Update app.py

app.py

@@ -4,22 +4,14 @@ import os
 from urllib.parse import quote
 
 # =====================================================
-# PAGE CONFIG
+# PAGE
 # =====================================================
 
-st.set_page_config(
-    page_title="Biodiversity GIS Explorer",
-    layout="wide"
-)
-
+st.set_page_config(page_title="Biodiversity GIS Explorer", layout="wide")
 st.title("🌍 Biodiversity GIS Explorer")
 
-st.write(
-    "Explore biodiversity species using GBIF + AI analysis."
-)
-
 # =====================================================
-# GEMINI API
+# GEMINI
 # =====================================================
 
 def call_gemini(prompt, api_key):
@@ -30,467 +22,231 @@ def call_gemini(prompt, api_key):
     url = f"https://generativelanguage.googleapis.com/v1beta/models/gemini-3.1-flash-lite:generateContent?key={api_key}"
 
     payload = {
-        "contents": [
-            {
-                "parts": [{"text": prompt}]
-            }
-        ]
+        "contents": [{"parts": [{"text": prompt}]}]
     }
 
     try:
-
-        response = requests.post(url, json=payload)
-
-        result = response.json()
-
-        return result["candidates"][0]["content"]["parts"][0]["text"]
-
+        r = requests.post(url, json=payload)
+        data = r.json()
+        return data["candidates"][0]["content"]["parts"][0]["text"]
     except Exception as e:
-
         return f"Gemini Error: {e}"
 
 
 # =====================================================
-# COUNTRY CODES
+# COUNTRY (fixed)
 # =====================================================
 
 country_codes = {
-
     "sudan": "SD",
     "السودان": "SD",
-
     "egypt": "EG",
     "مصر": "EG",
-
     "kenya": "KE",
     "كينيا": "KE",
-
     "ethiopia": "ET",
     "إثيوبيا": "ET",
-
     "saudi arabia": "SA",
     "السعودية": "SA",
-
-    "india": "IN",
-    "الهند": "IN",
-
     "usa": "US",
     "united states": "US",
     "america": "US",
     "امريكا": "US",
-
     "canada": "CA",
     "كندا": "CA"
 }
 
 
 # =====================================================
-# SPECIES FILTERS
+# TAXON FILTERS (FIXED - IMPORTANT)
 # =====================================================
 
 species_filters = {
-
-    "Birds": "class=Aves&rank=SPECIES",
-
-    "Plants": "kingdom=Plantae&rank=SPECIES",
-
-    "Mammals": "class=Mammalia&rank=SPECIES",
-
-    "Reptiles": "class=Reptilia&rank=SPECIES",
-
-    "Fish": "class=Actinopterygii&rank=SPECIES",
-
-    "Insects": "class=Insecta&rank=SPECIES"
+    "Birds": "taxon_key=212",
+    "Plants": "taxon_key=6",
+    "Mammals": "taxon_key=359",
+    "Reptiles": "taxon_key=358",
+    "Fish": "taxon_key=204",
+    "Insects": "taxon_key=216"
 }
 
 
 # =====================================================
-# GET SPECIES
+# GET SPECIES (CLEAN + FIXED GBIF LOGIC)
 # =====================================================
 
 def get_species(country_code, filter_value):
 
     url = (
-        f"https://api.gbif.org/v1/occurrence/search?"
+        "https://api.gbif.org/v1/occurrence/search?"
         f"country={country_code}"
         f"&{filter_value}"
-        f"&limit=200"
-        f"&has_coordinate=true"
-        f"&has_geospatial_issue=false"
-        f"&occurrence_status=PRESENT"
+        "&limit=200"
+        "&has_coordinate=true"
+        "&has_geospatial_issue=false"
+        "&occurrence_status=PRESENT"
     )
 
     try:
-
-        response = requests.get(url)
-
-        data = response.json()
-
+        r = requests.get(url)
+        data = r.json()
         results = data.get("results", [])
 
         species = []
 
         bad_words = [
-            "unclassified",
-            "sp.",
-            "cf.",
-            "aff.",
-            "unknown",
-            "incertae",
-            "birds",
-            "plants"
+            "unclassified", "sp.", "cf.", "aff.",
+            "unknown", "incertae", "plant", "bird"
         ]
 
         for item in results:
 
             sp = item.get("species")
+            kingdom = item.get("kingdom")
 
+            # FILTER 1: must exist
             if not sp:
                 continue
 
-            sp_lower = sp.lower()
-
-            # تنظيف البيانات
-            if any(word in sp_lower for word in bad_words):
+            # FILTER 2: remove junk names
+            sp_low = sp.lower()
+            if any(w in sp_low for w in bad_words):
                 continue
 
-            # تجاهل الأسماء القصيرة
+            # FILTER 3: must be real species name
             if len(sp.split()) < 2:
                 continue
 
-            # منع التكرار
+            # FILTER 4: strict bird check
+            if species_type == "Birds" and kingdom != "Animalia":
+                continue
+
             if sp not in species:
                 species.append(sp)
 
         return species[:10]
 
     except Exception as e:
-
         st.error(f"GBIF Error: {e}")
-
         return []
 
 
 # =====================================================
-# GET LINKS
+# LINKS (kept clean)
 # =====================================================
 
 def get_species_links(species_name, country):
 
     links = []
 
-    clean_name = " ".join(species_name.split()[:2])
-
-    encoded_species = quote(clean_name)
-
-    encoded_country = quote(country)
+    clean = " ".join(species_name.split()[:2])
+    q = quote(clean)
+    c = quote(country)
 
-    # -------------------------------------------------
     # GBIF
-    # -------------------------------------------------
-
-    gbif_api = (
-        f"https://api.gbif.org/v1/species/match?"
-        f"name={encoded_species}"
-    )
-
     try:
-
-        r = requests.get(gbif_api).json()
+        r = requests.get(
+            f"https://api.gbif.org/v1/species/match?name={q}"
+        ).json()
 
         if "usageKey" in r:
-
-            gbif_link = (
-                f"https://www.gbif.org/species/"
-                f"{r['usageKey']}"
-            )
-
             links.append(
-                f"- [GBIF Species Page]({gbif_link})"
+                f"- [GBIF](https://www.gbif.org/species/{r['usageKey']})"
             )
-
     except:
         pass
 
-    # -------------------------------------------------
     # Wikipedia
-    # -------------------------------------------------
-
-    wiki_name = clean_name.replace(" ", "_")
-
-    wiki_url = (
-        f"https://en.wikipedia.org/wiki/{wiki_name}"
-    )
-
-    links.append(
-        f"- [Wikipedia Article]({wiki_url})"
-    )
-
-    # -------------------------------------------------
-    # Google Scholar
-    # -------------------------------------------------
-
-    scholar_url = (
-        f"https://scholar.google.com/scholar?q="
-        f"{encoded_species}+{encoded_country}"
-    )
-
     links.append(
-        f"- [Google Scholar ({country})]({scholar_url})"
+        f"- [Wikipedia](https://en.wikipedia.org/wiki/{clean.replace(' ','_')})"
     )
 
-    # -------------------------------------------------
     # iNaturalist
-    # -------------------------------------------------
-
-    inaturalist = (
-        f"https://www.inaturalist.org/search?q="
-        f"{encoded_species}"
-    )
-
-    links.append(
-        f"- [iNaturalist Observations]({inaturalist})"
-    )
-
-    # -------------------------------------------------
-    # Encyclopedia of Life
-    # -------------------------------------------------
-
-    eol = (
-        f"https://eol.org/search?q="
-        f"{encoded_species}"
-    )
-
-    links.append(
-        f"- [Encyclopedia of Life]({eol})"
-    )
-
-    # -------------------------------------------------
-    # Biodiversity Heritage Library
-    # -------------------------------------------------
-
-    bhl = (
-        f"https://www.biodiversitylibrary.org/search?"
-        f"searchTerm={encoded_species}"
-    )
-
-    links.append(
-        f"- [Biodiversity Heritage Library]({bhl})"
-    )
-
-    # -------------------------------------------------
-    # Google Search
-    # -------------------------------------------------
-
-    google_search = (
-        f"https://www.google.com/search?q="
-        f"{encoded_species}+in+{encoded_country}"
-    )
-
-    links.append(
-        f"- [Google Search ({country})]({google_search})"
-    )
-
-    # -------------------------------------------------
-    # Images Search
-    # -------------------------------------------------
-
-    image_search = (
-        f"https://www.google.com/search?"
-        f"tbm=isch&q={encoded_species}+{encoded_country}"
-    )
-
     links.append(
-        f"- [Images Search]({image_search})"
-    )
-
-    # -------------------------------------------------
-    # Research Papers
-    # -------------------------------------------------
-
-    research = (
-        f"https://www.google.com/search?q="
-        f"{encoded_species}+{encoded_country}+research+paper"
+        f"- [iNaturalist](https://www.inaturalist.org/search?q={q})"
     )
 
+    # Scholar
     links.append(
-        f"- [Research Papers]({research})"
-    )
-
-    # -------------------------------------------------
-    # Conservation
-    # -------------------------------------------------
-
-    conservation = (
-        f"https://www.google.com/search?q="
-        f"{encoded_species}+{encoded_country}+conservation"
-    )
-
-    links.append(
-        f"- [Conservation Information]({conservation})"
+        f"- [Google Scholar](https://scholar.google.com/scholar?q={q}+{c})"
     )
 
     return links
 
 
 # =====================================================
-# USER INPUTS
+# UI
 # =====================================================
 
-country = st.text_input(
-    "Enter Country"
-)
+country = st.text_input("Enter Country")
 
 species_type = st.selectbox(
     "Select Species Type",
-    [
-        "Birds",
-        "Plants",
-        "Mammals",
-        "Reptiles",
-        "Fish",
-        "Insects"
-    ]
+    ["Birds", "Plants", "Mammals", "Reptiles", "Fish", "Insects"]
 )
 
 # =====================================================
-# SEARCH BUTTON
+# MAIN
 # =====================================================
 
 if st.button("Search Biodiversity"):
 
-    # -------------------------------------------------
-    # CLEAN COUNTRY INPUT
-    # -------------------------------------------------
-
+    # normalize country
     country_clean = country.strip().lower()
-
-    country_code = country_codes.get(
-        country_clean
-    )
-
-    # -------------------------------------------------
-    # VALIDATION
-    # -------------------------------------------------
+    country_code = country_codes.get(country_clean)
 
     if not country:
-
-        st.warning(
-            "Please enter a country."
-        )
-
+        st.warning("Enter country")
         st.stop()
 
     if not country_code:
-
-        st.error(
-            f"Country '{country}' not supported yet."
-        )
-
+        st.error(f"Country '{country}' not supported")
         st.stop()
 
-    filter_value = species_filters.get(
-        species_type
-    )
+    filter_value = species_filters[species_type]
 
-    # -------------------------------------------------
-    # FETCH SPECIES
-    # -------------------------------------------------
-
-    with st.spinner(
-        "Fetching biodiversity data..."
-    ):
-
-        species_list = get_species(
-            country_code,
-            filter_value
-        )
+    with st.spinner("Fetching GBIF data..."):
+        species_list = get_species(country_code, filter_value)
 
     if not species_list:
-
-        st.warning(
-            "No clean biodiversity data found."
-        )
-
+        st.warning("No clean species found")
         st.stop()
 
-    # -------------------------------------------------
-    # DISPLAY SPECIES
-    # -------------------------------------------------
-
     st.subheader("🐾 Species Found")
 
     for sp in species_list:
+        st.write("•", sp)
 
-        st.write(f"• {sp}")
-
-    # -------------------------------------------------
-    # AI ANALYSIS
-    # -------------------------------------------------
-
-    species_text = "\n".join(species_list)
+    # =================================================
+    # AI
+    # =================================================
 
     prompt = f"""
 You are a biodiversity scientist.
 
-Country:
-{country}
-
-Species Type:
-{species_type}
-
-Species Found:
-{species_text}
+Country: {country}
+Type: {species_type}
 
-Write:
-1. Biodiversity summary
-2. Ecological importance
-3. Conservation challenges
-4. GIS/environment insights
+Species:
+{chr(10).join(species_list)}
 
-IMPORTANT:
-Only discuss species actually found in the country.
-Avoid hallucinations.
+Only analyze real species.
 """
 
-    with st.spinner(
-        "Generating AI analysis..."
-    ):
+    st.subheader("🤖 AI Analysis")
 
-        ai_text = call_gemini(
-            prompt,
-            os.getenv("GEMINI_API_KEY")
-        )
-
-    st.subheader(
-        "🤖 AI Biodiversity Analysis"
+    st.write(
+        call_gemini(prompt, os.getenv("GEMINI_API_KEY"))
     )
 
-    st.write(ai_text)
-
     # =================================================
-    # LINKS SECTION
+    # LINKS
     # =================================================
 
-    st.subheader(
-        "🔗 Scientific & Country Related Links"
-    )
-
-    with st.spinner(
-        "Generating links..."
-    ):
+    st.subheader("🔗 Links")
 
-        for sp in species_list:
-
-            st.markdown(f"## 🔬 {sp}")
-
-            links = get_species_links(
-                sp,
-                country
-            )
-
-            for link in links:
-
-                st.markdown(link)
-
-            st.markdown("---")
+    for sp in species_list:
+        st.markdown(f"### {sp}")
+        for link in get_species_links(sp, country):
+            st.markdown(link)
+        st.markdown("---")