1-PDF解析
1.PDF读取解析
使用readPDF工具读取PDF，按页按行解析即可，获取数据逐个匹配塞进去
2.解析策略
/*
** chenle 新加S801研控部门的PDF解析
*/
PdfReader reader = null;
try {
reader = new PdfReader(fileName);
CheckReport checkReport = new CheckReport();
String faceType = "";
/*
** chenle 新加S801研控部门的PDF解析
*/
String isPass = "";
String testType="";
if(StringUtils.equals(site,"S801")){ //S801工厂的PDF解析策略
if(pageNum==0){
pageNum = reader.getNumberOfPages();
}
for(int i = 1;i <= pageNum;i++){
pageContent += PdfTextExtractor.getTextFromPage(reader, i);
}
checkReport.setModelName(modelName);
String[] contentSplit = pageContent.split("\n");
//S801的机种名称与其他的不一样，
// 单独一个循环判断机种是否重复上传
for(int i = 0;i < contentSplit.length;i++) {
String[] split;
String tempStr = contentSplit[i];
if (tempStr == null) {
break;
}
split = tempStr.split(" ");
if(i == 3){
StringBuffer sql_s801 = new StringBuffer();
sql_s801.append("select * from MESEXT_FIRST_CHECK_REPORT where model_name = '"+split[3]+"'");
List<Map<String, Object>> sqlResult_s801 = baseFunctionService.getSqlResult(sql_s801.toString());
if(CollectionUtil.isNotEmpty(sqlResult_s801)){
return OperationResult.buildFailureResult("已解析过的pdf不能再次解析");
}
break;
}
}
//S801的机种名称与其他的不一样，
// 单独一个循环判断机种是否重复上传
for(int i = 0;i < contentSplit.length;i++){
String [] split;
String tempStr = contentSplit[i];
// System.out.printf("第"+i+"行数据为："+tempStr);
if(tempStr == null){
break;
}
split = tempStr.split(" ");
switch (i){
case 1:
if(!split[1].equals("PASS")){
isPass = "NG";
throw new RuntimeException("failure的文件不能上传");
}else{
isPass = split[1];
checkReport.setResult(split[1]);
checkReport.setBOMVersion(split[3]);
checkReport.setPassCount(split[5]);
checkReport.setPassRate(split[7]);
break;
}
case 2:
checkReport.setCheckPerson(split[1]);
checkReport.setCheckTime(split[3]+" "+split[4]);
// checkReport.setTestTime(split[4]);
checkReport.setNoPassCount(split[6]);
break;
case 3:
// checkReport.setTestType(split[3]);
testType = "首件";
checkReport.setTestType(testType);
checkReport.setModelName(split[3]);
checkReport.setNoCheckCount(split[5]);
checkReport.setCheckCount(split[7]);
break;
case 4:
checkReport.setLineBody(split[1]);
checkReport.setOrderCode(split[3]);
checkReport.setProductCode(split[5]);
break;
case 5:
faceType = split[1];
if(StringUtils.equals(faceType,"T")){
checkReport.setABSurface("A");
break;
}
if(StringUtils.equals(faceType,"B")){
checkReport.setABSurface("B");
break;
}
else{
checkReport.setABSurface(faceType);
}
break;
default:
break;
}
}
checkReport.setSite(site);
checkReport.setFileId(fileId);
}
//其他工厂的PDF解析策略
else{
if(pageNum == 0) {
pageNum = reader.getNumberOfPages();
}
for(int i=1;i<=pageNum;i++){
pageContent += PdfTextExtractor.getTextFromPage(reader, i);//读取第i页的文档内容
//System.out.println(pageContent);
}
String shopOrder = modelName.split("-")[1];
String productCode = modelName.split("-")[0];
checkReport.setProductCode(productCode);
String[] contentSplit = pageContent.split("\n");
// 添加面别
//是否PASS或NG
for(int i = 0;i<contentSplit.length;i++){
String[] split;
String tempStr = contentSplit[i];
System.out.println(i+"行数据为："+contentSplit[i]);
if(tempStr == null){
break;
}
if(i==2){
split = tempStr.split(" ");
checkReport.setResult(split[1]);
if(!"PASS".equals(split[1])&&StringUtils.equals(site,"5001")){
throw new RuntimeException("failure的文件不能上传");
}
if(!"PASS".equals(split[1])&&!StringUtils.equals(site,"5001")){
isPass = "NG";
}
testType=split[3];
checkReport.setTestType(split[3]);
}
if(i==3){
split = tempStr.split(" ");
checkReport.setCompany(split[1]);
checkReport.setWorkClass(split[2]);
checkReport.setCheckCount(split[3]);
}
if(i==5){
split = tempStr.split(" ");
int length = split.length;
checkReport.setModelName(modelName);
checkReport.setABSurface(split[length-3]);
faceType = split[length-3];
checkReport.setPassCount(split[length-1]);
}
if(i==6){
split = tempStr.split(" ");
checkReport.setScanTime(split[1]);
checkReport.setNoPassCount(split[split.length-1]);
}
if(i==7){
split = tempStr.split(" ");
checkReport.setProcessTime(split[1]);
checkReport.setOrderCode(shopOrder);
checkReport.setNoCheckCount(split[split.length-1]);
}
if(i==9){
split = tempStr.split(" ");
//文档解析发生了变化 第9行变得不怎么符合规则 分割后数量发生了改变 导致数组下标溢出
// 修复时为了向下兼容 故移除中文字符
//List<String> list = Arrays.asList(split);
ArrayList<String> list = new ArrayList<>(Arrays.asList(split));
list.remove("测值时间:");
list.remove("生产线别:");
list.remove("通过率:");
checkReport.setCheckTime(list.get(0));
checkReport.setLineBody(list.get(1));
checkReport.setPassRate(list.get(2));
}
if(i==10){
split = tempStr.split(" ");
if (split.length==2){
//取11行数据
split =contentSplit[11].split(" ");
}
checkReport.setTestTime(split[1]);
checkReport.setProduceTime(split[3]+" "+split[4]);
if(split.length > 5) {
checkReport.setCheckPerson(split[6]);
}
}
}
checkReport.setSite(site);
checkReport.setFileId(fileId);
if(StringUtils.equals(faceType,"TOP")){
faceType = "A";
}
if(StringUtils.equals(faceType,"BOT")){
faceType = "B";
}
if(StringUtils.equals(faceType,"ALL")){
faceType = "A";
}
if (StringUtils.equals(faceType,"A")){
faceType = "A";
}
if (StringUtils.equals(faceType,"B")){
faceType = "B";
}
}
Map<String, Object> productPlan = getProductPlan(site, checkReport.getLineBody(), checkReport.getOrderCode(), faceType);
String planLogUuid = null;
if (Objects.nonNull(productPlan)){
planLogUuid = (String) productPlan.get("UUID");
}
//NG情况下进行停机
if("NG".equals(isPass)) {
//this.stopMounter(site, checkReport.getLineBody(), checkReport.getOrderCode(), faceType);
}
if(StringUtils.isNotBlank(planLogUuid)&&!StringUtils.equals(testType,"IPQC")){
checkReport.setPlanLogId(planLogUuid);
//双套料生产的工单 允许上传多个首件报告 单套料则清空关联
if (Integer.parseInt((String)productPlan.get("TL")) == 1){
clearOldRelation(planLogUuid);
}else{
// 2024-05-27 双套料工单上传多个首件报告，不删除只解除关联关系
String updateSql = "update mesext_first_check_report set plan_log_id = '' where id = :ID ";
Map<String, Object> params = new HashMap<>();
List<CheckReport> all = checkReportDao.findAllBySiteAndPlanLogIdOrderByCreatedDate(site, planLogUuid);
if (all.size()>1){
String id = all.get(0).getId();
if (Objects.nonNull(id)){
params.put("ID", id);
int update = baseFunctionService.getUpdate(updateSql, params);
// checkReportDao.deleteById(id);
}
}
}
}
if(StringUtils.isBlank(planLogUuid)&&StringUtils.equals(site,"5001")&&!StringUtils.equals(testType,"IPQC")){
throw new RuntimeException("未找到PDF中下达/激活状态的工单");
}
//获取当前线体下达的工单获取其产品编码 与 现有的产品编码对比是否一致
List<Map<String, Object>> results = baseFunctionService.getSqlResultThrowE(
"select a.shop_order,split(b.planned_item_bo,2) item from mesext_production_plan a " +
"left join shop_order b on a.site = b.site and a.shop_order = b.shop_order" +
" where a.site = '" + site + "' and a.line_no = '" + checkReport.getLineBody() + "' and a.status = '10'");
if (CollectionUtil.isEmpty(results)){
// throw new RuntimeException("当前线体"+checkReport.getLineBody()+"不存在下达工单");
}
System.out.printf(String.valueOf(results));
Map<String, Object> map = results.get(0);
Object itemBo = map.get("ITEM");
if (!StringUtils.equals(String.valueOf(itemBo ),checkReport.getProductCode())){
throw new RuntimeException("当前线体"+checkReport.getLineBody()+"下达工单的产品编码为："+itemBo+",与当前上传文件产品编码"+checkReport.getProductCode()+"不符");
}
checkReportDao.save(checkReport);
System.out.printf("解析完成");
} catch (Exception e) {
throw new RuntimeException("解析失败，请检查pdf格式是否正确" + e.getMessage(), e);
}finally {
if(reader != null) {
reader.close();
}
}
3.总结
本次主要进行的是PDF流解析的重写，新增一个工厂的解析方法，难度不大，重点还是熟悉业务流程和数据库，系统的操作。
3.1数据库连接
-
从yaml文件中找到oracle的信息，连接的时候记得修改为端口后面的字段
-
切记只用env和dev，不可使用pro
-
oracl不同于mysql，是以表空间为单位，本次操作的是在meswip中才找到了数据表，mesext_first_check_report存放的是首次报告检测的内容
3.2Git工具使用
-
首先在本地克隆远程master分支
-
new一个本地修改代码的分支：chenle_xxxx
-
本地修改代码
-
上测试机
-
改代码的分支chenle_xxxx进行commit
-
checkout到dev分支，merge分支chenle_xxxx到dev中
-
pull一次远程，然后push
-
-
上主机也是同理，将本地分支merge到master中，切记不要将dev合并到master中
————————————————————————————————————————
如果将dev合并到master将会导致别人的错误bug可能发布到正式机上，引起宕机
————————————————————————————————————————
3.3 8月23日~25日BUG排查日志
出现PDF Header not found的错误
- PDF文件出错，受损等等
- 排查read PDF类和PDFBOX类的影响
- 最终问题出在服务器的问题，上传PDF的服务器和测试的服务器不是同一个，但是数据库一样，会导致无法从服务器下载PDF下来并且进行解析
代码提交后依然无效，出现数组越界问题
- 发布正式机没发布对
- TODOLIST：学习docker指令