前言
项目进行到新的一个迭代了,赶了1周需求,接口终于处理完了。分享记录下迭代中处理导入、导出、下载模版功能的细节吧。
一、场景
- EasyExcel(阿里)实现Excel数据处理
- 三层表头,第二、三层表头动态
- 数据根据第二、三层表头动态
- 导出的模版动态数据区域,下拉选择
- 导入4的模版,导入业务数据
模版效果:
懂的都懂,没有其他啥好说的,直接上代码吧。
二、使用步骤
1.导出
controller导出方法
/*** 导出排班计划*/@ResourceAction(id = "exportShiftPlan", name = "导出排班计划")@GetMapping("/exportShiftPlan")public void exportShiftPlan(PotAttendShiftPlanQueryRequestVo requestVo, HttpServletResponse response) {try {// 设置响应头response.setContentType("application/vnd.ms-excel");response.setCharacterEncoding("utf-8");String fileName = URLEncoder.encode(DateUtil.format(DateUtil.date(), DatePattern.PURE_DATETIME_PATTERN) + "排班计划导出" + ".xlsx", StandardCharsets.UTF_8).replaceAll("\\+", "%20");response.setHeader("Content-disposition", "attachment;filename=" + fileName);//组织导出数据Map<String, Object> exportDataMap = service.exportShiftPlan(requestVo);List<List<String>> headList = (List<List<String>>) exportDataMap.get("headList");List<List<Object>> dataList = (List<List<Object>>) exportDataMap.get("dataList");ExcelWriterBuilder excelWriterBuilder = EasyExcel.write(response.getOutputStream());excelWriterBuilder.head(headList).sheet("排班计划").doWrite(dataList);} catch (Exception e) {throw new RuntimeException("导出排班计划失败");}}
exportShiftPlan方法实现
@Overridepublic Map<String, Object> exportShiftPlan(PotAttendShiftPlanQueryRequestVo requestVo) {Map<String, Object> dataMap = new HashMap<>();//处理动态列头List<List<String>> headList = new ArrayList<>();//前3列姓名、部门、员工账号List<String> xmHeard = new ArrayList<>();xmHeard.add("姓名");headList.add(xmHeard);List<String> bmHeard = new ArrayList<>();bmHeard.add("部门");headList.add(bmHeard);List<String> yghHeard = new ArrayList<>();yghHeard.add("员工账号");headList.add(yghHeard);//时间处理Date startTime = requestVo.getStartTime();Date endTime = requestVo.getEndTime();List<DateTime> rangeDateList = DateUtil.rangeToList(startTime, endTime, DateField.DAY_OF_MONTH);for (DateTime dateTime : rangeDateList) {List<String> dateHead = new ArrayList<>();//日期dateHead.add(DateUtil.format(dateTime, DatePattern.NORM_DATE_PATTERN));//星期int dayOfWeek = DateUtil.dayOfWeek(dateTime);String weekName = AttendConstant.weekDict.get(String.valueOf(dayOfWeek)).toString();dateHead.add(weekName);headList.add(dateHead);}//列头dataMap.put("headList", headList);//do your code,循环处理组织一行数据放到list,再放到map//动态列部分数据根据列头时间循环处理 dataMap.put("dataList", dataList);return dataMap;}
2.下载模版
controller方法
/*** 下载排班计划导入模版** @param response 响应*/@ResourceAction(id = "exportShiftPlanTemplate", name = "导出排班计划模版")@GetMapping("/exportShiftPlanTemplate")public void exportShiftPlanTemplate(PotAttendShiftPlanQueryRequestVo requestVo, HttpServletResponse response) {try {// 设置响应头response.setContentType("application/vnd.ms-excel");response.setCharacterEncoding("utf-8");String fileName = BusinessSerialNoUtil.genDateFmtSeqCode(redisService, "", "", 4, CommonConstant.SHIFT_EXPORT_TEMPLATE, "0", CommonConstant.COMPLET_BEFORE, DatePattern.PURE_DATE_PATTERN) + "排班计划模版.xlsx";//防止乱码fileName = URLEncoder.encode(fileName, StandardCharsets.UTF_8);response.setHeader("Content-disposition", "attachment;filename=" + fileName);//组织导出数据Map<String, Object> exportDataMap = service.exportShiftPlanTemplate(requestVo);List<List<String>> headList = (List<List<String>>) exportDataMap.get("headList");List<List<Object>> dataList = (List<List<Object>>) exportDataMap.get("dataList");List<PotAttendShiftDTO> shiftList = (List<PotAttendShiftDTO>) exportDataMap.get("shiftList");ExcelWriterBuilder excelWriterBuilder = EasyExcel.write(response.getOutputStream());//动态设置下拉框excelWriterBuilder.registerWriteHandler(new CustomCellWriteHeightStrategy((short) 30, (short) 30));excelWriterBuilder.registerWriteHandler(new ShiftSheetWriteHandler(shiftList, headList.size(), dataList.size()));ExcelWriterSheetBuilder excelWriterSheetBuilder = excelWriterBuilder.head(headList).sheet("排班计划");excelWriterSheetBuilder.doWrite(dataList);} catch (Exception e) {String errMsg = e.getMessage();if (StringUtils.isBlank(errMsg)) {errMsg = "导出导入模版失败";}throw new RuntimeException(errMsg);}}
实际上就是在导出的基础上增加了一个样式核心拦截、一个数据处理的handler
CustomCellWriteHeightStrategy
自定义单元格高度策略
/*** 自定义单元格高度策略* @author zwmac*/
public class CustomCellWriteHeightStrategy extends AbstractRowHeightStyleStrategy {/*** 默认高度*/private static final Integer DEFAULT_HEIGHT = 30;/*** 头部行高*/private Short headRowHeight = DEFAULT_HEIGHT.shortValue();/*** 内容行高*/private Short contentRowHeight = DEFAULT_HEIGHT.shortValue();public CustomCellWriteHeightStrategy(Short headRowHeight, Short contentRowHeight) {this.headRowHeight = headRowHeight;this.contentRowHeight = contentRowHeight;}@Overrideprotected void setHeadColumnHeight(Row row, int relativeRowIndex) {int maxHeight = autoCalMaxHeight(row);//row.setHeight((short) (maxHeight * DEFAULT_HEIGHT));row.setHeightInPoints(maxHeight * DEFAULT_HEIGHT);}@Overrideprotected void setContentColumnHeight(Row row, int relativeRowIndex) {int maxHeight = autoCalMaxHeight(row);//row.setHeight((short) (maxHeight * DEFAULT_HEIGHT));row.setHeightInPoints(maxHeight * DEFAULT_HEIGHT);}private int autoCalMaxHeight(Row row) {Iterator<Cell> cellIterator = row.cellIterator();if (!cellIterator.hasNext()) {return 1;}// 默认为 1行高度int maxHeight = 1;while (cellIterator.hasNext()) {Cell cell = cellIterator.next();switch (cell.getCellType()) {case STRING:if (cell.getStringCellValue().contains("\n")) {int length = cell.getStringCellValue().split("\n").length;maxHeight = Math.max(maxHeight, length);}break;default:break;}}return maxHeight;}
}
ShiftSheetWriteHandler
下拉框处理,这里用的是隐藏域方式,之前其实分享过另一种处理方式,也提到了这种方式,下拉框处理
/*** 排班sheetHandler** @author zwmac*/
public class ShiftSheetWriteHandler implements SheetWriteHandler {/*** 一行的列数*/private Integer rowCelNum = 0;/*** 行数*/private Integer rowNum = 0;/*** 班次数据*/private List<PotAttendShiftDTO> shiftList;/*public ShiftSheetWriteHandler(List<PotAttendShiftDTO> shiftList, int rowCelNum) {this.shiftList = shiftList;this.rowCelNum = rowCelNum;}@Overridepublic void beforeSheetCreate(WriteWorkbookHolder writeWorkbookHolder, WriteSheetHolder writeSheetHolder) {}@Overridepublic void afterSheetCreate(WriteWorkbookHolder writeWorkbookHolder, WriteSheetHolder writeSheetHolder) {//班次转mapMap<String, String[]> mapDropDown = new HashMap<>();String[] shiftTimeArr = new String[shiftList.size()];for (int i = 0; i < shiftList.size(); i++) {PotAttendShiftDTO shift = shiftList.get(i);String shiftName = shift.getShiftName();List<PotAttendShiftTime> shiftTimeList = shift.getAttendShiftTimeList();StringBuffer sbf = new StringBuffer();if (CollectionUtil.isNotEmpty(shiftTimeList)) {for (int j = 0; j < shiftTimeList.size(); j++) {PotAttendShiftTime shiftTime = shiftTimeList.get(j);String time = DateUtil.format(shiftTime.getStartTime(), "HH:mm") + "-" + DateUtil.format(shiftTime.getEndTime(),"HH:mm");sbf.append(time);if (j != shiftTimeList.size() - 1) {sbf.append("跳");}}}//长度估计是长了,报异常//shiftTimeArr[i] = shiftName + " " + sbf;shiftTimeArr[i] = shiftName;}mapDropDown.put("1", shiftTimeArr);//获取工作簿Sheet sheet = writeSheetHolder.getSheet();///开始设置下拉框DataValidationHelper helper = sheet.getDataValidationHelper();//设置下拉框for (Map.Entry<String, String[]> entry : mapDropDown.entrySet()) {*//*起始行、终止行、起始列、终止列 起始行为1即表示表头不设置**//*//这里设置65535可能又问题,因为这个是excel的最大行数,如果数据量超过这个数,就会报错CellRangeAddressList addressList = new CellRangeAddressList(2, 65535, 3, rowCelNum - 1);*//*设置下拉框数据**//*DataValidationConstraint constraint = helper.createExplicitListConstraint(entry.getValue());DataValidation dataValidation = helper.createValidation(constraint, addressList);//阻止输入非下拉选项的值dataValidation.setErrorStyle(DataValidation.ErrorStyle.STOP);dataValidation.setShowErrorBox(true);dataValidation.setSuppressDropDownArrow(true);dataValidation.createErrorBox("提示", "输入值与单元格定义格式不一致");dataValidation.createPromptBox("填写说明", "填写内容只能为下拉数据集中的类型");sheet.addValidationData(dataValidation);}}*/private Map<Integer, List<String>> selectMap;private int index;private char[] alphabet = new char[]{'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L','M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z'};public ShiftSheetWriteHandler(List<PotAttendShiftDTO> shiftList, int rowCelNum,Integer rowNum) {//转mapMap<Integer, List<String>> shiftMp = transShiftListToMap(shiftList);this.selectMap = shiftMp;this.index = 0;this.rowCelNum = rowCelNum;this.rowNum = rowNum;}private Map<Integer, List<String>> transShiftListToMap(List<PotAttendShiftDTO> shiftList) {Map<Integer, List<String>> mapDropDown = new HashMap<>();List<String> shiftTimeArr = new ArrayList<>();for (int i = 0; i < shiftList.size(); i++) {PotAttendShiftDTO shift = shiftList.get(i);String shiftName = shift.getShiftName();List<PotAttendShiftTime> shiftTimeList = shift.getAttendShiftTimeList();StringBuffer sbf = new StringBuffer();if (CollectionUtil.isNotEmpty(shiftTimeList)) {for (int j = 0; j < shiftTimeList.size(); j++) {PotAttendShiftTime shiftTime = shiftTimeList.get(j);String time = DateUtil.format(shiftTime.getStartTime(), "HH:mm") + "-" + DateUtil.format(shiftTime.getEndTime(), "HH:mm");sbf.append(time);if (j != shiftTimeList.size() - 1) {sbf.append("跳");}}}String fullName = DictTranslateor.translate(AttendConstant.SHIFT_NAME_DICT, shiftName);shiftTimeArr.add(fullName + " " + sbf);}mapDropDown.put(1, shiftTimeArr);return mapDropDown;}@Overridepublic void afterSheetCreate(WriteWorkbookHolder writeWorkbookHolder, WriteSheetHolder writeSheetHolder) {if (selectMap == null || selectMap.size() == 0) {return;}// 需要设置下拉框的sheet页Sheet curSheet = writeSheetHolder.getSheet();DataValidationHelper helper = curSheet.getDataValidationHelper();String dictSheetName = "字典sheet";Workbook workbook = writeWorkbookHolder.getWorkbook();// 数据字典的sheet页Sheet dictSheet = workbook.createSheet(dictSheetName);// 从第二个工作簿开始隐藏,为了用户的友好性,将字典sheet隐藏掉this.index++;// 设置隐藏workbook.setSheetHidden(this.index, true);for (Map.Entry<Integer, List<String>> entry : selectMap.entrySet()) {// 设置下拉单元格的首行、末行、首列、末列CellRangeAddressList rangeAddressList = new CellRangeAddressList(3, 3 + rowNum - 1, 3, rowCelNum - 1);int rowLen = entry.getValue().size();// 设置字典sheet页的值 每一列一个字典项for (int i = 0; i < rowLen; i++) {Row row = dictSheet.getRow(i);if (row == null) {row = dictSheet.createRow(i);}row.createCell(entry.getKey()).setCellValue(entry.getValue().get(i));}String excelColumn = getExcelColumn(entry.getKey());// 下拉框数据来源 eg:字典sheet!$B1:$B2String refers = dictSheetName + "!$" + excelColumn + "$1:$" + excelColumn + "$" + rowLen;// 创建可被其他单元格引用的名称Name name = workbook.createName();// 设置名称的名字name.setNameName("dict" + entry.getKey());// 设置公式name.setRefersToFormula(refers);// 设置引用约束DataValidationConstraint constraint = helper.createFormulaListConstraint("dict" + entry.getKey());// 设置约束DataValidation validation = helper.createValidation(constraint, rangeAddressList);if (validation instanceof HSSFDataValidation) {validation.setSuppressDropDownArrow(false);} else {validation.setSuppressDropDownArrow(true);validation.setShowErrorBox(true);}// 阻止输入非下拉框的值validation.setErrorStyle(DataValidation.ErrorStyle.STOP);validation.createErrorBox("提示", "此值与单元格定义格式不一致!");validation.setShowErrorBox(true);validation.setSuppressDropDownArrow(true);validation.createPromptBox("填写说明", "填写内容只能为下拉数据集中的类型");// 添加下拉框约束writeSheetHolder.getSheet().addValidationData(validation);}}/*** 将数字列转化成为字母列** @param num* @return*/private String getExcelColumn(int num) {String column = "";int len = alphabet.length - 1;int first = num / len;int second = num % len;if (num <= len) {column = alphabet[num] + "";} else {column = alphabet[first - 1] + "";if (second == 0) {column = column + alphabet[len] + "";} else {column = column + alphabet[second - 1] + "";}}return column;}
}
service实现部分就不贴了,跟导出的思路一致,多了一个下拉框数据shiftList而已。
3、导入
controller方法
/*** 导入排班计划2** @param file 文件*/@OperateLog(moduleName = moduleName, operateContent = "导入排班计划")@RequestMapping(value = "/importShiftPlan/{parkId}", method = RequestMethod.POST)public RestResponse importShiftPlan2(@RequestParam("file") MultipartFile file, @PathVariable("parkId") Long parkId) {try {service.importShiftPlan(file, parkId);return RestResponse.success();} catch (Exception e) {log.error("导入排班计划失败", e);throw new RuntimeException("导入排班计划失败");}}
importShiftPlan方法实现
@Overridepublic void importShiftPlan(MultipartFile file, Long parkId) throws IOException {//参数校验Assert.notNull(file, "文件为空");Assert.notNull(parkId, "园区id为空");//解析文件//EasyExcel.read(file.getInputStream(), AttendShiftPlanExcelDto.class, new AttendShiftPlanListener()).sheet(0).doRead();//List<Map<Integer,Object>> list = EasyExcel.read(file.getInputStream(), AttendShiftPlanExcelDto.class, new AttendShiftPlanListener()).sheet(0).doReadSync();//headRowNumber设置表示设置的行与之前的行都是列头,监听里的数据处理就从这个行之后开始读AttendShiftPlanListener attendShiftPlanListener = new AttendShiftPlanListener();List<Map<Integer, Object>> list = EasyExcel.read(file.getInputStream(), attendShiftPlanListener).sheet(0).headRowNumber(3).doReadSync();List<String> headList = attendShiftPlanListener.getHeaders();Assert.notEmpty(list, "导入数据为空");Assert.notEmpty(headList, "表头信息为空");//do your code//有表头了,有数据了,剩下的就是大家表演了}
AttendShiftPlanListener
监听处理数据,其实保存数据可以在监听里就处理了。
/*** 排班计划导入监听器** @author zwmac*/
@Slf4j
public class AttendShiftPlanListener extends AnalysisEventListener<Map<Integer, String>> {private static final int BATCH_COUNT = 5;private List<AttendShiftPlanExcelDto> cachedDataList = ListUtils.newArrayListWithExpectedSize(BATCH_COUNT);@Getterprivate List<String> headers = new ArrayList<>();// 存储自定义表头Map<Integer, String> customHeaders = new HashMap<>();@Overridepublic void onException(Exception exception, AnalysisContext context) throws Exception {super.onException(exception, context);}@Overridepublic void invokeHead(Map<Integer, ReadCellData<?>> headMap, AnalysisContext context) {//列头处理Integer rowIndex = context.readRowHolder().getRowIndex();if(rowIndex == 1) {for (int i = 3; i < headMap.size(); i++) {String header = headMap.get(i).getStringValue();if (!"null".equals(header) && !isFixedHeader(header)) {customHeaders.put(i, header);}}Map<Integer, String> hdMap = ConverterUtils.convertToStringMap(headMap, context);for (int i = 0; i < hdMap.size(); i++) {headers.add(hdMap.get(i));}}}/*** 判断是否是固定表头*/private boolean isFixedHeader(String header) {//定义固定表头集合List<String> list = Arrays.asList("姓名", "部门", "账号");return list.contains(header);}@Overridepublic void invoke(Map<Integer, String> data, AnalysisContext context) {//数据处理AttendShiftPlanExcelDto shiftPlanExcelDto = new AttendShiftPlanExcelDto();StringBuilder sb = new StringBuilder();shiftPlanExcelDto.setUserName(data.get(0));shiftPlanExcelDto.setOrgName(data.get(1));shiftPlanExcelDto.setUserAccount(data.get(2));JSONObject jsonObject = new JSONObject();for (int i = 0; i < data.size() - 1; i++) {jsonObject.put(customHeaders.get(i), data.get(i));}shiftPlanExcelDto.setShiftPlanDataJson(jsonObject.toJSONString());cachedDataList.add(shiftPlanExcelDto);if (cachedDataList.size() >= BATCH_COUNT) {//saveData();其实这里就可以直接处理保存cachedDataList.clear();} else {log.info("解析到一条数据:{}", data);}}@Overridepublic void doAfterAllAnalysed(AnalysisContext context) {}
}
AttendShiftPlanExcelDto
import com.alibaba.excel.annotation.ExcelIgnore;
import com.alibaba.excel.annotation.ExcelProperty;
import lombok.Data;/*** @author zwmac*/
@Data
public class AttendShiftPlanExcelDto {@ExcelProperty(value = "姓名", index = 0)private String userName;@ExcelProperty(value = "部门", index = 1)private String orgName;@ExcelProperty(value = "账号", index = 2)private String userAccount;/*** 动态数据列*/@ExcelIgnoreprivate String shiftPlanDataJson;
}
前3列是固定的,后面是动态,当成一个字段,实际是json串
总结
- EasyExcel阿里系的还是挺强,目前我们的微服务也是阿里系的(其实我个人不太喜欢阿里系的微服务)
- 导出可以使用模版,这里横向、纵向都是动态,感觉太麻烦,还不如编码
- 其实主要一点还是要自己有大致的思路,自己要敢想
还是那句话,思路很重要,就分享到这里,希望能帮到大家!uping!