txt doc rtf html,JAVA读取WORD,EXCEL,PDF,TXT,RTF,HTML文件文本内容的方法示例.docx

JAVA读取WORD,EXCEL,PDF,TXT,RTF,HTML文件文本内容的方法示例

JAVA读取WORD,EXCEL,PDF,TXT,RTF,HTML文件文本内容的方法示例??2012-06-29 17:13:08|??分类：JAVA |??标签： |字号大中小?订阅以下是Java对几种文本文件内容读取代码。其中，OFFICE文档(WORD,EXCEL)使用了POI控件，PDF使用了PDFBOX控件。查看相关控件的下载地址和配置方法。转自：/allan811112/blog/item/d77b70492f0a65fb82025c3b.htmlWORD Java代码package textReader; import java.io.*; import org.apache.poi.hwpf.extractor.WordExtractor; public class WordReader { public WordReader(){ } /** * @param filePath 文件路径* @return 读出的Word的内容*/ public String getTextFromWord(String filePath){ String result = null; File file = new File(filePath); try{ FileInputStream fis = new FileInputStream(file); WordExtractor wordExtractor = new WordExtractor(fis); result = wordExtractor.getText(); }catch(FileNotFoundException e){ e.printStackTrace(); }catch(IOException e){ e.printStackTrace(); }; return result; } } EXCEL Java代码package textReader; import org.apache.poi.hssf.usermodel.HSSFWorkbook; import org.apache.poi.hssf.usermodel.HSSFSheet; import org.apache.poi.hssf.usermodel.HSSFRow; import org.apache.poi.hssf.usermodel.HSSFCell; import java.io.FileInputStream; import java.io.FileNotFoundException; import java.io.IOException; public class ExcelReader { @SuppressWarnings("deprecation") /** * @param filePath 文件路径* @return 读出的Excel的内容*/ public String getTextFromExcel(String filePath) { StringBuffer buff = new StringBuffer(); try { //创建对Excel工作簿文件的引用HSSFWorkbook wb = new HSSFWorkbook(new FileInputStream(filePath)); //创建对工作表的引用。????????? for (int numSheets = 0; numSheets < wb.getNumberOfSheets(); numSheets++) { if (null != wb.getSheetAt(numSheets)) { HSSFSheet aSheet = wb.getSheetAt(numSheets);//获得一个sheet for (int rowNumOfSheet = 0; rowNumOfSheet <= aSheet.getLastRowNum(); rowNumOfSheet++) { if (null != aSheet.getRow(rowNumOfSheet)) { HSSFRow aRow = aSheet.getRow(rowNumOfSheet); //获得一个行for (int cellNumOfRow = 0; cellNumOfRow <= aRow.getLastCellNum(); cellNumOfRow++) { if (null != aRow.getCell(cellNumOfRow)) { HSSFCell aCell = aRow.getCell(cellNumOfRow);//获得列值switch(aCell.getCellType()){ case HSSFC