POI对Excel自定义日期格式的读取

Posted 2020-10-23 史红星

cell.getCellStyle().getDataFormat();根据这个值进行时间、日期格式的判断；

POI读取出来的结果也是有些变化的；需要在实际项目中进行确认；

时间格式的遍历：

 /**
     * 处理数据类型
     * 
     * @param attributes
     */
    public void setNextDataType(Attributes attributes) {
        nextDataType = CellDataType.NUMBER;
        formatIndex = -1;
        formatString = null;
        String cellType = attributes.getValue("t");
        String cellStyleStr = attributes.getValue("s");
        String columData = attributes.getValue("r");
        if ("b".equals(cellType)) {
            nextDataType = CellDataType.BOOL;
        } else if ("e".equals(cellType)) {
            nextDataType = CellDataType.ERROR;
        } else if ("inlineStr".equals(cellType)) {
            nextDataType = CellDataType.INLINESTR;
        } else if ("s".equals(cellType)) {
            nextDataType = CellDataType.SSTINDEX;
        } else if ("str".equals(cellType)) {
            nextDataType = CellDataType.FORMULA;
        }
        if (cellStyleStr != null) {
            int styleIndex = Integer.parseInt(cellStyleStr);
            XSSFCellStyle style = this.stylesTable.getStyleAt(styleIndex);
            formatIndex = style.getDataFormat();
            formatString = style.getDataFormatString();
            short format=this.formatIndex;
             if (format == 14 || format == 31 || format == 57 || 
                        format == 58 || (176 <= format && format <= 178)
                        || (182 <= format && format <= 196) || 
                        (210 <= format && format <= 213) || (208 == format)) 
                { // 日期
                    this.formatString = "yyyy-MM-dd";
                    nextDataType = CellDataType.NUMBER;
                } else if (format == 20 || format == 32 || format == 183 || (200 <= format && format <= 209)) { // 时间
                    this.formatString = "HH:mm";
                    nextDataType = CellDataType.NUMBER;
                }
             if (this.formatString == null)
                {
                     nextDataType = CellDataType.NULL;
                     formatString = BuiltinFormats.getBuiltinFormat(formatIndex);
                }
        }
    }

 

 

package com.beyondsoft.util.base;

import java.io.InputStream;
import java.text.SimpleDateFormat;
import java.util.ArrayList;
import java.util.Date;
import java.util.Iterator;
import java.util.List;

import org.apache.commons.lang.StringUtils;
import org.apache.poi.hssf.usermodel.HSSFDateUtil;
import org.apache.poi.openxml4j.opc.OPCPackage;
import org.apache.poi.ss.usermodel.BuiltinFormats;
import org.apache.poi.ss.usermodel.DataFormatter;
import org.apache.poi.xssf.eventusermodel.ReadOnlySharedStringsTable;
import org.apache.poi.xssf.eventusermodel.XSSFReader;
import org.apache.poi.xssf.model.StylesTable;
import org.apache.poi.xssf.usermodel.XSSFCellStyle;
import org.apache.poi.xssf.usermodel.XSSFRichTextString;
import org.xml.sax.Attributes;
import org.xml.sax.InputSource;
import org.xml.sax.SAXException;
import org.xml.sax.XMLReader;
import org.xml.sax.helpers.DefaultHandler;
import org.xml.sax.helpers.XMLReaderFactory;

import com.beyondsoft.util.base.XLSXCovertCSVReader2.xssfDataType;

/**
 * 抽象Excel2007读取器，excel2007的底层数据结构是xml文件，采用SAX的事件驱动的方法解析
 * xml，需要继承DefaultHandler，在遇到文件内容时，事件会触发，这种做法可以大大降低
 * 内存的耗费，特别使用于大数据量的文件。
 * 作者：史红星
 *
 */
public class Excel2007Reader extends DefaultHandler {
    //共享字符串表
    private   ReadOnlySharedStringsTable sharedStringsTable;
    //上一次的内容
    private String lastContents;
    private boolean nextIsString;

   // Set when V start element is seen  
    private boolean vIsOpen;  
    private int sheetIndex = -1;
    private List<String> rowlist = new ArrayList<String>();
    //当前行
    private int curRow = 0;
    //当前列
    private int curCol = 0;
    private int preCol = 0;     //上一列列索引  
    //日期标志
    private boolean dateFlag;
    //数字标志
    private boolean numberFlag;
      /**
     * T元素标识
     */
    private boolean isTElement;
    
    /**
     * 单元格数据类型，默认为字符串类型
     */
    private CellDataType nextDataType = CellDataType.SSTINDEX;

    private final DataFormatter formatter = new DataFormatter();
    
    // 定义前一个元素和当前元素的位置，用来计算其中空的单元格数量，如A6和A8等
    private String preRef = null, ref = null;

    // 定义该文档一行最大的单元格数，用来补全一行最后可能缺失的单元格
    private String maxRef = null;
    
    // 定义该文档一行最小单元格，用来补全一行最后可能缺失的单元格
    private Integer minIndex = null;
    
    private short formatIndex;

    private String formatString;
    
    //首列是否为空
    private boolean isRowStartNull;
    
    /**
     * 单元格
     */
    private StylesTable stylesTable;

    //填充字符串
    private static final String CELL_FILL_STR = "@";
    //列的最大位数
    private static final int MAX_CELL_BIT = 3;
    private IExcelRowReader rowReader;
    
    public void setRowReader(IExcelRowReader rowReader){
        this.rowReader = rowReader;
    }
    
    /**只遍历一个电子表格，其中sheetId为要遍历的sheet索引，从1开始，1-3
     * @param filename
     * @param sheetId
     * @throws Exception
     */
    public void processOneSheet(String filename,int sheetId) throws Exception {
        OPCPackage pkg = OPCPackage.open(filename);
        XSSFReader r = new XSSFReader(pkg);
        this.sharedStringsTable = new ReadOnlySharedStringsTable(pkg);
        this.stylesTable = r.getStylesTable();
        XMLReader parser = fetchSheetParser(this.sharedStringsTable);
        // 根据 rId# 或 rSheet# 查找sheet
        InputStream sheet2 = r.getSheet("rId"+sheetId);
        sheetIndex++;
        InputSource sheetSource = new InputSource(sheet2);
        parser.parse(sheetSource);
        sheet2.close();
    }

    /**
     * 遍历工作簿中所有的电子表格
     * @param filename
     * @throws Exception
     */
    public void process(String filename) throws Exception {
        OPCPackage pkg = OPCPackage.open(filename);
        XSSFReader r = new XSSFReader(pkg);
        ReadOnlySharedStringsTable strings = new ReadOnlySharedStringsTable(pkg);
        this.stylesTable = r.getStylesTable();
        XMLReader parser = fetchSheetParser(this.sharedStringsTable);
        Iterator<InputStream> sheets = r.getSheetsData();
        while (sheets.hasNext()) {
            curRow = 0;
            sheetIndex++;
            InputStream sheet = sheets.next();
            InputSource sheetSource = new InputSource(sheet);
            parser.parse(sheetSource);
            sheet.close();
        }
    }
    /**
     * 遍历工作簿中所有的电子表格
     * @param filename
     * @throws Exception
     */
    public void process(InputStream filename) throws Exception {
        OPCPackage pkg = OPCPackage.open(filename);
        XSSFReader r = new XSSFReader(pkg);
        this.stylesTable = r.getStylesTable();
        //this.sharedStringsTable = new ReadOnlySharedStringsTable(pkg);
        ReadOnlySharedStringsTable strings = new ReadOnlySharedStringsTable(pkg);
        XMLReader parser = fetchSheetParser(strings);
        Iterator<InputStream> sheets = (XSSFReader.SheetIterator)r.getSheetsData();
        while (sheets.hasNext()) {
            curRow = 0;
            sheetIndex++;
            InputStream sheet = sheets.next();
            InputSource sheetSource = new InputSource(sheet);
            parser.parse(sheetSource);
            sheet.close();
        }
    }

    public XMLReader fetchSheetParser(ReadOnlySharedStringsTable sharedStringsTable)
            throws SAXException {
        XMLReader parser = XMLReaderFactory
                .createXMLReader("org.apache.xerces.parsers.SAXParser");
        this.sharedStringsTable = sharedStringsTable;
        parser.setContentHandler(this);
        return parser;
    }
    //得到列索引，每一列c元素的r属性构成为字母加数字的形式，字母组合为列索引，数字组合为行索引，  
    //如AB45,表示为第（A-A+1）*26+（B-A+1）*26列，45行  
    public int getRowIndex(String rowStr){  
        rowStr = rowStr.replaceAll("[^A-Z]", "");  
        byte[] rowAbc = rowStr.getBytes();  
        int len = rowAbc.length;  
        float num = 0;  
        for (int i=0;i<len;i++){  
            num += (rowAbc[i]-\'A\'+1)*Math.pow(26,len-i-1 );  
        }  
        return (int) num;  
    } 
    
    /**
     * 单元格中的数据可能的数据类型
     */
    enum CellDataType {
        BOOL, ERROR, FORMULA, INLINESTR, SSTINDEX, NUMBER, NULL,DATE
    }
    /**
     * 处理数据类型
     * 
     * @param attributes
     */
    public void setNextDataType(Attributes attributes) {
        nextDataType = CellDataType.NUMBER;
        formatIndex = -1;
        formatString = null;
        String cellType = attributes.getValue("t");
        String cellStyleStr = attributes.getValue("s");
        String columData = attributes.getValue("r");
        if ("b".equals(cellType)) {
            nextDataType = CellDataType.BOOL;
        } else if ("e".equals(cellType)) {
            nextDataType = CellDataType.ERROR;
        } else if ("inlineStr".equals(cellType)) {
            nextDataType = CellDataType.INLINESTR;
        } else if ("s".equals(cellType)) {
            nextDataType = CellDataType.SSTINDEX;
        } else if ("str".equals(cellType)) {
            nextDataType = CellDataType.FORMULA;
        }
        if (cellStyleStr != null) {
            int styleIndex = Integer.parseInt(cellStyleStr);
            XSSFCellStyle style = this.stylesTable.getStyleAt(styleIndex);
            formatIndex = style.getDataFormat();
            formatString = style.getDataFormatString();
            short format=this.formatIndex;
             if (format == 14 || format == 31 || format == 57 || 
                        format == 58 || (176 <= format && format <= 178)
                        || (182 <= format && format <= 196) || 
                        (210 <= format && format <= 213) || (208 == format)) 
                { // 日期
                    this.formatString = "yyyy-MM-dd";
                    nextDataType = CellDataType.NUMBER;
                } else if (format == 20 || format == 32 || format == 183 || (200 <= format && format <= 209)) { // 时间
                    this.formatString = "HH:mm";
                    nextDataType = CellDataType.NUMBER;
                }
             if (this.formatString == null)
                {
                     nextDataType = CellDataType.NULL;
                     formatString = BuiltinFormats.getBuiltinFormat(formatIndex);
                }
        }
    }
    
    /**
     * 对解析出来的数据进行类型处理
     * 
     * @param value
     *            单元格的值（这时候是一串数字）
     * @param thisStr
     *            一个空字符串
     * @return
     */
    @SuppressWarnings("deprecation")
    public String getDataValue(String value, String thisStr) 
    {
        switch (nextDataType) {
        // 这几个的顺序不能随便交换，交换了很可能会导致数据错误
        case BOOL:
            char first = value.charAt(0);
            thisStr = first == \'0\' ? "FALSE" : "TRUE";
            break;
        case ERROR:
            thisStr = "\\"ERROR:" + value.toString() + \'"\';
            break;
        case FORMULA:
            thisStr = \'"\' + value.toString() + \'"\';
            break;
        case INLINESTR:
            XSSFRichTextString rtsi = new XSSFRichTextString(value.toString());

            thisStr = rtsi.toString();
            rtsi = null;
            break;
        case SSTINDEX:
            String sstIndex = value.toString();
            try {
                int idx = Integer.parseInt(sstIndex);
                XSSFRichTextString rtss = new XSSFRichTextString(sharedStringsTable.getEntryAt(idx));
                thisStr = rtss.toString();
                rtss = null;
            } catch (NumberFormatException ex) {
                thisStr = value.toString();
            }
            break;
        case NUMBER:
            String n = value.toString();
            // 判断是否是日期格式
            if (HSSFDateUtil.isADateFormat(this.formatIndex, n)) {
                Double d = Double.parseDouble(n);
                Date date = HSSFDateUtil.getJavaDate(d);
                thisStr = formateDateToString(date);
            } else if (this.formatString != null) {
                thisStr = this.formatter.formatRawCellContents(
                        Double.parseDouble(n), this.formatIndex,
                        this.formatString);
            } else {
                thisStr = n;
            }
            break;
        default:
            thisStr = " ";

            break;
        }

        return thisStr;
    }

   /**
     * 字符串的填充
     * 
     * @param str
     * @param len
     * @param let
     * @param isPre
     * @return
     */
    String fillChar(String str, int len, char let, boolean isPre) {
        int len_1 = str.length();
        if (len_1 < len) {
            if (isPre) {
                for (int i = 0; i < (len - len_1); i++) {
                    str = let + str;
                }
            } else {
                for (int i = 0; i < (len - len_1); i++) {
                    str = str + let;
                }
            }
        }
        return str;
    }

    public static void main(String[] args) {
        String ref="C1";
        String preRef="A1";
        Excel2007Reader Excel2007Reader=new Excel2007Reader();
        System.out.println(Excel2007Reader.countNullCell(ref,preRef));
    }

       /**
     * 计算两个单元格之间的单元格数目(同一行)
     * 
     * @param ref
     * @param preRef
     * @return
     */
    public int countNullCell(String ref, String preRef) {
        // excel2007最大行数是1048576，最大列数是16384，最后一列列名是XFD
        String xfd = ref.replaceAll("\\\\d+", "");
        String xfd_1 = preRef.replaceAll("\\\\d+", "");

        xfd = fillChar(xfd, 3, \'@\', true);
        xfd_1 = fillChar(xfd_1, 3, \'@\', true);

        char[] letter = xfd.toCharArray();
        char[] letter_1 = xfd_1.toCharArray();
        int res = (letter[0] - letter_1[0]) * 26 * 26 + (letter[1] - letter_1[1]) * 26 + (letter[2] - letter_1[2]);
        return res;
    }
    
    /**
     * 填充空白单元格
     *
     * @param curCoordinate
     * @param preCoordinate
     */
    private void fillBlackCell(String curCoordinate, String preCoordinate, boolean isEnd) {
        if (!curCoordinate.equals(preCoordinate)) {
            int len = calNullCellCount(curCoordinate, preCoordinate, isEnd);
//            for (int i = 0; i < len; i++) {
//                rowCellList.add(curCell++, "");
//            }
        }
    }
    private String formateDateToString(Date date) {
        SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");// 格式化日期
        return sdf.format(date);

    }
    /**
     * 计算当前单元格和前一个单元格之间的空白单元格数量
     * 如果是尾部则不减1
     *
     * @param curCoordinate
     * @param preCoordinate
     * @return
     */
    private int calNullCellCount(String curCoordinate, String preCoordinate, boolean isEnd) {
        // excel2007最大行数是1048576，最大列数是16384，最后一列列名是XFD
        String curCellCoordinate = curCoordinate.replaceAll("\\\\d+", "");
        String preCellCoordinate = preCoordinate.replaceAll("\\\\d+", "");

        curCellCoordinate = fillChar(curCellCoordinate, MAX_CELL_BIT, CELL_FILL_STR);
        preCellCoordinate = fillChar(preCellCoordinate, MAX_CELL_BIT, CELL_FILL_STR);

        char[] cur = curCellCoordinate.toCharArray();
        char[] pre = preCellCoordinate.toCharArray();

        int len = (cur[0] - pre[0]) * 26 * 26 + (cur[1] - pre[1]) * 26 + (cur[2] - pre[2]);
        if (!isEnd) {
            len = len - 1;
        }
        return len;
    }

    /**
     * 将不足指定位数的字符串补全，高位补上指定字符串
     *
     * @param cellCoordinate
     * @param maxLen
     * @param fillChar
     * @return
     */
    private String fillChar(String cellCoordinate, int maxLen, String fillChar) {
        int coordinateLen = cellCoordinate.length();
        if (coordinateLen < maxLen) {
            for (int i = 0; i < (maxLen - coordinateLen); i++) {
                cellCoordinate = fillChar + cellCoordinate;
            }
        }
        return cellCoordinate;
    }
    public void startElement(String uri, String localName, String name,
            Attributes attributes) throws SAXException {
        if ("inlineStr".equals(name) || "v".equals(name)) {
            vIsOpen = true;
            // Clear contents cache
            lastContents = "";
        }
        // c => cell
        else if ("c".equals(name)) {
            // 前一个单元格的位置
            // 当前单元格的位置
            ref = attributes.getValue("r");// 设定单元格类型
            this.setNextDataType(attributes);
            // Figure out if the value is an index in the SST
            String cellType = attributes.getValue("t");
            if (cellType != null && cellType.equals("s")) {
                nextIsString = true;
            } else {
                nextIsString = false;
            }
        }
        //当元素为t时
        if("t".equals(name)){
            isTElement = true;
        } else {
            isTElement = false;
        }
        
        // 置空
        lastContents = "";
    }

 
    public void endElement(String uri, String localName, String name)
            throws SAXException {
        if(isTElement){
            String value = lastContents.trim();
            rowlist.add(curCol, value);
            curCol++;
            isTElement = false;
            if(ref!=null&&preRef!=null)
            {
                preRef = ref;    
            } 
            // v => 单元格的值，如果单元格是字符串则v标签的值为该字符串在SST中的索引
            // 将单元格内容加入rowlist中，在这之前先去掉字符串前后的空白符
        } else if ("v".equals(name)) {
               // v => 单元格的值，如果单元格是字符串则v标签的值为该字符串在SST中的索引
            
            String value = this.getDataValue(lastContents.trim(), "");
        
            if (preRef==null) {
                if(this.maxRef!=null&&this.getRowIndex(maxRef)>0)
                {
                    char index_row = (char) (\'A\'+minIndex);  
                    preRef=""+index_row+(curRow+1);
                    if(!ref.equals(preRef))
                    {
                         // 补全单元格之间的空单元格
                        int len = calNullCellCount(ref, preRef,false);
                        for (int i = 0; i <=len; i++) {
                            rowlist.add(curCol, "");
                            curCol++;
                        }
                    }
                }
            }
            else {
                if(!ref.equals(preRef)){
                    int len = calNullCellCount(ref, preRef,false);
                    for (int i = 0; i <len; i++) {
                        rowlist.add(curCol, "");
                        curCol++;
                    }
                }
            }
            rowlist.add(curCol, value);
            curCol++;
            preRef = ref;

        }else {
            //如果标签名称为 row ，这说明已到行尾，调用 optRows() 方法
            if (name.equals("row")) {
                // 默认第一行为表头，以该行单元格数目为最大数目
                if (curRow == 0) {
                    maxRef = ref;
                    minIndex=this.getRowIndex(ref)-rowlist.size();
                }
                // 补全一行尾部可能缺失的单元格
                if (maxRef != null) {
                    int len = calNullCellCount(maxRef, ref,true);
                    for (int i = 0; i <len; i++) {
                        rowlist.add(curCol, "");
                        curCol++;
                    }
                }
                rowReader.getRows(sheetIndex,curRow,rowlist);
                rowlist.clear();
                curRow++;
                curCol = 0;
                //preCol = 0;  
                 preRef = null;
                ref = null;
                //isRowStartNull=false;
            }
        }
        
    }

    @Override
    public void characters(char[] ch, int start, int length) throws SAXException {
        // 得到单元格内容的值
          if (this.vIsOpen) {
              lastContents += new String(ch, start, length);
          }
        
    }
}