app/api/documents/process/route.ts

import { NextRequest, NextResponse } from 'next/server';
import mammoth from 'mammoth';
import ExcelJS from 'exceljs';
import fs from 'fs';
import path from 'path';

// Use /data for Hugging Face Spaces persistent storage
const DATA_DIR = process.env.SPACE_ID
  ? '/data'
  : path.join(process.cwd(), 'public', 'data');
const PUBLIC_DIR = path.join(DATA_DIR, 'public');

export async function POST(request: NextRequest) {
  try {
    const body = await request.json();
    const { fileName, isPublic = false, operation = 'read', key } = body;

    if (!fileName) {
      return NextResponse.json(
        { success: false, error: 'File name is required' },
        { status: 400 }
      );
    }

    let targetDir = PUBLIC_DIR;
    if (!isPublic) {
      if (!key) {
        return NextResponse.json(
          { success: false, error: 'Passkey (key) is required for non-public files' },
          { status: 401 }
        );
      }
      const sanitizedKey = key.replace(/[^a-zA-Z0-9_-]/g, '');
      targetDir = path.join(DATA_DIR, sanitizedKey);
    }

    // Get file buffer
    const filePath = path.join(targetDir, fileName);
    if (!fs.existsSync(filePath)) {
      return NextResponse.json(
        { success: false, error: 'File not found' },
        { status: 404 }
      );
    }

    const fileBuffer = fs.readFileSync(filePath);
    const ext = fileName.split('.').pop()?.toLowerCase();
    let content: any = {};

    switch (ext) {
      case 'docx':
        try {
          const result = await mammoth.extractRawText({ buffer: fileBuffer });
          content = {
            type: 'docx',
            text: result.value,
            messages: result.messages
          };

          const htmlResult = await mammoth.convertToHtml({ buffer: fileBuffer });
          content.html = htmlResult.value;
        } catch (error) {
          content = {
            type: 'docx',
            error: 'Failed to process Word document',
            details: error
          };
        }
        break;

      case 'xlsx':
      case 'xls':
        try {
          const workbook = new ExcelJS.Workbook();
          await workbook.xlsx.load(fileBuffer as any);

          const sheets: any[] = [];
          workbook.eachSheet((worksheet) => {
            const sheetData: any = {
              name: worksheet.name,
              rowCount: worksheet.rowCount,
              columnCount: worksheet.columnCount,
              data: []
            };

            worksheet.eachRow((row, rowNumber) => {
              const rowData: any[] = [];
              row.eachCell((cell, colNumber) => {
                rowData.push({
                  value: cell.value,
                  type: cell.type,
                  formula: cell.formula
                });
              });
              sheetData.data.push(rowData);
            });

            sheets.push(sheetData);
          });

          content = {
            type: 'excel',
            sheets,
            sheetCount: sheets.length
          };
        } catch (error) {
          content = {
            type: 'excel',
            error: 'Failed to process Excel spreadsheet',
            details: error
          };
        }
        break;

      case 'pdf':
        try {
          const pdf = require('pdf-parse');
          const data = await pdf(fileBuffer);
          content = {
            type: 'pdf',
            text: data.text,
            info: data.info,
            metadata: data.metadata,
            version: data.version,
            numpages: data.numpages
          };
        } catch (error) {
          content = {
            type: 'pdf',
            error: 'Failed to process PDF document',
            details: error
          };
        }
        break;

      case 'pptx':
      case 'ppt':
        content = {
          type: 'powerpoint',
          fileName,
          size: fileBuffer.length,
          message: 'PowerPoint processing requires additional libraries'
        };
        break;

      case 'txt':
      case 'md':
      case 'json':
      case 'csv':
        content = {
          type: ext,
          text: fileBuffer.toString('utf-8')
        };
        break;

      default:
        content = {
          type: 'unknown',
          fileName,
          size: fileBuffer.length,
          message: 'Unknown file type'
        };
    }

    if (operation === 'analyze' && content.text) {
      const text = content.text || '';
      content.analysis = {
        characterCount: text.length,
        wordCount: text.split(/\s+/).filter(Boolean).length,
        lineCount: text.split('\n').length,
        paragraphCount: text.split('\n\n').filter(Boolean).length
      };
    }

    return NextResponse.json({
      success: true,
      fileName,
      operation,
      content
    });
  } catch (error) {
    console.error('Error processing document:', error);
    return NextResponse.json(
      { success: false, error: 'Failed to process document' },
      { status: 500 }
    );
  }
}

export async function GET() {
  return NextResponse.json({
    message: 'Document processing endpoint',
    endpoint: '/api/documents/process',
    method: 'POST',
    body: {
      fileName: 'Name of the file to process',
      isPublic: 'true/false - whether file is in public folder',
      key: 'Passkey for secure storage (required if not public)',
      operation: 'Operation to perform: read (default), analyze'
    },
    supportedFormats: [
      'docx - Word documents (text extraction)',
      'xlsx/xls - Excel spreadsheets (data extraction)',
      'pdf - PDF files (metadata only)',
      'pptx/ppt - PowerPoint (metadata only)',
      'txt/md/json/csv - Text files (full content)'
    ]
  });
}