TieMeasureFlow/server/services/report_service.py

"""Report generation service — PDF reports via Jinja2 + Plotly/Kaleido + WeasyPrint."""
import base64
from datetime import datetime
from pathlib import Path

import plotly.graph_objects as go
import plotly.io as pio
from jinja2 import Environment, FileSystemLoader
from sqlalchemy import and_, select
from sqlalchemy.ext.asyncio import AsyncSession
from weasyprint import HTML

from config import settings
from models.measurement import Measurement
from models.recipe import Recipe, RecipeVersion
from models.setting import SystemSetting
from models.task import RecipeSubtask, RecipeTask
from models.user import User
from services.spc_service import (
    compute_capability,
    compute_control_chart,
    compute_histogram,
    compute_summary,
)

# Jinja2 environment for report templates
_templates_dir = Path(__file__).parent.parent / "templates" / "reports"
_jinja_env = Environment(
    loader=FileSystemLoader(str(_templates_dir)),
    autoescape=True,
)


async def _query_measurements(
    db: AsyncSession,
    recipe_id: int,
    version_id: int | None = None,
    subtask_id: int | None = None,
    date_from: datetime | None = None,
    date_to: datetime | None = None,
    operator_id: int | None = None,
    lot_number: str | None = None,
    serial_number: str | None = None,
) -> list[Measurement]:
    """Query measurements with common filters (duplicated from statistics router)."""
    filters = []

    if version_id is not None:
        filters.append(Measurement.version_id == version_id)
    else:
        version_ids = select(RecipeVersion.id).where(
            RecipeVersion.recipe_id == recipe_id
        )
        filters.append(Measurement.version_id.in_(version_ids))

    if subtask_id is not None:
        filters.append(Measurement.subtask_id == subtask_id)
    if date_from is not None:
        filters.append(Measurement.measured_at >= date_from)
    if date_to is not None:
        filters.append(Measurement.measured_at <= date_to)
    if operator_id is not None:
        filters.append(Measurement.measured_by == operator_id)
    if lot_number is not None:
        filters.append(Measurement.lot_number == lot_number)
    if serial_number is not None:
        filters.append(Measurement.serial_number == serial_number)

    query = (
        select(Measurement)
        .where(and_(*filters) if filters else True)
        .order_by(Measurement.measured_at.asc())
    )
    result = await db.execute(query)
    return list(result.scalars().all())


async def _get_company_info(db: AsyncSession) -> dict:
    """Read company logo path and company name from system_settings."""
    info = {"logo_base64": None, "company_name": "TieMeasureFlow"}

    # Company name
    result = await db.execute(
        select(SystemSetting).where(SystemSetting.setting_key == "company_name")
    )
    setting = result.scalar_one_or_none()
    if setting:
        info["company_name"] = setting.setting_value

    # Company logo
    result = await db.execute(
        select(SystemSetting).where(SystemSetting.setting_key == "company_logo_path")
    )
    setting = result.scalar_one_or_none()
    if setting and setting.setting_value:
        logo_path = settings.upload_path / setting.setting_value
        if logo_path.exists():
            with open(logo_path, "rb") as f:
                logo_bytes = f.read()
            # Detect mime type
            suffix = logo_path.suffix.lower()
            mime = {".png": "image/png", ".jpg": "image/jpeg", ".jpeg": "image/jpeg", ".svg": "image/svg+xml"}.get(suffix, "image/png")
            info["logo_base64"] = f"data:{mime};base64,{base64.b64encode(logo_bytes).decode()}"

    return info


async def _get_recipe_info(db: AsyncSession, recipe_id: int) -> dict:
    """Get recipe code and name."""
    result = await db.execute(select(Recipe).where(Recipe.id == recipe_id))
    recipe = result.scalar_one_or_none()
    if recipe:
        return {"code": recipe.code, "name": recipe.name, "id": recipe.id}
    return {"code": "???", "name": "Unknown", "id": recipe_id}


async def _get_subtask_info(db: AsyncSession, subtask_id: int) -> dict | None:
    """Get subtask details including tolerances."""
    result = await db.execute(
        select(RecipeSubtask).where(RecipeSubtask.id == subtask_id)
    )
    st = result.scalar_one_or_none()
    if st is None:
        return None
    return {
        "id": st.id,
        "marker_number": st.marker_number,
        "description": st.description,
        "unit": st.unit,
        "nominal": float(st.nominal) if st.nominal is not None else None,
        "utl": float(st.utl) if st.utl is not None else None,
        "uwl": float(st.uwl) if st.uwl is not None else None,
        "lwl": float(st.lwl) if st.lwl is not None else None,
        "ltl": float(st.ltl) if st.ltl is not None else None,
    }


async def _get_operator_map(db: AsyncSession, user_ids: set[int]) -> dict[int, str]:
    """Build a map of user_id -> display name."""
    if not user_ids:
        return {}
    result = await db.execute(select(User).where(User.id.in_(user_ids)))
    users = result.scalars().all()
    return {u.id: u.display_name or u.username for u in users}


def _generate_control_chart_svg(chart_data, subtask_info: dict | None) -> str:
    """Generate control chart SVG using Plotly + Kaleido."""
    if not chart_data.values:
        return ""

    fig = go.Figure()

    # Measurement values
    x_vals = list(range(1, len(chart_data.values) + 1))
    fig.add_trace(go.Scatter(
        x=x_vals, y=chart_data.values,
        mode="lines+markers",
        name="Values",
        line=dict(color="#2563EB", width=2),
        marker=dict(size=5),
    ))

    # Mean line
    fig.add_hline(y=chart_data.mean, line=dict(color="#64748B", dash="dash", width=1.5),
                  annotation_text=f"Mean: {chart_data.mean:.4f}")

    # UCL / LCL
    fig.add_hline(y=chart_data.ucl, line=dict(color="#EF4444", dash="dot", width=1),
                  annotation_text=f"UCL: {chart_data.ucl:.4f}")
    fig.add_hline(y=chart_data.lcl, line=dict(color="#EF4444", dash="dot", width=1),
                  annotation_text=f"LCL: {chart_data.lcl:.4f}")

    # Tolerance limits
    if chart_data.utl is not None:
        fig.add_hline(y=chart_data.utl, line=dict(color="#DC2626", width=1.5),
                      annotation_text=f"UTL: {chart_data.utl}")
    if chart_data.ltl is not None:
        fig.add_hline(y=chart_data.ltl, line=dict(color="#DC2626", width=1.5),
                      annotation_text=f"LTL: {chart_data.ltl}")

    # Out of control points
    if chart_data.out_of_control:
        ooc_x = [x_vals[i] for i in chart_data.out_of_control]
        ooc_y = [chart_data.values[i] for i in chart_data.out_of_control]
        fig.add_trace(go.Scatter(
            x=ooc_x, y=ooc_y,
            mode="markers",
            name="Out of control",
            marker=dict(color="#EF4444", size=10, symbol="x"),
        ))

    fig.update_layout(
        title="Control Chart",
        xaxis_title="Measurement #",
        yaxis_title="Value",
        template="plotly_white",
        width=700, height=350,
        margin=dict(l=60, r=30, t=40, b=40),
        showlegend=False,
    )

    svg_bytes = pio.to_image(fig, format="svg", engine="kaleido")
    return svg_bytes.decode("utf-8")


def _generate_histogram_svg(hist_data, subtask_info: dict | None) -> str:
    """Generate histogram SVG using Plotly + Kaleido."""
    if not hist_data.bins:
        return ""

    fig = go.Figure()

    # Histogram bars
    bin_centers = [(hist_data.bins[i] + hist_data.bins[i + 1]) / 2 for i in range(len(hist_data.counts))]
    bin_width = hist_data.bins[1] - hist_data.bins[0] if len(hist_data.bins) > 1 else 1

    fig.add_trace(go.Bar(
        x=bin_centers, y=hist_data.counts,
        width=bin_width * 0.9,
        marker_color="#2563EB",
        opacity=0.7,
        name="Frequency",
    ))

    # Normal curve overlay
    if hist_data.normal_x and hist_data.normal_y:
        fig.add_trace(go.Scatter(
            x=hist_data.normal_x, y=hist_data.normal_y,
            mode="lines",
            name="Normal curve",
            line=dict(color="#EF4444", width=2),
        ))

    # Tolerance lines
    if subtask_info:
        if subtask_info.get("utl") is not None:
            fig.add_vline(x=subtask_info["utl"], line=dict(color="#DC2626", width=1.5, dash="dash"),
                          annotation_text="UTL")
        if subtask_info.get("ltl") is not None:
            fig.add_vline(x=subtask_info["ltl"], line=dict(color="#DC2626", width=1.5, dash="dash"),
                          annotation_text="LTL")
        if subtask_info.get("nominal") is not None:
            fig.add_vline(x=subtask_info["nominal"], line=dict(color="#22C55E", width=1.5),
                          annotation_text="Nom")

    fig.update_layout(
        title="Histogram",
        xaxis_title="Value",
        yaxis_title="Frequency",
        template="plotly_white",
        width=700, height=350,
        margin=dict(l=60, r=30, t=40, b=40),
        bargap=0.05,
        showlegend=False,
    )

    svg_bytes = pio.to_image(fig, format="svg", engine="kaleido")
    return svg_bytes.decode("utf-8")


async def generate_spc_report(
    db: AsyncSession,
    recipe_id: int,
    subtask_id: int,
    version_id: int | None = None,
    date_from: datetime | None = None,
    date_to: datetime | None = None,
    operator_id: int | None = None,
    lot_number: str | None = None,
    serial_number: str | None = None,
) -> bytes:
    """Generate SPC PDF report and return bytes.

    Includes: summary, capability indices, control chart SVG, histogram SVG.
    """
    # 1. Query measurements
    measurements = await _query_measurements(
        db, recipe_id, version_id, subtask_id,
        date_from, date_to, operator_id, lot_number, serial_number,
    )

    # 2. Compute SPC data
    pass_fail_values = [m.pass_fail for m in measurements]
    values = [float(m.value) for m in measurements]
    timestamps = [m.measured_at for m in measurements]

    summary = compute_summary(pass_fail_values)

    subtask_info = await _get_subtask_info(db, subtask_id)
    tol_utl = subtask_info["utl"] if subtask_info else None
    tol_ltl = subtask_info["ltl"] if subtask_info else None
    tol_uwl = subtask_info["uwl"] if subtask_info else None
    tol_lwl = subtask_info["lwl"] if subtask_info else None
    tol_nominal = subtask_info["nominal"] if subtask_info else None

    capability = compute_capability(values, tol_utl, tol_ltl, tol_nominal)
    control_chart = compute_control_chart(
        values, timestamps, tol_utl, tol_uwl, tol_lwl, tol_ltl, tol_nominal
    )
    histogram = compute_histogram(values)

    # 3. Generate SVG charts
    control_chart_svg = _generate_control_chart_svg(control_chart, subtask_info)
    histogram_svg = _generate_histogram_svg(histogram, subtask_info)

    # 4. Get company info and recipe info
    company = await _get_company_info(db)
    recipe_info = await _get_recipe_info(db, recipe_id)

    # 5. Build filter description
    filters_desc = []
    if version_id:
        filters_desc.append(f"Version: {version_id}")
    if date_from:
        filters_desc.append(f"From: {date_from.strftime('%Y-%m-%d')}")
    if date_to:
        filters_desc.append(f"To: {date_to.strftime('%Y-%m-%d')}")
    if lot_number:
        filters_desc.append(f"Lot: {lot_number}")
    if serial_number:
        filters_desc.append(f"Serial: {serial_number}")

    # 6. Render HTML template
    template = _jinja_env.get_template("spc_report.html")
    html_content = template.render(
        company=company,
        recipe=recipe_info,
        subtask=subtask_info,
        summary=summary,
        capability=capability,
        control_chart_svg=control_chart_svg,
        histogram_svg=histogram_svg,
        filters_desc=filters_desc,
        generated_at=datetime.now(),
        n_measurements=len(measurements),
    )

    # 7. Convert HTML to PDF
    pdf_bytes = HTML(string=html_content).write_pdf()
    return pdf_bytes


async def generate_measurement_report(
    db: AsyncSession,
    recipe_id: int,
    subtask_id: int | None = None,
    version_id: int | None = None,
    date_from: datetime | None = None,
    date_to: datetime | None = None,
    operator_id: int | None = None,
    lot_number: str | None = None,
    serial_number: str | None = None,
) -> bytes:
    """Generate measurement table PDF report and return bytes."""
    # 1. Query measurements
    measurements = await _query_measurements(
        db, recipe_id, version_id, subtask_id,
        date_from, date_to, operator_id, lot_number, serial_number,
    )

    # 2. Get recipe info
    recipe_info = await _get_recipe_info(db, recipe_id)
    company = await _get_company_info(db)

    # 3. Get subtask map and operator map
    subtask_ids = {m.subtask_id for m in measurements}
    operator_ids = {m.measured_by for m in measurements}

    subtask_map = {}
    for sid in subtask_ids:
        info = await _get_subtask_info(db, sid)
        if info:
            subtask_map[sid] = info

    operator_map = await _get_operator_map(db, operator_ids)

    # 4. Build table rows
    rows = []
    for i, m in enumerate(measurements, 1):
        st_info = subtask_map.get(m.subtask_id, {})
        rows.append({
            "num": i,
            "subtask": f"#{st_info.get('marker_number', '?')} — {st_info.get('description', '?')}",
            "value": f"{float(m.value):.4f}",
            "unit": st_info.get("unit", "mm"),
            "pass_fail": m.pass_fail,
            "measured_at": m.measured_at.strftime("%Y-%m-%d %H:%M"),
            "operator": operator_map.get(m.measured_by, f"ID {m.measured_by}"),
            "lot_number": m.lot_number or "—",
            "serial_number": m.serial_number or "—",
        })

    # 5. Summary
    pass_fail_values = [m.pass_fail for m in measurements]
    summary = compute_summary(pass_fail_values)

    # 6. Build filter description
    filters_desc = []
    if subtask_id:
        si = subtask_map.get(subtask_id)
        if si:
            filters_desc.append(f"Point: #{si['marker_number']} — {si['description']}")
    if version_id:
        filters_desc.append(f"Version: {version_id}")
    if date_from:
        filters_desc.append(f"From: {date_from.strftime('%Y-%m-%d')}")
    if date_to:
        filters_desc.append(f"To: {date_to.strftime('%Y-%m-%d')}")
    if lot_number:
        filters_desc.append(f"Lot: {lot_number}")
    if serial_number:
        filters_desc.append(f"Serial: {serial_number}")

    # 7. Render template
    template = _jinja_env.get_template("measurement_report.html")
    html_content = template.render(
        company=company,
        recipe=recipe_info,
        rows=rows,
        summary=summary,
        filters_desc=filters_desc,
        generated_at=datetime.now(),
        n_measurements=len(measurements),
    )

    # 8. Convert to PDF
    pdf_bytes = HTML(string=html_content).write_pdf()
    return pdf_bytes