src/pages/data_entry.py

"""
pages/data_entry.py
────────────────────────────────────────────────────────────────────
SPJIMR Waste Analytics — Daily Data Entry Form

Replaces the Excel upload workflow entirely.
Staff enter wet & dry waste kg per block directly in this page.
Data is persisted in SQLite (data/waste_log.db).

After saving, the shared session state key "waste_all_df" is
refreshed so the Analytics and Gamification pages immediately
reflect the new data — no re-upload needed.

Add to app.py:
    from pages.data_entry import render_data_entry
    # in your navigation block:
    render_data_entry()
"""

from __future__ import annotations
import io
import logging
from datetime import date, timedelta

import pandas as pd
import plotly.graph_objects as go
import streamlit as st

from core.waste_db import WasteDB
from core.waste_parser import LOCATIONS, LOCATION_LABELS, LOCATION_GROUPS

logger = logging.getLogger(__name__)

# ── Session state key shared across all pages ──────────────────────────────────
_ALL_DF_KEY = "waste_all_df"
_DB_KEY     = "waste_db_instance"

# ── Block display order — group them visually ──────────────────────────────────
DISPLAY_ORDER = [
    # Academic
    "A&B Block", "C&D Block",
    # Hostels
    "L H Hostel", "Hostel no -25", "Hostel no -26", "Hostel no -27",
    "Hostel no -28", "Hostel no -29", "Hostel no -30",
    # Dining
    "cantean", "MESS",
]

GROUP_ICONS = {"Academic": "🏫", "Hostels": "🏠", "Dining": "🍽️"}

def _loc_group(loc: str) -> str:
    return next((g for g, locs in LOCATION_GROUPS.items() if loc in locs), "Other")


# ── CSS ────────────────────────────────────────────────────────────────────────
_CSS = """
<style>
@import url('https://fonts.googleapis.com/css2?family=DM+Sans:wght@300;400;500;700&family=Space+Grotesk:wght@400;600;700;800&display=swap');

[data-testid="stAppViewContainer"] { background: #0B1622; }
[data-testid="stSidebar"]          { background: #07111D; }
h1,h2,h3,h4 { font-family:'Space Grotesk',sans-serif !important; }
p,div,span,label,input { font-family:'DM Sans',sans-serif !important; }

/* ── Block input card ── */
.block-card {
    background: linear-gradient(145deg,#111E2E,#172840);
    border: 1px solid rgba(0,201,167,0.15);
    border-radius: 14px;
    padding: 16px 18px 14px;
    margin-bottom: 12px;
    transition: border-color 0.2s, box-shadow 0.2s;
}
.block-card:hover {
    border-color: rgba(0,201,167,0.4);
    box-shadow: 0 4px 20px rgba(0,201,167,0.08);
}
.block-card.filled {
    border-color: rgba(0,201,167,0.35);
    background: linear-gradient(145deg,#0E2220,#14352E);
}
.block-card.warning {
    border-color: rgba(245,166,35,0.4);
    background: linear-gradient(145deg,#1E1800,#2A2000);
}

.block-title {
    font-family:'Space Grotesk',sans-serif;
    font-size: 0.95rem; font-weight: 700;
    color: #E8F4F8; margin-bottom: 2px;
}
.block-group {
    font-size: 0.72rem; color: #7A9BB5;
    text-transform: uppercase; letter-spacing: 0.08em;
    margin-bottom: 10px;
}
.block-total {
    font-family:'Space Grotesk',sans-serif;
    font-size: 1.05rem; font-weight: 700; color: #00C9A7;
    text-align: right; margin-top: 6px;
}

/* ── Section headers ── */
.section-hdr {
    font-family:'Space Grotesk',sans-serif;
    font-size: 1.1rem; font-weight: 700; color: #E8F4F8;
    border-left: 3px solid #00C9A7; padding-left: 12px;
    margin: 28px 0 14px;
}
.group-hdr {
    font-family:'Space Grotesk',sans-serif;
    font-size: 0.82rem; font-weight: 600; color: #7A9BB5;
    text-transform: uppercase; letter-spacing: 0.12em;
    margin: 20px 0 8px; padding-bottom: 4px;
    border-bottom: 1px solid rgba(255,255,255,0.06);
}

/* ── Summary bar ── */
.summary-bar {
    background: linear-gradient(135deg,#0D2B3E,#112840);
    border: 1px solid rgba(0,201,167,0.25);
    border-radius: 12px; padding: 16px 24px;
    display: flex; justify-content: space-between;
    align-items: center; margin-bottom: 20px;
}
.summary-val {
    font-family:'Space Grotesk',sans-serif;
    font-size: 1.6rem; font-weight: 800; color: #00C9A7;
}
.summary-lbl { font-size: 0.76rem; color: #7A9BB5; text-transform: uppercase; letter-spacing: 0.1em; }

/* ── Status chips ── */
.chip {
    display: inline-block; padding: 3px 10px; border-radius: 20px;
    font-size: 0.74rem; font-weight: 600; margin: 2px;
}
.chip-ok   { background: rgba(0,201,167,0.15); color: #00C9A7; border: 1px solid rgba(0,201,167,0.3); }
.chip-warn { background: rgba(245,166,35,0.15); color: #F5A623; border: 1px solid rgba(245,166,35,0.3); }
.chip-err  { background: rgba(255,107,107,0.15); color: #FF6B6B; border: 1px solid rgba(255,107,107,0.3); }

/* ── Calendar completion dots ── */
.cal-cell {
    display: inline-block; width: 28px; height: 28px;
    border-radius: 6px; line-height: 28px; text-align: center;
    font-size: 0.72rem; font-weight: 600; margin: 2px;
}
.cal-full    { background: #00C9A7; color: #0B1622; }
.cal-partial { background: #F5A623; color: #0B1622; }
.cal-empty   { background: rgba(255,255,255,0.06); color: #7A9BB5; }

/* ── History table rows ── */
.hist-row {
    display: flex; align-items: center; justify-content: space-between;
    background: rgba(17,30,46,0.9); border: 1px solid rgba(0,201,167,0.1);
    border-radius: 10px; padding: 12px 18px; margin-bottom: 8px;
}
.hist-date { font-family:'Space Grotesk',sans-serif; font-size:0.9rem; font-weight:700; color:#E8F4F8; width:110px; }
.hist-stat { font-size:0.82rem; color:#7A9BB5; }
.hist-total { font-family:'Space Grotesk',sans-serif; font-size:1rem; font-weight:700; color:#00C9A7; width:100px; text-align:right; }
</style>
"""

CHART_LAYOUT = dict(
    paper_bgcolor="rgba(0,0,0,0)", plot_bgcolor="rgba(0,0,0,0)",
    font=dict(family="DM Sans", color="#E8F4F8", size=12),
    margin=dict(l=10, r=10, t=36, b=10),
    legend=dict(bgcolor="rgba(17,30,46,0.8)", bordercolor="rgba(0,201,167,0.2)", borderwidth=1),
    xaxis=dict(gridcolor="rgba(255,255,255,0.05)", linecolor="rgba(255,255,255,0.08)"),
    yaxis=dict(gridcolor="rgba(255,255,255,0.05)", linecolor="rgba(255,255,255,0.08)"),
)


# ── DB singleton via session state ─────────────────────────────────────────────
def _get_db() -> WasteDB:
    if _DB_KEY not in st.session_state:
        st.session_state[_DB_KEY] = WasteDB()
    return st.session_state[_DB_KEY]


def _refresh_session_df(db: WasteDB) -> None:
    """Rebuild the shared DataFrame so analytics/gamification pages update instantly."""
    df = db.to_dataframe()
    st.session_state[_ALL_DF_KEY] = df if not df.empty else None


# ── Helpers ────────────────────────────────────────────────────────────────────
def _group_header(group: str) -> str:
    icon = GROUP_ICONS.get(group, "📦")
    return f'<div class="group-hdr">{icon} {group}</div>'


def _completion_chip(pct: float) -> str:
    if pct == 100:
        return '<span class="chip chip-ok">✓ Complete</span>'
    elif pct > 0:
        return f'<span class="chip chip-warn">⚡ {pct:.0f}%</span>'
    return '<span class="chip chip-err">○ No data</span>'


# ══════════════════════════════════════════════════════════════════════════════
# Tab 1 — Daily Entry Form
# ══════════════════════════════════════════════════════════════════════════════

def _render_entry_form(db: WasteDB) -> None:
    st.markdown('<div class="section-hdr">📅 Select Date</div>', unsafe_allow_html=True)

    col_date, col_nav = st.columns([2, 3])
    with col_date:
        selected_date = st.date_input(
            "Entry Date",
            value=date.today(),
            max_value=date.today(),
            key="entry_date_picker",
            label_visibility="collapsed",
        )
    with col_nav:
        recorded = db.all_dates()
        if recorded:
            b1, b2, b3 = st.columns(3)
            with b1:
                if st.button("⬅ Previous day", use_container_width=True, key="nav_prev"):
                    idx = recorded.index(selected_date) if selected_date in recorded else -1
                    if idx > 0:
                        st.session_state["entry_date_picker"] = recorded[idx - 1]
                        st.rerun()
            with b2:
                if st.button("Today", use_container_width=True, key="nav_today"):
                    st.session_state["entry_date_picker"] = date.today()
                    st.rerun()
            with b3:
                if st.button("Next day ➡", use_container_width=True, key="nav_next"):
                    idx = recorded.index(selected_date) if selected_date in recorded else -1
                    if idx >= 0 and idx < len(recorded) - 1:
                        st.session_state["entry_date_picker"] = recorded[idx + 1]
                        st.rerun()

    # Load existing values for this date
    existing = db.get_day(selected_date)
    comp     = db.date_completion(selected_date)

    # Status summary bar
    status_chip = _completion_chip(comp["pct"])
    st.markdown(
        f"""<div class="summary-bar">
            <div>
                <div class="summary-lbl">Selected Date</div>
                <div class="summary-val">{selected_date.strftime("%d %b %Y")}</div>
            </div>
            <div style="text-align:center">
                <div class="summary-lbl">Blocks filled</div>
                <div class="summary-val">{comp["filled"]} / {comp["total"]}</div>
            </div>
            <div style="text-align:right">
                {status_chip}
            </div>
        </div>""",
        unsafe_allow_html=True,
    )

    st.markdown('<div class="section-hdr">⚖️ Enter Waste Data (kg)</div>', unsafe_allow_html=True)
    st.caption("Enter 0 for blocks with no data collected that day.")

    # ── Build form inputs grouped by category ─────────────────────────────────
    form_values: dict[str, dict[str, float]] = {}
    prev_group = None

    # Use a single st.form so all 11 blocks save atomically
    with st.form(key=f"waste_entry_{selected_date}", clear_on_submit=False):
        for loc in DISPLAY_ORDER:
            group = _loc_group(loc)
            label = LOCATION_LABELS.get(loc, loc)
            ex    = existing.get(loc, {"wet": 0.0, "dry": 0.0})

            # Group separator header
            if group != prev_group:
                st.markdown(_group_header(group), unsafe_allow_html=True)
                prev_group = group

            # Card header with location name
            is_filled  = ex["wet"] > 0 or ex["dry"] > 0
            card_class = "block-card filled" if is_filled else "block-card"
            last_edit  = f"Last saved: {ex['updated_at'][:16].replace('T',' ')}" if ex.get("updated_at") else ""
            st.markdown(
                f"""<div class="{card_class}">
                    <div class="block-title">{label}</div>
                    <div class="block-group">{group} &nbsp;·&nbsp; {last_edit}</div>
                </div>""",
                unsafe_allow_html=True,
            )

            c1, c2 = st.columns(2)
            with c1:
                wet = st.number_input(
                    f"🟢 Wet waste (kg) — {label}",
                    min_value=0.0, max_value=5000.0,
                    value=float(ex["wet"]),
                    step=0.5, format="%.1f",
                    key=f"wet_{loc}",
                    label_visibility="visible",
                )
            with c2:
                dry = st.number_input(
                    f"🟡 Dry waste (kg) — {label}",
                    min_value=0.0, max_value=5000.0,
                    value=float(ex["dry"]),
                    step=0.5, format="%.1f",
                    key=f"dry_{loc}",
                    label_visibility="visible",
                )

            total_now = wet + dry
            if total_now > 0:
                st.markdown(
                    f'<div class="block-total">Total: {total_now:.1f} kg</div>',
                    unsafe_allow_html=True,
                )

            form_values[loc] = {"wet": wet, "dry": dry}

        st.markdown("---")

        # Live preview inside form
        grand_wet   = sum(v["wet"] for v in form_values.values())
        grand_dry   = sum(v["dry"] for v in form_values.values())
        grand_total = grand_wet + grand_dry

        pcol1, pcol2, pcol3, pcol4 = st.columns(4)
        pcol1.metric("Total Wet",   f"{grand_wet:.1f} kg")
        pcol2.metric("Total Dry",   f"{grand_dry:.1f} kg")
        pcol3.metric("Grand Total", f"{grand_total:.1f} kg")
        pcol4.metric("Blocks with data", sum(1 for v in form_values.values() if v["wet"]+v["dry"] > 0))

        submitted = st.form_submit_button(
            "💾  Save Entry",
            use_container_width=True,
            type="primary",
        )

    if submitted:
        with st.spinner("Saving…"):
            db.upsert_day(selected_date, form_values)
            _refresh_session_df(db)
        filled = sum(1 for v in form_values.values() if v["wet"] + v["dry"] > 0)
        st.success(f"✅ Saved {filled} block(s) for **{selected_date.strftime('%d %b %Y')}**. "
                   f"Analytics and Gamification pages are now updated.")
        st.rerun()


# ══════════════════════════════════════════════════════════════════════════════
# Tab 2 — History & Edit
# ══════════════════════════════════════════════════════════════════════════════

def _render_history(db: WasteDB) -> None:
    recorded = db.all_dates()
    if not recorded:
        st.info("No entries yet. Use the **Enter Data** tab to add your first record.")
        return

    st.markdown('<div class="section-hdr">📋 Recorded Days</div>', unsafe_allow_html=True)

    # Completion calendar
    df_all = db.to_dataframe()
    if not df_all.empty:
        months = sorted(df_all["month"].unique(),
                        key=lambda m: pd.to_datetime("01 " + m, format="%d %b %Y"))
        sel_cal_month = st.selectbox("Month", months, index=len(months)-1, key="hist_cal_month")
        mdf = df_all[df_all["month"] == sel_cal_month]

        st.markdown("**Completeness calendar** (🟢 full · 🟡 partial · ⬛ missing):")
        cal_dates = sorted(mdf["date"].dt.date.unique())
        dt_comp   = {d: db.date_completion(d) for d in cal_dates}

        cells_html = ""
        for d in cal_dates:
            c = dt_comp[d]
            if c["pct"] == 100:
                css = "cal-full"
            elif c["pct"] > 0:
                css = "cal-partial"
            else:
                css = "cal-empty"
            cells_html += f'<span class="cal-cell {css}" title="{d}: {c["filled"]}/{c["total"]}">{d.day}</span>'
        st.markdown(cells_html, unsafe_allow_html=True)

    st.markdown("---")
    st.markdown('<div class="section-hdr">🗂️ All Entries</div>', unsafe_allow_html=True)

    # Search / filter
    fc1, fc2 = st.columns([2, 3])
    with fc1:
        filter_loc = st.selectbox(
            "Filter by block",
            ["All blocks"] + [LOCATION_LABELS[l] for l in DISPLAY_ORDER],
            key="hist_filter_loc",
        )
    with fc2:
        sort_by = st.radio(
            "Sort by",
            ["Date (newest first)", "Date (oldest first)", "Total waste (highest)", "Total waste (lowest)"],
            horizontal=True, key="hist_sort",
        )

    if df_all.empty:
        st.info("No data recorded yet.")
        return

    disp = df_all.copy()
    if filter_loc != "All blocks":
        disp = disp[disp["label"] == filter_loc]

    sort_map = {
        "Date (newest first)":     ("date", False),
        "Date (oldest first)":     ("date", True),
        "Total waste (highest)":   ("total_kg", False),
        "Total waste (lowest)":    ("total_kg", True),
    }
    col, asc = sort_map[sort_by]
    disp = disp.sort_values(col, ascending=asc)

    # Render as styled rows grouped by date
    grouped = disp.groupby("date")
    for dt, grp in grouped:
        day_str   = pd.Timestamp(dt).strftime("%d %b %Y (%A)")
        day_total = grp["total_kg"].sum()
        filled    = (grp["total_kg"] > 0).sum()
        chip      = _completion_chip(filled / len(LOCATIONS) * 100)

        with st.expander(f"📅  {day_str}  —  {day_total:,.1f} kg total  {chip}", expanded=False):
            # Mini table
            t_df = grp[["label","group","wet_kg","dry_kg","total_kg"]].rename(columns={
                "label":"Block","group":"Category",
                "wet_kg":"Wet (kg)","dry_kg":"Dry (kg)","total_kg":"Total (kg)"
            })
            st.dataframe(t_df, use_container_width=True, hide_index=True)

            # Quick bar chart for this day
            fig = go.Figure()
            fig.add_bar(name="Wet", x=grp["label"], y=grp["wet_kg"],
                        marker_color="#2E9E6B", text=grp["wet_kg"].round(1), textposition="inside")
            fig.add_bar(name="Dry", x=grp["label"], y=grp["dry_kg"],
                        marker_color="#F5A623", text=grp["dry_kg"].round(1), textposition="inside")
            fig.update_layout(barmode="stack", height=260,
                              title=f"Waste breakdown — {day_str}", **CHART_LAYOUT)
            fig.update_xaxes(tickangle=-30)
            st.plotly_chart(fig, use_container_width=True)

            # Delete this day
            if st.button(f"🗑️ Delete all entries for {pd.Timestamp(dt).strftime('%d %b %Y')}",
                         key=f"del_day_{dt}", type="secondary"):
                deleted = db.delete_day(pd.Timestamp(dt).date())
                _refresh_session_df(db)
                st.warning(f"Deleted {deleted} record(s) for {pd.Timestamp(dt).strftime('%d %b %Y')}.")
                st.rerun()


# ══════════════════════════════════════════════════════════════════════════════
# Tab 3 — Database Health & Export
# ══════════════════════════════════════════════════════════════════════════════

def _render_db_health(db: WasteDB) -> None:
    st.markdown('<div class="section-hdr">🗄️ Database Status</div>', unsafe_allow_html=True)

    row_count = db.row_count()
    all_dates = db.all_dates()
    months_df = db.monthly_completion()

    mc1, mc2, mc3 = st.columns(3)
    mc1.metric("Total Records",  f"{row_count:,}")
    mc2.metric("Days Recorded",  f"{len(all_dates)}")
    mc3.metric("Months Covered", f"{len(months_df)}" if not months_df.empty else "0")

    if not months_df.empty:
        st.markdown('<div class="section-hdr">📊 Monthly Completeness</div>', unsafe_allow_html=True)

        fig_comp = go.Figure(go.Bar(
            x=months_df["month"],
            y=months_df["completeness"],
            marker=dict(
                color=months_df["completeness"],
                colorscale=[[0,"#FF6B6B"],[0.5,"#F5A623"],[1.0,"#00C9A7"]],
                showscale=False,
            ),
            text=months_df["completeness"].apply(lambda x: f"{x:.0f}%"),
            textposition="outside",
        ))
        fig_comp.add_hline(y=100, line_dash="dot", line_color="rgba(0,201,167,0.4)",
                            annotation_text="100% target")
        fig_comp.update_layout(
            title="Data Completeness by Month (%)",
            **CHART_LAYOUT,
        )
        # yaxis=dict(range=[0, 115], title="%"),
        fig_comp.update_yaxes(range=[0, 115], title="%")
        st.plotly_chart(fig_comp, use_container_width=True)

        st.dataframe(
            months_df.rename(columns={
                "month":"Month","days":"Days with data",
                "location_days":"Block-days recorded","completeness":"Completeness %"
            }),
            use_container_width=True, hide_index=True,
        )

    st.markdown('<div class="section-hdr">📥 Export Data</div>', unsafe_allow_html=True)

    ec1, ec2 = st.columns(2)
    with ec1:
        if st.button("📊 Refresh Analytics & Gamification", use_container_width=True, type="primary"):
            _refresh_session_df(db)
            st.success("✅ Analytics and Gamification pages updated with latest data.")

    with ec2:
        df_export = db.to_dataframe()
        if not df_export.empty:
            csv_bytes = df_export.to_csv(index=False).encode()
            st.download_button(
                label="⬇️ Download Full Dataset (CSV)",
                data=csv_bytes,
                file_name="spjimr_waste_data.csv",
                mime="text/csv",
                use_container_width=True,
            )
        else:
            st.info("No data to export yet.")

    # Optional: import from Excel on this tab as a one-time migration
    st.markdown('<div class="section-hdr">📤 One-time Excel Import</div>', unsafe_allow_html=True)
    st.caption("Already have historical Excel files? Import them once — data is saved to the database and you won't need Excel files again.")

    uploaded_files = st.file_uploader(
        "Upload historical Excel files (one per month)",
        type=["xlsx"], accept_multiple_files=True, key="db_import_xlsx",
    )
    if uploaded_files and st.button("⬆️ Import into Database", type="primary", key="do_import"):
        from core.waste_parser import parse_waste_excel
        total_imported = 0
        for f in uploaded_files:
            try:
                df_xl = parse_waste_excel(io.BytesIO(f.read()))
                if df_xl.empty:
                    st.warning(f"⚠️ No data found in {f.name}")
                    continue
                for _, row in df_xl.iterrows():
                    db.upsert(
                        row["date"].date(),
                        row["location"],
                        row["wet_kg"],
                        row["dry_kg"],
                    )
                total_imported += len(df_xl)
                st.success(f"✅ Imported {len(df_xl)} records from **{f.name}**")
            except Exception as exc:
                st.error(f"❌ Failed to import {f.name}: {exc}")
        if total_imported:
            _refresh_session_df(db)
            st.success(f"🎉 Total {total_imported} records imported. Analytics pages are now live.")
            st.rerun()


# ══════════════════════════════════════════════════════════════════════════════
# Main entry point
# ══════════════════════════════════════════════════════════════════════════════

def render_data_entry() -> None:
    st.markdown(_CSS, unsafe_allow_html=True)
    st.markdown("## 📝 Daily Waste Data Entry")
    st.markdown(
        "<p style='color:#7A9BB5;margin-top:-10px;'>"
        "Enter daily block-wise waste directly — no Excel file needed</p>",
        unsafe_allow_html=True,
    )

    db = _get_db()

    # Auto-refresh shared DataFrame on page load so analytics stay in sync
    if st.session_state.get(_ALL_DF_KEY) is None and db.row_count() > 0:
        _refresh_session_df(db)

    # Quick status banner at the top
    row_count = db.row_count()
    all_dates = db.all_dates()
    if row_count > 0:
        last_date = max(all_dates).strftime("%d %b %Y")
        st.markdown(
            f'<div style="background:rgba(0,201,167,0.08);border:1px solid rgba(0,201,167,0.2);'
            f'border-radius:10px;padding:10px 18px;margin-bottom:16px;font-size:0.88rem;color:#7A9BB5;">'
            f'🗄️ Database contains <strong style="color:#00C9A7">{row_count:,} records</strong> across '
            f'<strong style="color:#00C9A7">{len(all_dates)} days</strong>. '
            f'Last entry: <strong style="color:#E8F4F8">{last_date}</strong></div>',
            unsafe_allow_html=True,
        )

    tab1, tab2, tab3 = st.tabs(["✏️ Enter Data", "📋 History & Edit", "🗄️ Database & Export"])

    with tab1:
        _render_entry_form(db)

    with tab2:
        _render_history(db)

    with tab3:
        _render_db_health(db)