package com.product.lucene.util;

import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileReader;
import java.io.IOException;
import java.io.InputStream;
import java.util.List;

import org.apache.poi.hssf.usermodel.HSSFCell;
import org.apache.poi.hssf.usermodel.HSSFRow;
import org.apache.poi.hssf.usermodel.HSSFSheet;
import org.apache.poi.hssf.usermodel.HSSFWorkbook;
import org.apache.poi.hwpf.HWPFDocument;
import org.apache.poi.hwpf.usermodel.Range;
import org.apache.poi.xssf.usermodel.XSSFCell;
import org.apache.poi.xssf.usermodel.XSSFRow;
import org.apache.poi.xssf.usermodel.XSSFSheet;
import org.apache.poi.xssf.usermodel.XSSFWorkbook;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFParagraph;
import org.apache.poi.xwpf.usermodel.XWPFTable;
import org.apache.poi.xwpf.usermodel.XWPFTableCell;
import org.apache.poi.xwpf.usermodel.XWPFTableRow;


public class FileUtils {

	/**
	 * 	文件转string
	 * @param file
	 * @return
	 */
	public static String FileToString(File file) {
		String file_content="";
    	if ((file != null) && (file.isFile())) {
    		if (file.getName().toLowerCase().endsWith(".txt")) {
    			file_content = txtToString(file);
			} else if ((file.getName().toLowerCase().endsWith(".doc"))){
				file_content = docToString(file);
			} else if ((file.getName().toLowerCase().endsWith(".docx"))) {
				file_content = docxToString(file);
			} else if (file.getName().toLowerCase().endsWith(".xls")) {
				file_content = readXls(file);
			} else if (file.getName().toLowerCase().endsWith(".xlsx")) {
				file_content = readXlsx(file);
			}
		}
    	return file_content;
    }
	
	/**
	 * 	txt文件读取
	 * @param file
	 * @return
	 */
	public static String txtToString(File file) {
		String result = "";
		try (BufferedReader br = new BufferedReader(new FileReader(file))){// 构造一个BufferedReader类来读取文件
			String s = null;
			// 使用readLine方法，一次读一行
			while ((s = br.readLine()) != null) {
				result = result + "\n" + s;
			}
		} catch (Exception e) {
			e.printStackTrace();
		}
		return result;
	}
	
	/**
	 * 	doc转文本
	 * @param file
	 * @return
	 */
	public static String docToString(File file) {
		String result = "";
		try (
			FileInputStream fileInputStream = new FileInputStream(file);
			HWPFDocument doc = new HWPFDocument(fileInputStream);
		){
			Range rang = doc.getRange();
			result += rang.text();
		} catch (Exception e) {
			e.printStackTrace();
		}
		return result;
	}
	
	/**
	 * 	docx转文本
	 * @param file
	 * @return
	 * @throws IOException
	 */
	public static String docxToString(File file) {
		StringBuffer s=new StringBuffer();
		try (
			InputStream inputStream = new FileInputStream(file);
			XWPFDocument doc = new XWPFDocument(inputStream);
		){
			List<XWPFParagraph> paras = doc.getParagraphs();
			for (XWPFParagraph para : paras) {
				// 当前段落的属性
				// CTPPr pr = para.getCTP().getPPr();
				s.append( para.getText());
			}
			// 获取文档中所有的表格
			List<XWPFTable> tables = doc.getTables();
			List<XWPFTableRow> rows;
			List<XWPFTableCell> cells;
			for (XWPFTable table : tables) {
				// 表格属性
				// CTTblPr pr = table.getCTTbl().getTblPr();
				// 获取表格对应的行
				rows = table.getRows();
				for (XWPFTableRow row : rows) {
					// 获取行对应的单元格
					cells = row.getTableCells();
					for (XWPFTableCell cell : cells) {
						s.append( cell.getText());
					}
				}
			}
		} catch (Exception e) {
			e.printStackTrace();
		}
		return s.toString();
	}
	
	/**
	 * `xls转字符
	 * @param f
	 * @return
	 * @throws IOException
	 */
	public static String readXls(File file) {
		StringBuffer info = new StringBuffer();
		try(
			InputStream inputStream = new FileInputStream(file);
			HSSFWorkbook hssfWorkbook = new HSSFWorkbook(inputStream);
		)
		{
			// Read the Sheet
			for (int numSheet = 0; numSheet < hssfWorkbook.getNumberOfSheets(); numSheet++) {
				HSSFSheet hssfSheet = hssfWorkbook.getSheetAt(numSheet);
				if (hssfSheet == null) {
					continue;
				}
				// Read the Row
				for (int rowNum = 0; rowNum <= hssfSheet.getLastRowNum(); rowNum++) {
					HSSFRow hssfRow = hssfSheet.getRow(rowNum);
					if (hssfRow != null) {
						int cols = hssfRow.getLastCellNum();
						for (int j = 0; j <= cols; j++) {
							HSSFCell no = hssfRow.getCell(j);
							info.append(getValue(no)).append(" ");
						}
					}
				}
			}
		} catch (Exception e) {
			e.printStackTrace();
		}
		return info.toString();
	}
	
	/**
	 * 	xlsx转字符
	 * @param file
	 * @return
	 * @throws IOException
	 */
	public static String readXlsx(File file) {
		StringBuffer info = new StringBuffer();
		try(
				InputStream is = new FileInputStream(file);
				XSSFWorkbook xssfWorkbook = new XSSFWorkbook(is);
		){
		
		for (int numSheet = 0; numSheet < xssfWorkbook.getNumberOfSheets(); numSheet++) {
			XSSFSheet xssfSheet = xssfWorkbook.getSheetAt(numSheet);
			if (xssfSheet == null) {
				continue;
			}
			// Read the Row
			for (int rowNum = 0; rowNum <= xssfSheet.getLastRowNum(); rowNum++) {
				XSSFRow xssfRow = xssfSheet.getRow(rowNum);
				if (xssfRow != null) {
					int tdLength = xssfRow.getLastCellNum();
					for (int j = 0; j <= tdLength; j++) {
						XSSFCell no = xssfRow.getCell(j);
						info.append(getValue(no)).append(" ");
					}
				}
			}
		}
		} catch (Exception e) {
			e.printStackTrace();
		}
		return info.toString();
	}
	
	/**
	 * 	获取xls单元格值
	 * @param hssfCell
	 * @return
	 */
	private static String getValue(HSSFCell hssfCell) {
		if (hssfCell == null)
			return "";
		if (hssfCell.getCellType() == hssfCell.CELL_TYPE_BOOLEAN) {
			return String.valueOf(hssfCell.getBooleanCellValue());
		} else if (hssfCell.getCellType() == hssfCell.CELL_TYPE_NUMERIC) {
			return String.valueOf(hssfCell.getNumericCellValue());
		} else {
			hssfCell.setCellType(hssfCell.CELL_TYPE_STRING);
			return String.valueOf(hssfCell.getStringCellValue());
		}
	}
	
	/**
	 * 	获取xlxs单元格值
	 * @param xssfRow
	 * @return
	 */
	private static String getValue(XSSFCell xssfRow) {
		if (xssfRow == null)
			return "";
		if (xssfRow.getCellType() == xssfRow.CELL_TYPE_BOOLEAN) {
			return String.valueOf(xssfRow.getBooleanCellValue());
		} else if (xssfRow.getCellType() == xssfRow.CELL_TYPE_NUMERIC) {
			return String.valueOf(xssfRow.getNumericCellValue());
		} else {
			return String.valueOf(xssfRow.getStringCellValue());
		}
	}
	
	public static void main(String[] args) {
		File f1=new File("D://ceshi/a.txt");
		File f2=new File("D://ceshi/b.doc");
		File f3=new File("D://ceshi/c.docx");
		File f4=new File("D://ceshi/d.xls");
		File f5=new File("D://ceshi/e.xlsx");
		System.out.println(txtToString(f1));
		System.out.println("+++++++++++++++++++++++++++++++++++++++++++");
		System.out.println(docToString(f2));
		System.out.println("+++++++++++++++++++++++++++++++++++++++++++");
		System.out.println(docxToString(f3));
		System.out.println("+++++++++++++++++++++++++++++++++++++++++++");
		System.out.println(readXls(f4));
		System.out.println("+++++++++++++++++++++++++++++++++++++++++++");
		System.out.println(readXlsx(f5));
	}
}