Update backup18.app.py

backup18.app.py

@@ -106,6 +106,10 @@ if 'last_query' not in st.session_state:
 if 'marquee_content' not in st.session_state:
     st.session_state['marquee_content'] = "🚀 Welcome to TalkingAIResearcher | 🤖 Your Research Assistant"
 
+# New: default AutoRun to False (off)
+if 'autorun' not in st.session_state:
+    st.session_state['autorun'] = False
+
 # API Keys
 openai_api_key = os.getenv('OPENAI_API_KEY', "")
 anthropic_key = os.getenv('ANTHROPIC_API_KEY_3', "")
@@ -124,7 +128,14 @@ API_URL = os.getenv('API_URL')
 FILE_EMOJIS = {
     "md": "📝",
     "mp3": "🎵",
-    "wav": "🔊"
+    "wav": "🔊",
+    "pdf": "📕",
+    "mp4": "🎥",
+    "csv": "📈",
+    "xlsx": "📊",
+    "html": "🌐",
+    "py": "🐍",
+    "txt": "📄"
 }
 
 # ─────────────────────────────────────────────────────────
@@ -439,35 +450,45 @@ def display_papers_in_sidebar(papers):
                 st.markdown(f"**Summary:** {paper['summary'][:300]}...")
 
 # ─────────────────────────────────────────────────────────
-# 4. ZIP FUNCTION
+# 4. ZIP & DELETE-ALL UTILS
 # ─────────────────────────────────────────────────────────
 
-def create_zip_of_files(md_files, mp3_files, wav_files, input_question):
+def create_zip_of_all_files():
     """
-    Zip up all relevant files, limiting the final zip name to ~20 chars
+    Zip up all recognized file types, limiting the final zip name to ~20 chars
     to avoid overly long base64 strings. 
     """
-    md_files = [f for f in md_files if os.path.basename(f).lower() != 'readme.md']
-    all_files = md_files + mp3_files + wav_files
+    # Patterns for .md, .pdf, .mp4, .mp3, .wav, .csv, .xlsx, .html, .py, .txt
+    file_patterns = [
+        "*.md", "*.pdf", "*.mp4", "*.mp3", "*.wav",
+        "*.csv", "*.xlsx", "*.html", "*.py", "*.txt"
+    ]
+    all_files = []
+    for pat in file_patterns:
+        all_files.extend(glob.glob(pat))
+    all_files = list(set(all_files))  # unique
+
     if not all_files:
         return None
 
+    # Combine content for naming
     all_content = []
     for f in all_files:
-        if f.endswith('.md'):
-            with open(f, 'r', encoding='utf-8') as file:
-                all_content.append(file.read())
-        elif f.endswith('.mp3') or f.endswith('.wav'):
-            basename = os.path.splitext(os.path.basename(f))[0]
-            words = basename.replace('_', ' ')
-            all_content.append(words)
-    
-    all_content.append(input_question)
+        if f.endswith(".md"):
+            with open(f, "r", encoding="utf-8") as fin:
+                all_content.append(fin.read())
+        else:
+            all_content.append(os.path.basename(f))
+
+    # Add last query if relevant
+    if st.session_state['last_query']:
+        all_content.append(st.session_state['last_query'])
+
     combined_content = " ".join(all_content)
     info_terms = get_high_info_terms(combined_content, top_n=10)
     
     timestamp = format_timestamp_prefix()
-    name_text = '-'.join(term for term in info_terms[:5])  
+    name_text = '-'.join(term for term in info_terms[:5])
     short_zip_name = (timestamp + "_" + name_text)[:20] + ".zip"
 
     with zipfile.ZipFile(short_zip_name, 'w') as z:
@@ -475,6 +496,16 @@ def create_zip_of_files(md_files, mp3_files, wav_files, input_question):
             z.write(f)
     return short_zip_name
 
+def delete_all_files():
+    """Removes all recognized file types from the directory."""
+    file_patterns = [
+        "*.md", "*.pdf", "*.mp4", "*.mp3", "*.wav",
+        "*.csv", "*.xlsx", "*.html", "*.py", "*.txt"
+    ]
+    for pat in file_patterns:
+        for f in glob.glob(pat):
+            os.remove(f)
+
 # ─────────────────────────────────────────────────────────
 # 5. MAIN LOGIC: AI LOOKUP & VOICE INPUT
 # ─────────────────────────────────────────────────────────
@@ -483,9 +514,6 @@ def perform_ai_lookup(q, vocal_summary=True, extended_refs=False,
                      titles_summary=True, full_audio=False):
     """Main routine that uses Anthropic (Claude) + Gradio ArXiv RAG pipeline."""
     start = time.time()
-    ai_constitution = """
-    You are a talented AI coder and songwriter...
-    """
 
     # --- 1) Claude API
     client = anthropic.Anthropic(api_key=anthropic_key)
@@ -516,7 +544,6 @@ def perform_ai_lookup(q, vocal_summary=True, extended_refs=False,
         "mistralai/Mixtral-8x7B-Instruct-v0.1",
         api_name="/update_with_rag_md"
     )[0]
-
     r2 = client.predict(
         q, 
         "mistralai/Mixtral-8x7B-Instruct-v0.1", 
@@ -572,17 +599,21 @@ def process_voice_input(text):
 
 def display_file_history_in_sidebar():
     """
-    Shows a history of each local .md, .mp3, .wav file in descending
+    Shows a history of each recognized file in descending
     order of modification time, with quick icons and optional download links.
     """
     st.sidebar.markdown("---")
     st.sidebar.markdown("### 📂 File History")
 
-    # Gather all files
-    md_files = glob.glob("*.md")
-    mp3_files = glob.glob("*.mp3")
-    wav_files = glob.glob("*.wav")
-    all_files = md_files + mp3_files + wav_files
+    # Patterns for .md, .mp3, .wav, .pdf, .mp4, .csv, .xlsx, .html, .py, .txt
+    patterns = [
+        "*.md", "*.pdf", "*.mp4", "*.mp3", "*.wav",
+        "*.csv", "*.xlsx", "*.html", "*.py", "*.txt"
+    ]
+    all_files = []
+    for p in patterns:
+        all_files.extend(glob.glob(p))
+    all_files = list(set(all_files))  # unique
 
     if not all_files:
         st.sidebar.write("No files found.")
@@ -617,103 +648,93 @@ def display_file_history_in_sidebar():
 # ─────────────────────────────────────────────────────────
 
 def main():
-    # 1) Setup marquee UI in the sidebar
+    """
+    Main Streamlit app. 
+    Now includes:
+      1) Voice & AutoRun at the top of the sidebar,
+      2) File Tools (Delete All / Zip All) in the sidebar,
+      3) A new '📤 Upload' tab, 
+      4) Everything else from your original code snippet.
+    """
+
+    # -- 1) Voice & AutoRun at top of sidebar --
+    st.sidebar.title("Global Settings")
+    selected_voice = st.sidebar.selectbox(
+        "TTS Voice",
+        options=EDGE_TTS_VOICES,
+        index=EDGE_TTS_VOICES.index(st.session_state['tts_voice'])
+    )
+    # Autorun defaults to off (False)
+    st.session_state.autorun = st.sidebar.checkbox("AutoRun", value=st.session_state.autorun)
+
+    # Audio format
+    audio_format = st.sidebar.radio("Audio Format", ["MP3","WAV"], index=0)
+    if selected_voice != st.session_state['tts_voice']:
+        st.session_state['tts_voice'] = selected_voice
+        st.experimental_rerun()
+    if audio_format.lower() != st.session_state['audio_format']:
+        st.session_state['audio_format'] = audio_format.lower()
+        st.experimental_rerun()
+
+    # -- 2) File Tools: Delete All / Zip All
+    st.sidebar.markdown("---")
+    st.sidebar.markdown("### 🗃 File Tools")
+    col_del, col_zip = st.sidebar.columns(2)
+    with col_del:
+        if st.button("🗑 Delete All"):
+            delete_all_files()
+            st.sidebar.success("All recognized files removed!")
+            st.experimental_rerun()
+    with col_zip:
+        if st.button("📦 Zip All"):
+            zip_name = create_zip_of_all_files()
+            if zip_name:
+                st.sidebar.markdown(get_download_link(zip_name, "zip"), unsafe_allow_html=True)
+
+    # -- 3) Marquee Settings
     update_marquee_settings_ui()
     marquee_settings = get_marquee_settings()
 
-    # 2) Display the marquee welcome
+    # -- 4) File History in sidebar
+    display_file_history_in_sidebar()
+
+    # -- 5) Display marquee
     display_marquee(st.session_state['marquee_content'], 
                     {**marquee_settings, "font-size": "28px", "lineHeight": "50px"},
                     key_suffix="welcome")
 
-    # 3) Main action tabs
-    tab_main = st.radio("Action:", ["🎤 Voice", "📸 Media", "🔍 ArXiv", "📝 Editor"], 
-                        horizontal=True)
-
-    # Example custom component usage
-    mycomponent = components.declare_component("mycomponent", path="mycomponent")
-    val = mycomponent(my_input_value="Hello")
-
-    if val:
-        val_stripped = val.replace('\\n', ' ')
-        edited_input = st.text_area("✏️ Edit Input:", value=val_stripped, height=100)
-        run_option = st.selectbox("Model:", ["Arxiv"])
-        col1, col2 = st.columns(2)
-        with col1:
-            autorun = st.checkbox("⚙ AutoRun", value=True)
-        with col2:
-            full_audio = st.checkbox("📚FullAudio", value=False)
-
-        input_changed = (val != st.session_state.old_val)
-
-        if autorun and input_changed:
-            st.session_state.old_val = val
-            st.session_state.last_query = edited_input
-            perform_ai_lookup(edited_input, 
-                              vocal_summary=True, 
-                              extended_refs=False, 
-                              titles_summary=True, 
-                              full_audio=full_audio)
-        else:
-            if st.button("▶ Run"):
-                st.session_state.old_val = val
-                st.session_state.last_query = edited_input
-                perform_ai_lookup(edited_input, 
-                                  vocal_summary=True, 
-                                  extended_refs=False, 
-                                  titles_summary=True, 
-                                  full_audio=full_audio)
-    
-    # ─────────────────���───────────────────────────────────────
-    # TAB: ArXiv
-    # ─────────────────────────────────────────────────────────
-    if tab_main == "🔍 ArXiv":
-        st.subheader("🔍 Query ArXiv")
-        q = st.text_input("🔍 Query:", key="arxiv_query")
-        
-        st.markdown("### 🎛 Options")
-        vocal_summary = st.checkbox("🎙ShortAudio", value=True, key="option_vocal_summary")
-        extended_refs = st.checkbox("📜LongRefs", value=False, key="option_extended_refs")
-        titles_summary = st.checkbox("🔖TitlesOnly", value=True, key="option_titles_summary")
-        full_audio = st.checkbox("📚FullAudio", value=False, key="option_full_audio")
-        full_transcript = st.checkbox("🧾FullTranscript", value=False, key="option_full_transcript")
-        
-        if q and st.button("🔍Run"):
-            st.session_state.last_query = q
-            result = perform_ai_lookup(q, vocal_summary=vocal_summary, extended_refs=extended_refs, 
-                                       titles_summary=titles_summary, full_audio=full_audio)
-            if full_transcript:
-                create_file(q, result, "md")
+    # -- 6) Main action tabs
+    tab_main = st.radio(
+        "Action:", 
+        ["📤 Upload", "🎤 Voice", "📸 Media", "🔍 ArXiv", "📝 Editor"], 
+        horizontal=True
+    )
 
-    # ─────────────────────────────────────────────────────────
-    # TAB: Voice
-    # ─────────────────────────────────────────────────────────
+    # 6a) Upload Tab
+    if tab_main == "📤 Upload":
+        st.header("📤 Upload Files")
+        accepted_types = [
+            # We'll accept basically everything (None in file_uploader),
+            # but let's specify for clarity:
+            "text/plain", "text/markdown", "audio/mpeg", "audio/wav",
+            "image/png", "image/jpeg", "video/mp4", "application/pdf",
+            "application/vnd.ms-excel",
+            "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+            "text/html", "application/octet-stream",
+        ]
+        uploaded = st.file_uploader("Select files to upload:", 
+                                    accept_multiple_files=True, 
+                                    type=None)
+        if uploaded:
+            for uf in uploaded:
+                with open(uf.name, "wb") as outfile:
+                    outfile.write(uf.read())
+            st.success("Uploaded!")
+            st.session_state.should_rerun = True
+
+    # 6b) Voice Tab
     elif tab_main == "🎤 Voice":
         st.subheader("🎤 Voice Input")
-        
-        st.markdown("### 🎤 Voice Settings")
-        selected_voice = st.selectbox(
-            "Select TTS Voice:",
-            options=EDGE_TTS_VOICES,
-            index=EDGE_TTS_VOICES.index(st.session_state['tts_voice'])
-        )
-        
-        st.markdown("### 🔊 Audio Format")
-        selected_format = st.radio(
-            "Choose Audio Format:",
-            options=["MP3", "WAV"],
-            index=0
-        )
-
-        # Update session state if voice/format changes
-        if selected_voice != st.session_state['tts_voice']:
-            st.session_state['tts_voice'] = selected_voice
-            st.rerun()
-        if selected_format.lower() != st.session_state['audio_format']:
-            st.session_state['audio_format'] = selected_format.lower()
-            st.rerun()
-
-        # Input text
         user_text = st.text_area("💬 Message:", height=100)
         user_text = user_text.strip().replace('\n', ' ')
 
@@ -725,16 +746,10 @@ def main():
             st.write("**You:**", c["user"])
             st.write("**Response:**", c["claude"])
 
-    # ─────────────────────────────────────────────────────────
-    # TAB: Media
-    # ─────────────────────────────────────────────────────────
+    # 6c) Media Tab
     elif tab_main == "📸 Media":
         st.header("📸 Media Gallery")
-
-        # By default, show audio first
         tabs = st.tabs(["🎵 Audio", "🖼 Images", "🎥 Video"])
-        
-        # AUDIO sub-tab
         with tabs[0]:
             st.subheader("🎵 Audio Files")
             audio_files = glob.glob("*.mp3") + glob.glob("*.wav")
@@ -747,8 +762,6 @@ def main():
                         st.markdown(dl_link, unsafe_allow_html=True)
             else:
                 st.write("No audio files found.")
-        
-        # IMAGES sub-tab
         with tabs[1]:
             st.subheader("🖼 Image Files")
             imgs = glob.glob("*.png") + glob.glob("*.jpg") + glob.glob("*.jpeg")
@@ -760,8 +773,6 @@ def main():
                         st.image(Image.open(f), use_container_width=True)
             else:
                 st.write("No images found.")
-        
-        # VIDEO sub-tab
         with tabs[2]:
             st.subheader("🎥 Video Files")
             vids = glob.glob("*.mp4") + glob.glob("*.mov") + glob.glob("*.avi")
@@ -772,30 +783,45 @@ def main():
             else:
                 st.write("No videos found.")
 
-    # ─────────────────────────────────────────────────────────
-    # TAB: Editor
-    # ─────────────────────────────────────────────────────────
-    elif tab_main == "📝 Editor":
-        st.write("Select or create a file to edit. (Currently minimal demo)")
+    # 6d) ArXiv Tab
+    elif tab_main == "🔍 ArXiv":
+        st.subheader("🔍 Query ArXiv")
+        q = st.text_input("🔍 Query:", key="arxiv_query")
+        
+        st.markdown("### 🎛 Options")
+        st.write("(AutoRun is in the sidebar.)")
+        extended_refs = st.checkbox("📜LongRefs", value=False, key="option_extended_refs")
+        titles_summary = st.checkbox("🔖TitlesOnly", value=True, key="option_titles_summary")
+        full_audio = st.checkbox("📚FullAudio", value=False, key="option_full_audio")
+        full_transcript = st.checkbox("🧾FullTranscript", value=False, key="option_full_transcript")
+        
+        if q and st.button("🔍Run"):
+            st.session_state.last_query = q
+            result = perform_ai_lookup(q, 
+                                       extended_refs=extended_refs, 
+                                       titles_summary=titles_summary, 
+                                       full_audio=full_audio)
+            if full_transcript:
+                create_file(q, result, "md")
 
-    # ─────────────────────────────────────────────────────────
-    # SIDEBAR: FILE HISTORY
-    # ─────────────────────────────────────────────────────────
-    display_file_history_in_sidebar()
+        # If AutoRun is ON and user typed something
+        if st.session_state.autorun and q:
+            st.session_state.last_query = q
+            result = perform_ai_lookup(q, 
+                                       extended_refs=extended_refs, 
+                                       titles_summary=titles_summary, 
+                                       full_audio=full_audio)
+            if full_transcript:
+                create_file(q, result, "md")
 
-    # Some light CSS styling
-    st.markdown("""
-    <style>
-        .main { background: linear-gradient(to right, #1a1a1a, #2d2d2d); color: #fff; }
-        .stMarkdown { font-family: 'Helvetica Neue', sans-serif; }
-        .stButton>button { margin-right: 0.5rem; }
-    </style>
-    """, unsafe_allow_html=True)
+    # 6e) Editor Tab
+    elif tab_main == "📝 Editor":
+        st.write("Select or create a file to edit. (Currently minimal demo)")
 
     # Rerun if needed
     if st.session_state.should_rerun:
         st.session_state.should_rerun = False
-        st.rerun()
+        st.experimental_rerun()
 
 if __name__ == "__main__":
     main()