Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,4 +1,3 @@
-import os
 import tempfile
 import logging
 import urllib.parse
@@ -6,78 +5,36 @@ import urllib.parse
 import requests
 import streamlit as st
 from gtts import gTTS
-from dotenv import load_dotenv
 from translatepy import Translator
 
-# -------------------- Load Environment --------------------
-load_dotenv()
-
-# -------------------- Configuration --------------------
-WIKIPEDIA_API_BASE_URL_PREFIX = ".wikipedia.org/w/api.php"
-ENGLISH_SUMMARY_SENTENCES = 5
-WIKIMEDIA_HEADERS = {
-    'User-Agent': 'WikiHealthNavigator/1.0 (contact@example.com)'
-}
-REQUEST_TIMEOUT = 30
+# -------------------- CONFIG --------------------
+REQUEST_TIMEOUT = 20
+HEADERS = {'User-Agent': 'WikiHealthNavigator/1.0'}
 translator = Translator()
 
-logger = logging.getLogger(__name__)
-logger.setLevel(logging.DEBUG)
+logging.basicConfig(level=logging.INFO)
 
-# -------------------- Wikipedia Summary Fetch --------------------
+# -------------------- FETCH SUMMARY --------------------
 @st.cache_data(ttl=3600)
-def fetch_wikipedia_content(query, lang="en", is_fallback=False):
+def fetch_summary(query, lang):
     if not query:
-        return None, None, lang
+        return None, None
 
-    title = query.strip()
-    encoded_title = urllib.parse.quote(title, safe="")
+    title = urllib.parse.quote(query.strip())
+    url = f"https://{lang}.wikipedia.org/api/rest_v1/page/summary/{title}"
 
-    rest_url = f"https://{lang}.wikipedia.org/api/rest_v1/page/summary/{encoded_title}"
     try:
-        resp = requests.get(rest_url, headers=WIKIMEDIA_HEADERS, timeout=REQUEST_TIMEOUT)
-        resp.raise_for_status()
-        j = resp.json()
-        summary = j.get("extract") or j.get("description")
-        full_url = j.get("content_urls", {}).get("desktop", {}).get("page")
-        if summary:
-            return summary, full_url, lang
-    except Exception as e:
-        logger.debug("REST failed: %s", e)
+        r = requests.get(url, headers=HEADERS, timeout=REQUEST_TIMEOUT)
+        r.raise_for_status()
+        data = r.json()
+        return data.get("extract"), data.get("content_urls", {}).get("desktop", {}).get("page")
+    except Exception:
+        return None, None
 
-    try:
-        api = f"https://{lang}{WIKIPEDIA_API_BASE_URL_PREFIX}"
-        params = {
-            "action": "query",
-            "format": "json",
-            "formatversion": 2,
-            "prop": "extracts|info",
-            "inprop": "url",
-            "titles": title,
-            "redirects": 1,
-            "explaintext": True,
-            **({"exsentences": ENGLISH_SUMMARY_SENTENCES} if lang == "en" else {})
-        }
-        res = requests.get(api, params=params, headers=WIKIMEDIA_HEADERS, timeout=REQUEST_TIMEOUT)
-        res.raise_for_status()
-        pages = res.json().get("query", {}).get("pages", [])
-        if pages and not pages[0].get("missing"):
-            return pages[0].get("extract"), pages[0].get("fullurl"), lang
-    except Exception as e:
-        logger.debug("Query API failed: %s", e)
-
-    if not is_fallback and lang != "en":
-        return fetch_wikipedia_content(query, lang="en", is_fallback=True)
-
-    return None, None, lang
-
-# -------------------- Wikipedia Image Fetch --------------------
+# -------------------- FETCH IMAGES (EN ONLY – STABLE) --------------------
 @st.cache_data(ttl=3600)
-def fetch_wikipedia_images(query, lang="en", max_images=3):
-    if not query:
-        return []
-
-    api = f"https://{lang}.wikipedia.org/w/api.php"
+def fetch_images(query, max_images=3):
+    api = "https://en.wikipedia.org/w/api.php"
     params = {
         "action": "query",
         "format": "json",
@@ -88,88 +45,72 @@ def fetch_wikipedia_images(query, lang="en", max_images=3):
     }
 
     try:
-        res = requests.get(api, params=params, headers=WIKIMEDIA_HEADERS, timeout=REQUEST_TIMEOUT)
-        res.raise_for_status()
-        pages = res.json().get("query", {}).get("pages", {})
+        r = requests.get(api, params=params, headers=HEADERS, timeout=REQUEST_TIMEOUT)
+        r.raise_for_status()
+        pages = r.json()["query"]["pages"]
         images = []
 
-        for page in pages.values():
-            if "original" in page:
-                images.append(page["original"]["source"])
+        for p in pages.values():
+            if "original" in p:
+                images.append(p["original"]["source"])
 
         return images[:max_images]
-    except Exception as e:
-        logger.debug("Image fetch failed: %s", e)
+    except Exception:
         return []
 
-# -------------------- Narration --------------------
-def narrate_text(text, lang_code):
-    try:
-        clean = text.strip().replace("\n", " ")
-        if not clean:
-            st.warning("Nothing to narrate.")
-            return
-
-        tts = gTTS(text=clean, lang=lang_code)
-        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as fp:
-            tts.save(fp.name)
-            audio_path = fp.name
-
-        st.audio(audio_path, format="audio/mp3")
-    except Exception as e:
-        st.error(f"Narration error: {e}")
-
-# -------------------- Translation --------------------
-def translate_summary(text, target_lang):
-    try:
-        return translator.translate(text, target_lang).result
-    except Exception:
-        return text
+# -------------------- NARRATION --------------------
+def narrate(text, lang):
+    tts = gTTS(text=text, lang=lang)
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as f:
+        tts.save(f.name)
+        st.audio(f.name)
 
-# -------------------- Streamlit UI --------------------
-st.set_page_config(page_title="WikiHealth Navigator", layout="centered")
+# -------------------- UI --------------------
+st.set_page_config("WikiHealth Navigator", layout="centered")
 st.title("🩺 WikiHealth Navigator")
-st.markdown("Search health topics using Wikipedia (with images & audio)")
 
 languages = {
     "English": "en",
-    "हिन्दी": "hi",
-    "తెలుగు": "te",
-    "தமிழ்": "ta"
+    "Hindi": "hi",
+    "Telugu": "te",
+    "Tamil": "ta"
 }
 
-col1, col2 = st.columns([2, 3])
-with col1:
-    selected = st.selectbox("Select Language:", list(languages.keys()))
-    lang_code = languages[selected]
-with col2:
-    query = st.text_input("Enter a health topic:")
+lang_name = st.selectbox("Select Language", list(languages.keys()))
+lang_code = languages[lang_name]
 
-if st.button("🔍 Search") and query:
+query = st.text_input("Enter health topic")
+
+search = st.button("🔍 Search")
+
+if search and query:
     with st.spinner("Searching Wikipedia..."):
-        summary, url, actual_lang = fetch_wikipedia_content(query, lang_code)
+        summary, url = fetch_summary(query, lang_code)
+
+        if not summary and lang_code != "en":
+            summary, url = fetch_summary(query, "en")
 
         if summary:
-            if actual_lang != lang_code:
-                summary = translate_summary(summary, lang_code)
+            if lang_code != "en":
+                summary = translator.translate(summary, lang_code).result
 
             st.subheader("📝 Summary")
             st.write(summary)
 
-            st.subheader("🖼️ Related Images")
-            images = fetch_wikipedia_images(query, lang_code)
+            st.subheader("🖼️ Images")
+            images = fetch_images(query)
 
             if images:
                 cols = st.columns(len(images))
-                for col, img in zip(cols, images):
-                    col.image(img, use_container_width=True)
+                for c, img in zip(cols, images):
+                    c.image(img, use_container_width=True)
             else:
                 st.info("No images found.")
 
-            if st.button("🔊 Narrate Summary"):
-                narrate_text(summary, lang_code)
+            if st.button("🔊 Narrate"):
+                narrate(summary, lang_code)
 
             if url:
-                st.markdown(f"[Read full article on Wikipedia]({url})")
+                st.markdown(f"[Read full article]({url})")
         else:
-            st.warning("No information found.")
+            st.error("No data found.")