import streamlit as st
import pandas as pd
from api_logic import fetch_keyword_ideas, rank_and_filter_keywords, KeywordRequestModel


# -----------------------------
# Display functions
# -----------------------------
def display_keyword_results(data: dict):
    """
    Display keyword ideas returned from fetch_keyword_ideas()

    Parameters
    ----------
    data : dict
        {
            "results": [...],
            "aggregateMetricResults": {...}
        }
    """

    # When no results are found
    if not data or not data.get("results"):
        st.warning("No keyword ideas found.")
        return

    results = data["results"]

    st.subheader("📌 Keyword Ideas Overview")

    # ------- 1 Display main metrics as a table -------
    df_main = pd.DataFrame([
        {
            "Keyword": r["keyword"],
            "Avg Monthly Searches": r["avg_monthly_searches"],
            "Competition": r["competition"],
            "Competition Index": r["competition_index"],
            "Low Bid (micros)": r["low_top_of_page_bid_micros"],
            "High Bid (micros)": r["high_top_of_page_bid_micros"],
            #"Avg CPC (micros)": r["average_cpc_micros"],
            ## Added Features
            "Log CPC": r["log_cpc"],
            "Normalized CPC": r["log_cpc_norm_0_100"],
            
        }
        for r in results
    ])

    st.dataframe(df_main, use_container_width=True)

    # ------- 2 Monthly search volumes (inside expanders) -------
    st.subheader("📅 Monthly Search Volumes")

    for r in results:
        with st.expander(f"{r['keyword']} - Monthly Trends"):
            mv = r.get("monthly_search_volumes", [])

            if mv:
                df_mv = pd.DataFrame(mv)
                df_mv = df_mv.rename(columns={
                    "month": "Month",
                    "year": "Year",
                    "monthly_searches": "Monthly Searches"
                })
                st.dataframe(df_mv, use_container_width=True)
            else:
                st.write("No monthly search volume data available.")


    # ------- 3 Aggregate metrics (device-based searches) -------
    st.subheader("📱 Aggregate Metric Results")

    agg = data.get("aggregateMetricResults", {})
    device_searches = agg.get("device_searches", [])

    if device_searches:
        df_device = pd.DataFrame(device_searches)
        st.dataframe(df_device, use_container_width=True)
    else:
        st.write("No aggregate device metrics.")


# -----------------------------
# Streamlit UI
# -----------------------------
def main():

    st.title("🔍 Google Ads Keyword Ideas ToolV007")
    with st.expander("🛠️ Keyword Planner Demo – Development Progress  (click to view)"):
        st.markdown("""
1. Obtain Google Ads API authentication token

2. Implement `fetch_keyword_ideas` in `api_logic.py`  
   This function communicates with the Google Ads API and retrieves keyword ideas.

3. Implement methodology to sort data    **⚠️(currently in progress)⚠️** requires 1 day more. 
   - Methodology is taken from **["Rule Book For GKP With Code"]**   
     3.1 Volume Search Section  
     3.2 Competition Score Section   
     3.3 CPC Score Section **(Completed)**   
     3.4 Intent Section  
     3.5 Trend Section 

4. Implement `app.py`  
   Integrate the results into the Streamlit UI and deploy  
                    **Hugging Face Link will be shared soon ** 
                        
                """)   

    st.markdown("Enter keyword(s) and your Google Ads settings to fetch keyword ideas.")

    # Input UI
    customer_id = st.text_input("Customer ID", value="")
    keywords = st.text_input("Keywords (comma separated)", value="")
    geo_target = st.text_input("Geo Target ID (optional)", value="2840")
    language = st.text_input("Language ID (optional)", value="1000")

    st.write("---")

    if st.button("🚀 Get Keyword Ideas"):

        if not customer_id or not keywords:
            st.error("Please enter Customer ID and keywords.")
            return

        keyword_list = [k.strip() for k in keywords.split(",")]

        # Build KeywordRequestModel
        request_model = KeywordRequestModel(
            customer_id=customer_id,
            keywords=keyword_list,
            geo_target=geo_target if geo_target else None,
            language=language if language else None,
        )

        with st.spinner("Fetching keyword ideas..."):
            response = fetch_keyword_ideas(request_model)

        # Filter top 20 items
        top20 = rank_and_filter_keywords(response, limit=20)

        # Build display-friendly response structure
        filtered_response = {
            "results": top20,
            "aggregateMetricResults": response.get("aggregateMetricResults")
        }

        st.success("✔ Completed")
        display_keyword_results(filtered_response)

# Run application
if __name__ == "__main__":
    main()