.15556.pdf)
+1. 🧬 Input Seq → 2. 🔍 DB Search → 3. 🧩 MSA
+4. 📑 Templates → 5. 🔄 Evoformer → 6. 🧱 Structure
+7. 🎯 3D Predict → 8. ♻️ Recycle x3"""
+
+st.sidebar.markdown(sidebar_md)
+
+if __name__ == "__main__":
     main()