Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,5 +1,5 @@
 # ==========================================================
-# streamlit_app.py — Stable Layout (English Only)
+# streamlit_app.py — Commit 2 (Stable)
 # ==========================================================
 import os
 import re
@@ -29,50 +29,6 @@ os.environ.update({
     "HF_MODULES_CACHE": CACHE_DIR,
 })
 
-# ==========================================================
-# 🧠 SMART SUGGESTION GENERATOR
-# ==========================================================
-def generate_dynamic_suggestions_from_toc(toc, chunks, doc_name="Document"):
-    if not toc or not chunks:
-        return ["How do I start using this guide?", "What does this document cover?"]
-
-    titles = []
-    for sec, raw_title in toc:
-        title = re.sub(r"^\s*[\dA-Za-z.\-]+\s*", "", raw_title)
-        title = re.sub(r"\.{2,}\s*\d+$", "", title).strip()
-        if 4 < len(title) < 120:
-            titles.append(title)
-
-    context_sample = " ".join(chunks[:3])[:4000]
-    prompt = f"""
-You are a content assistant. Based on the TOC and text, generate 5–7 short natural questions.
-Each question <18 words, ends with '?', sounds human. Document: "{doc_name}"
-TOC:
-{chr(10).join(['- ' + t for t in titles[:8]])}
-Sample:
-{context_sample}
-"""
-
-    try:
-        ai_response = genai_generate(prompt)
-        lines = [ln.strip() for ln in ai_response.splitlines() if ln.strip()]
-        out = []
-        for ln in lines:
-            q = re.sub(r"^[\-\u2022\*\d\.\)\s]+", "", ln).strip()
-            if not q.endswith("?") and re.match(r"(?i)^(what|how|why|where|who|when|which|can|does|is|are)\b", q):
-                q += "?"
-            if 8 <= len(q) <= 140:
-                out.append(q)
-        uniq = []
-        seen = set()
-        for q in out:
-            if q.lower() not in seen:
-                seen.add(q.lower())
-                uniq.append(q)
-        return uniq[:7] or [f"What should I know about {t.rstrip('.')}?" for t in titles[:7]]
-    except Exception:
-        return ["How do I start using this guide?", "What does this document cover?"]
-
 # ==========================================================
 # 🎨 STYLING
 # ==========================================================
@@ -80,17 +36,12 @@ st.markdown("""
 <style>
 div.block-container {padding-top:1.2rem;max-width:1080px;}
 h1,h2,h3{color:#f3f4f6;font-weight:600;}
-.suggest-chip{background:#0f1724;border:1px solid #374151;border-radius:14px;
-color:#e6eef8;padding:8px 12px;cursor:pointer;font-size:13px;margin:6px 6px 6px 0;
-display:inline-block;transition:background 0.2s,transform 0.1s;}
-.suggest-chip:hover{background:#1e3a8a;transform:translateY(-2px);}
 .answer-box{background:linear-gradient(180deg,#0b1220,#071027);
 border-left:4px solid #3b82f6;border-radius:8px;padding:16px 18px;color:#e6eef8;
 margin-top:12px;box-shadow:0 4px 14px rgba(0,0,0,0.35);}
 .stTextInput>div>div>input{background-color:#0f172a!important;color:#f1f5f9!important;
 border-radius:6px!important;border:1px solid #334155!important;padding:8px 10px!important;
 font-size:15px!important;}
-.small-link{font-size:13px;color:#60a5fa;cursor:pointer;}
 </style>
 """, unsafe_allow_html=True)
 
@@ -107,14 +58,13 @@ with st.sidebar:
     )
     st.markdown("---")
 
-    # 📚 Registry display
     if "registry" in st.session_state:
         registry = st.session_state["registry"]
-        registered = registry.list_docs() if hasattr(registry, "list_docs") else []
-        if registered:
+        registered_docs = registry.list_docs() if hasattr(registry, "list_docs") else []
+        if registered_docs:
             with st.expander("📚 Registered Documents", expanded=False):
-                for i, d in enumerate(registered, 1):
-                    st.markdown(f"**{i}. {d.get('name','?')}** — {d.get('num_chunks','?')} chunks *(TOC: {d.get('toc_source','—')})*")
+                for i, doc in enumerate(registered_docs, start=1):
+                    st.markdown(f"**{i}. {doc.get('name','?')}** — {doc.get('num_chunks','?')} chunks *(TOC: {doc.get('toc_source','—')})*")
         else:
             st.caption("📭 No documents registered yet.")
     else:
@@ -123,7 +73,6 @@ with st.sidebar:
     st.markdown("---")
     show_dev = st.checkbox("Show advanced settings (for developers)", value=False)
     if show_dev:
-        st.markdown("### ⚙️ Developer Options")
         chunk_size = st.slider("Chunk Size", 200, 1500, 1000, step=50)
         overlap = st.slider("Chunk Overlap", 50, 200, 120, step=10)
         top_k = st.slider("Top K Results", 1, 10, 5)
@@ -132,32 +81,12 @@ with st.sidebar:
     st.markdown("---")
     st.caption("✨ Built by Shubham Sharma")
 
-# ==========================================================
-# 🧠 SESSION STATE
-# ==========================================================
-for key, val in {
-    "user_query_input": "",
-    "show_more": False,
-    "selected_suggestion": None,
-    "query_suggestions_fixed": None,
-    "last_doc": None,
-    "doc_lang": "en",
-}.items():
-    if key not in st.session_state:
-        st.session_state[key] = val
-
-def set_user_query(q, idx):
-    st.session_state["user_query_input"] = q
-    st.session_state["selected_suggestion"] = idx
-    st.experimental_rerun()
-
 # ==========================================================
 # 📄 MAIN SECTION
 # ==========================================================
 st.title("📄 Enterprise Knowledge Assistant")
 st.caption("Query SAP documentation and enterprise PDFs — powered by reasoning and retrieval.")
 
-# ✅ FIXED: must be defined before document-handling logic
 doc_choice = st.radio(
     "Select a document:",
     ["-- Select --", "Sample PDF", "Upload Custom PDF"],
@@ -165,7 +94,7 @@ doc_choice = st.radio(
 )
 
 # ==========================================================
-# 📂 DOCUMENT HANDLING
+# 📂 DOCUMENT HANDLING (Commit 2)
 # ==========================================================
 def _hash_content(file_path):
     h = hashlib.sha256()
@@ -174,11 +103,6 @@ def _hash_content(file_path):
             h.update(chunk)
     return h.hexdigest()[:12]
 
-def refresh_suggestions(doc_name, toc, chunks):
-    st.session_state["query_suggestions_fixed"] = generate_dynamic_suggestions_from_toc(
-        toc, chunks, doc_name
-    )
-    st.session_state.update({"user_query_input": "", "selected_suggestion": None, "show_more": False})
 
 if doc_choice == "-- Select --":
     st.info("⬅️ Select or upload a document to begin.")
@@ -198,96 +122,84 @@ else:
 
     if temp_path:
         doc_name = os.path.basename(temp_path)
-        doc_hash = _hash_content(temp_path)
-        doc_identifier = f"{doc_name}_{doc_hash}"
+        file_hash = _hash_content(temp_path)
+        doc_identifier = f"{doc_name}_{file_hash}"
 
-        # --- registry check
+        # ✅ Registry initialization
         if "registry" not in st.session_state:
             st.session_state["registry"] = DocumentRegistry()
         registry = st.session_state["registry"]
 
-        existing = next((d for d in registry.list_docs() if d["name"] == doc_name), None)
-        if existing:
-            doc_data = registry.get_doc(existing["name"])
+        # ✅ Reuse if already processed
+        if doc_name in [d["name"] for d in registry.list_docs()]:
+            st.session_state["status_text"] = f"✅ {doc_name} already processed — loaded from registry."
+            doc_data = registry.get_doc(doc_name)
             st.session_state.update({
                 "text": doc_data.get("text", ""),
                 "toc": doc_data.get("toc", []),
                 "chunks": doc_data.get("chunks", []),
                 "embeddings": doc_data.get("embeddings"),
                 "index": doc_data.get("index"),
-                "doc_ready": True,
-                "active_doc": existing["name"],
-                "status_text": f"✅ {doc_name} already processed — loaded from registry."
+                "doc_ready": True
             })
-            refresh_suggestions(existing["name"], st.session_state["toc"], st.session_state["chunks"])
-            st.experimental_rerun()
-
-        status = st.empty()
-        status.info("📤 Upload complete — reading document...")
-
-        text, toc, toc_source = extract_text_from_pdf(temp_path)
-        status.info("📑 Parsing and chunking document...")
-        chunks = chunk_text(text, chunk_size=chunk_size, overlap=overlap)
+        else:
+            status = st.empty()
+            status.info("📤 Upload complete — reading document...")
 
-        status.info("🧠 Building embeddings and search index...")
-        embeddings = cache_embeddings(doc_name, chunks, embed_chunks)
-        index = build_faiss_index(embeddings)
+            text, toc, toc_source = extract_text_from_pdf(temp_path)
+            status.info("📑 Parsing and chunking document...")
+            chunks = chunk_text(text, chunk_size=chunk_size, overlap=overlap)
+            status.info("🧠 Building embeddings and search index...")
+            embeddings = cache_embeddings(doc_name, chunks, embed_chunks)
+            index = build_faiss_index(embeddings)
 
-        doc_id = registry.register(temp_path, chunks, embeddings, index)
-        st.session_state["active_doc"] = doc_id
-        status.success("✅ Document processed successfully — ready to query!")
-        refresh_suggestions(doc_name, toc, chunks)
+            doc_id = registry.register(temp_path, chunks, embeddings, index)
+            st.session_state["active_doc"] = doc_id
+            status.success("✅ Document processed successfully — ready to query!")
 
-        st.session_state.update({
-            "text": text,
-            "toc": toc,
-            "chunks": chunks,
-            "embeddings": embeddings,
-            "index": index,
-            "doc_ready": True,
-            "last_doc": doc_identifier,
-            "status_text": "✅ Document processed successfully — ready to query!"
-        })
-        st.experimental_rerun()
+            st.session_state.update({
+                "text": text,
+                "toc": toc,
+                "chunks": chunks,
+                "embeddings": embeddings,
+                "index": index,
+                "doc_ready": True,
+                "last_doc": doc_identifier,
+                "status_text": "✅ Document processed successfully — ready to query!"
+            })
 
-if st.session_state.get("doc_ready"):
-    st.info(st.session_state.get("status_text", "📄 Ready for queries."))
-    st.markdown("### 💬 Ask the Assistant")
+    # --- Ask section ---
+    if st.session_state.get("doc_ready"):
+        st.info(st.session_state.get("status_text", "📄 Ready for queries."))
+        st.markdown("### 💬 Ask the Assistant")
 
-    suggs = st.session_state.get("query_suggestions_fixed", [])
-    if suggs:
-        visible = suggs if st.session_state["show_more"] else suggs[:3]
+        query_suggestions = ["How do I start using this guide?", "What are the prerequisites?", "What is covered in this document?"]
+        visible = query_suggestions if st.session_state["show_more"] else query_suggestions[:3]
         cols = st.columns(min(3, len(visible)))
         for i, q in enumerate(visible):
             if cols[i % 3].button(f"💬 {q}", key=f"sugg_{i}"):
-                set_user_query(q, i)
+                st.session_state["user_query_input"] = q
+                st.experimental_rerun()
 
         toggle_text = "Show less ▲" if st.session_state["show_more"] else "Show more ▼"
         if st.button(toggle_text):
             st.session_state["show_more"] = not st.session_state["show_more"]
             st.experimental_rerun()
 
-    user_query = st.text_input("Type your question or click one above:", key="user_query_input")
-    if user_query.strip():
-        reasoning_mode = mode == "Extended (Document + General)"
-        with st.spinner("💭 Generating your answer..."):
-            retrieved = retrieve_chunks(
-                user_query,
-                st.session_state["index"],
-                st.session_state["chunks"],
-                top_k=top_k,
-                embeddings=st.session_state["embeddings"]
-            )
-            answer = generate_answer(user_query, retrieved, reasoning_mode=reasoning_mode)
-            st.session_state["retrieved"] = retrieved
-        st.markdown("### 🤖 Assistant’s Answer")
-        st.markdown(f"<div class='answer-box'>{answer}</div>", unsafe_allow_html=True)
-
-# ==========================================================
-# 🎨 Sidebar scroll style
-# ==========================================================
-st.markdown("""
-<style>
-section[data-testid="stSidebar"] div.stExpander {max-height:480px;overflow-y:auto;}
-</style>
-""", unsafe_allow_html=True)
+        user_query = st.text_input("Type your question or click one above:", key="user_query_input")
+
+        if user_query.strip():
+            reasoning_mode = mode == "Extended (Document + General)"
+            with st.spinner("💭 Generating your answer..."):
+                retrieved = retrieve_chunks(
+                    user_query,
+                    st.session_state["index"],
+                    st.session_state["chunks"],
+                    top_k=top_k,
+                    embeddings=st.session_state["embeddings"]
+                )
+                answer = generate_answer(user_query, retrieved, reasoning_mode=reasoning_mode)
+                st.session_state["retrieved"] = retrieved
+
+            st.markdown("### 🤖 Assistant’s Answer")
+            st.markdown(f"<div class='answer-box'>{answer}</div>", unsafe_allow_html=True)