BeyondCX_Insights/src/exports/excel_export.py

"""
CXInsights - Excel Export

Exports analysis results to Excel format with multiple sheets.
"""

from datetime import datetime
from pathlib import Path

from src.aggregation.models import BatchAggregation
from src.models.call_analysis import CallAnalysis

# Try to import openpyxl, provide fallback
try:
    from openpyxl import Workbook
    from openpyxl.styles import Font, PatternFill, Alignment, Border, Side
    from openpyxl.utils import get_column_letter
    OPENPYXL_AVAILABLE = True
except ImportError:
    OPENPYXL_AVAILABLE = False


def export_to_excel(
    batch_id: str,
    aggregation: BatchAggregation,
    analyses: list[CallAnalysis],
    output_dir: Path,
) -> Path:
    """
    Export results to Excel file.

    Creates workbook with sheets:
    - Summary: High-level metrics
    - Lost Sales Drivers: Driver frequencies and severity
    - Poor CX Drivers: Driver frequencies and severity
    - Call Details: Individual call results
    - Emergent Patterns: New patterns found

    Args:
        batch_id: Batch identifier
        aggregation: Aggregation results
        analyses: Individual call analyses
        output_dir: Output directory

    Returns:
        Path to Excel file
    """
    if not OPENPYXL_AVAILABLE:
        raise ImportError(
            "openpyxl is required for Excel export. "
            "Install with: pip install openpyxl"
        )

    output_dir = Path(output_dir)
    output_dir.mkdir(parents=True, exist_ok=True)

    wb = Workbook()

    # Style definitions
    header_font = Font(bold=True, color="FFFFFF")
    header_fill = PatternFill(start_color="4472C4", end_color="4472C4", fill_type="solid")
    header_alignment = Alignment(horizontal="center", vertical="center")

    # Sheet 1: Summary
    ws_summary = wb.active
    ws_summary.title = "Summary"
    _create_summary_sheet(ws_summary, batch_id, aggregation, analyses, header_font, header_fill)

    # Sheet 2: Lost Sales Drivers
    ws_lost_sales = wb.create_sheet("Lost Sales Drivers")
    _create_drivers_sheet(
        ws_lost_sales,
        aggregation.lost_sales_frequencies,
        aggregation.lost_sales_severities,
        header_font,
        header_fill,
    )

    # Sheet 3: Poor CX Drivers
    ws_poor_cx = wb.create_sheet("Poor CX Drivers")
    _create_drivers_sheet(
        ws_poor_cx,
        aggregation.poor_cx_frequencies,
        aggregation.poor_cx_severities,
        header_font,
        header_fill,
    )

    # Sheet 4: Call Details
    ws_calls = wb.create_sheet("Call Details")
    _create_calls_sheet(ws_calls, analyses, header_font, header_fill)

    # Sheet 5: Emergent Patterns
    if aggregation.emergent_patterns:
        ws_emergent = wb.create_sheet("Emergent Patterns")
        _create_emergent_sheet(ws_emergent, aggregation.emergent_patterns, header_font, header_fill)

    # Save workbook
    output_path = output_dir / f"{batch_id}_analysis.xlsx"
    wb.save(output_path)

    return output_path


def _create_summary_sheet(ws, batch_id, aggregation, analyses, header_font, header_fill):
    """Create summary sheet."""
    # Title
    ws["A1"] = "CXInsights Analysis Report"
    ws["A1"].font = Font(bold=True, size=16)
    ws.merge_cells("A1:D1")

    ws["A2"] = f"Batch ID: {batch_id}"
    ws["A3"] = f"Generated: {datetime.utcnow().strftime('%Y-%m-%d %H:%M UTC')}"

    # Metrics section
    row = 5
    ws[f"A{row}"] = "Key Metrics"
    ws[f"A{row}"].font = Font(bold=True, size=12)

    metrics = [
        ("Total Calls Analyzed", aggregation.total_calls_processed),
        ("Successful Analyses", aggregation.successful_analyses),
        ("Failed Analyses", aggregation.failed_analyses),
        ("Success Rate", f"{aggregation.successful_analyses / aggregation.total_calls_processed * 100:.1f}%" if aggregation.total_calls_processed > 0 else "N/A"),
    ]

    if aggregation.rca_tree:
        tree = aggregation.rca_tree
        metrics.extend([
            ("", ""),
            ("Calls with Lost Sales Issues", tree.calls_with_lost_sales),
            ("Calls with Poor CX Issues", tree.calls_with_poor_cx),
            ("Calls with Both Issues", tree.calls_with_both),
        ])

    for i, (label, value) in enumerate(metrics):
        ws[f"A{row + 1 + i}"] = label
        ws[f"B{row + 1 + i}"] = value

    # Top drivers section
    row = row + len(metrics) + 3
    ws[f"A{row}"] = "Top Lost Sales Drivers"
    ws[f"A{row}"].font = Font(bold=True, size=12)

    for i, freq in enumerate(aggregation.lost_sales_frequencies[:5]):
        ws[f"A{row + 1 + i}"] = freq.driver_code
        ws[f"B{row + 1 + i}"] = f"{freq.call_rate:.1%}"

    row = row + 7
    ws[f"A{row}"] = "Top Poor CX Drivers"
    ws[f"A{row}"].font = Font(bold=True, size=12)

    for i, freq in enumerate(aggregation.poor_cx_frequencies[:5]):
        ws[f"A{row + 1 + i}"] = freq.driver_code
        ws[f"B{row + 1 + i}"] = f"{freq.call_rate:.1%}"

    # Adjust column widths
    ws.column_dimensions["A"].width = 30
    ws.column_dimensions["B"].width = 20


def _create_drivers_sheet(ws, frequencies, severities, header_font, header_fill):
    """Create driver analysis sheet."""
    headers = [
        "Rank",
        "Driver Code",
        "Occurrences",
        "Calls Affected",
        "Call Rate",
        "Avg Confidence",
        "Severity Score",
        "Impact Level",
    ]

    # Write headers
    for col, header in enumerate(headers, 1):
        cell = ws.cell(row=1, column=col, value=header)
        cell.font = header_font
        cell.fill = header_fill

    # Create severity lookup
    sev_map = {s.driver_code: s for s in severities}

    # Write data
    for row, freq in enumerate(frequencies, 2):
        sev = sev_map.get(freq.driver_code)

        ws.cell(row=row, column=1, value=row - 1)
        ws.cell(row=row, column=2, value=freq.driver_code)
        ws.cell(row=row, column=3, value=freq.total_occurrences)
        ws.cell(row=row, column=4, value=freq.calls_affected)
        ws.cell(row=row, column=5, value=f"{freq.call_rate:.1%}")
        ws.cell(row=row, column=6, value=f"{freq.avg_confidence:.2f}")
        ws.cell(row=row, column=7, value=f"{sev.severity_score:.1f}" if sev else "N/A")
        ws.cell(row=row, column=8, value=sev.impact_level.value if sev else "N/A")

    # Adjust column widths
    for col in range(1, len(headers) + 1):
        ws.column_dimensions[get_column_letter(col)].width = 15


def _create_calls_sheet(ws, analyses, header_font, header_fill):
    """Create call details sheet."""
    headers = [
        "Call ID",
        "Outcome",
        "Status",
        "Lost Sales Drivers",
        "Poor CX Drivers",
        "Audio Duration (s)",
    ]

    # Write headers
    for col, header in enumerate(headers, 1):
        cell = ws.cell(row=1, column=col, value=header)
        cell.font = header_font
        cell.fill = header_fill

    # Write data
    for row, analysis in enumerate(analyses, 2):
        lost_sales = ", ".join(d.driver_code for d in analysis.lost_sales_drivers)
        poor_cx = ", ".join(d.driver_code for d in analysis.poor_cx_drivers)

        ws.cell(row=row, column=1, value=analysis.call_id)
        ws.cell(row=row, column=2, value=analysis.outcome.value)
        ws.cell(row=row, column=3, value=analysis.status.value)
        ws.cell(row=row, column=4, value=lost_sales or "-")
        ws.cell(row=row, column=5, value=poor_cx or "-")
        ws.cell(row=row, column=6, value=analysis.observed.audio_duration_sec)

    # Adjust column widths
    ws.column_dimensions["A"].width = 15
    ws.column_dimensions["B"].width = 20
    ws.column_dimensions["C"].width = 12
    ws.column_dimensions["D"].width = 40
    ws.column_dimensions["E"].width = 40
    ws.column_dimensions["F"].width = 18


def _create_emergent_sheet(ws, emergent_patterns, header_font, header_fill):
    """Create emergent patterns sheet."""
    headers = [
        "Proposed Label",
        "Occurrences",
        "Avg Confidence",
        "Sample Evidence",
    ]

    # Write headers
    for col, header in enumerate(headers, 1):
        cell = ws.cell(row=1, column=col, value=header)
        cell.font = header_font
        cell.fill = header_fill

    # Write data
    for row, pattern in enumerate(emergent_patterns, 2):
        evidence = "; ".join(pattern.get("sample_evidence", [])[:2])

        ws.cell(row=row, column=1, value=pattern.get("proposed_label", "N/A"))
        ws.cell(row=row, column=2, value=pattern.get("occurrences", 0))
        ws.cell(row=row, column=3, value=f"{pattern.get('avg_confidence', 0):.2f}")
        ws.cell(row=row, column=4, value=evidence[:100] if evidence else "-")

    # Adjust column widths
    ws.column_dimensions["A"].width = 30
    ws.column_dimensions["B"].width = 12
    ws.column_dimensions["C"].width = 15
    ws.column_dimensions["D"].width = 60