java实现解析pdf格式发票

为了减少用户工作量及误操作的可能性，需要实现用户上传PDF格式的发票，系统通过解析PDF文件获取发票内容，并直接将其写入表单。以下文章记录了功能实现的代码。

发票样式

发票内容解析

引用Maven

使用pdfbox

<dependency><groupId>org.apache.pdfbox</groupId><artifactId>pdfbox</artifactId><version>2.0.24</version> <!-- 请检查最新版本 -->
</dependency>

获取PDF内容

设置 sortByPosition 为 true 可以按文本位置提取内容，否则获取到的内容错乱，无法获取到真正需要的内容

    @RequestMapping("uploadReceiptsTest")@ResponseBodypublic Map<String,String> uploadReceiptsTest() throws Exception{String filePath = "D:/apache-tomcat-8.5.98/webapps/tspspic/发票文件/24372000000145100092.pdf"; // 保存路径PDDocument document = PDDocument.load(new File(filePath));PDFTextStripper pdfStripper = new PDFTextStripper();// 排序文本行按其位置pdfStripper.setSortByPosition(true);String text = pdfStripper.getText(document);document.close();Map<String, String> map = pdfStr(text);return map;}

文本内容(部分内容以*替代)

public static void main(String[] args) {String invoiceInfo = "电子发票（普通发票） 发票号码：******\n" +"开票日期：******\n" +"购 名称：****** 销 名称：******\n" +"买 售\n" +"方 方\n" +"信 统一社会信用代码/纳税人识别号：****** 信 统一社会信用代码/纳税人识别号：******\n" +"息 息\n" +"项目名称 规格型号 单 位 数 量 单 价 金 额 税率/征收率 税 额\n" +"******* 100ml：5g 袋 800 4.070796 3256.64 13% 423.36\n" +"******\n" +"合 计 ¥3256.64 ¥423.36\n" +"价税合计（大写） 叁仟陆佰捌拾圆整 （小写）¥3680.00\n" +"批号：******/ 生产日期：2024-05-15/ 有效期至:2026-04-30/ 含税单价：4.6000/ 生产厂家:******/ 批准文号:******/\n" +"备 注\n" +"开票人：王宁\n" +"王宁";

解析文件内容，返回数据

初版测试

    public Map<String,String> pdfStr(String invoiceInfo) {//因解析出的括号不确定为中文还是英文，统一替换为英文字符invoiceInfo = invoiceInfo.replaceAll("（","(").replaceAll("）",")");// 定义正则表达式模式Pattern patternInvoiceNumber = Pattern.compile("发票号码：(\\d+)");Pattern patternInvoiceDate = Pattern.compile("开票日期：(\\d{4}年\\d{1,2}月\\d{1,2}日)");Pattern patternBuyerName = Pattern.compile("购 名称：(.+?) 销 名称：(.+?)\n");//由上图可以发现“项目名称 规格型号 单 位 数 量 单 价 金 额 税率/征收率 税 额”所需要的内容在下一行数据，使用笨方法直接获取“税额与合计”之间的数据通过之后的空格分割进行获取数据Pattern patternItemDetails = Pattern.compile("税 额\\s+(.*?)合 计", Pattern.DOTALL);Pattern patternTotal = Pattern.compile("\\(小写\\)¥(\\d+(\\.\\d+)?)");Pattern patternBatchNumber = Pattern.compile("批号：(.+?)/");Pattern patternProductionDate = Pattern.compile("生产日期：(\\d{4}-\\d{1,2}-\\d{1,2})/");Pattern patternExpirationDate = Pattern.compile("有效期至:(\\d{4}-\\d{1,2}-\\d{1,2})/");Pattern patternTaxIncludedPrice = Pattern.compile("含税单价：(\\d+(\\.\\d+)?)");Pattern patternManufacturer = Pattern.compile("生产厂家:(.+?)/");Pattern patternApprovalNumber = Pattern.compile("批准文号:(.+?)/");Pattern patternIssuer = Pattern.compile("开票人：(.+)");// 创建Matcher对象Matcher matcherInvoiceNumber = patternInvoiceNumber.matcher(invoiceInfo);Matcher matcherInvoiceDate = patternInvoiceDate.matcher(invoiceInfo);Matcher matcherBuyerName = patternBuyerName.matcher(invoiceInfo);Matcher matcherItemDetails = patternItemDetails.matcher(invoiceInfo);Matcher matcherTotal = patternTotal.matcher(invoiceInfo);Matcher matcherBatchNumber = patternBatchNumber.matcher(invoiceInfo);Matcher matcherProductionDate = patternProductionDate.matcher(invoiceInfo);Matcher matcherExpirationDate = patternExpirationDate.matcher(invoiceInfo);Matcher matcherTaxIncludedPrice = patternTaxIncludedPrice.matcher(invoiceInfo);Matcher matcherManufacturer = patternManufacturer.matcher(invoiceInfo);Matcher matcherApprovalNumber = patternApprovalNumber.matcher(invoiceInfo);Matcher matcherIssuer = patternIssuer.matcher(invoiceInfo);// 提取数据String invoiceNumber = "";String invoiceDate = "";String buyerName = "";String sellerName = "";String productName = "";String specification = "";String unit = "";int quantity = 0;double unitPrice = 0.0;double amount = 0.0;String taxRate = "";double taxAmount = 0.0;double total = 0.0;String batchNumber = "";String productionDate = "";String expirationDate = "";double taxIncludedPrice = 0.0;String manufacturer = "";String approvalNumber = "";String issuer = "";if (matcherInvoiceNumber.find()) {invoiceNumber = matcherInvoiceNumber.group(1);}if (matcherInvoiceDate.find()) {invoiceDate = matcherInvoiceDate.group(1);}if (matcherBuyerName.find()) {buyerName = matcherBuyerName.group(1);sellerName = matcherBuyerName.group(2);}// 处理项目名称、规格型号、单位、数量、单价、金额、税率/征收率、税额if (matcherItemDetails.find()) {String itemDetailsLine = matcherItemDetails.group(1).trim();itemDetailsLine = itemDetailsLine.replace("\n"," ");String[] details = itemDetailsLine.split(" "); // 按空格分割if (details.length >= 8) { // 确保有足够的字段//因部分名称过长，换行数据解析到最后进行拼接productName = details[0].trim(); // 项目名称if (details.length >= 9){productName = details[0].trim()+details[8].trim(); // 项目名称}specification = details[1].trim(); // 规格型号unit = details[2].trim(); // 单位quantity = Integer.parseInt(details[3].trim()); // 数量unitPrice = Double.parseDouble(details[4].trim()); // 单价amount = Double.parseDouble(details[5].trim()); // 金额taxRate = details[6].trim(); // 税率/征收率taxAmount = Double.parseDouble(details[7].trim()); // 税额System.out.println("项目名称: " + productName);System.out.println("规格型号: " + specification);System.out.println("单位: " + unit);System.out.println("数量: " + quantity);System.out.println("单价: " + unitPrice);System.out.println("金额: " + amount);System.out.println("税率/征收率: " + taxRate);System.out.println("税额: " + taxAmount);}}if (matcherTotal.find()) {total = Double.parseDouble(matcherTotal.group(1));}if (matcherBatchNumber.find()) {batchNumber = matcherBatchNumber.group(1);}if (matcherProductionDate.find()) {productionDate = matcherProductionDate.group(1);}if (matcherExpirationDate.find()) {expirationDate = matcherExpirationDate.group(1);}if (matcherTaxIncludedPrice.find()) {taxIncludedPrice = Double.parseDouble(matcherTaxIncludedPrice.group(1));}if (matcherManufacturer.find()) {manufacturer = matcherManufacturer.group(1);}if (matcherApprovalNumber.find()) {approvalNumber = matcherApprovalNumber.group(1);}if (matcherIssuer.find()) {issuer = matcherIssuer.group(1);}// 输出其他结果System.out.println("发票号码: " + invoiceNumber);System.out.println("开票日期: " + invoiceDate);System.out.println("购买方名称: " + buyerName);System.out.println("销售方名称: " + sellerName);System.out.println("价税合计: " + total);System.out.println("批号: " + batchNumber);System.out.println("生产日期: " + productionDate);System.out.println("有效期至: " + expirationDate);System.out.println("含税单价: " + taxIncludedPrice);System.out.println("生产厂家: " + manufacturer);System.out.println("批准文号: " + approvalNumber);System.out.println("开票人: " + issuer);}

优化代码

• Map存储正则表达式：将所有正则表达式模式和对应的字段名称存储在一个Map中，遍历Map并执行匹配，从而避免了为每个字段都写单独的匹配代码。

• 抽取通用逻辑：将匹配逻辑抽象成一个通用方法，简化了代码结构，减少了重复代码。

• 处理商品详情：在匹配完itemDetails后，再拆分字符串并填充对应的字段。

    public static Map<String, String> pdfStr(String invoiceInfo) {invoiceInfo = invoiceInfo.replaceAll("（", "(").replaceAll("）", ")");// 定义正则表达式模式Map<String, String> patterns = new HashMap<>();patterns.put("invoiceNumber", "发票号码：(\\d+)");patterns.put("invoiceDate", "开票日期：(\\d{4}年\\d{1,2}月\\d{1,2}日)");patterns.put("buyerName", "购 名称：(.+?) 销 名称：(.+?)\n");patterns.put("itemDetails", "税 额\\s+(.*?)合 计");patterns.put("total", "\\(小写\\)¥(\\d+(\\.\\d+)?)");patterns.put("batchNumber", "批号：(.+?)/");patterns.put("productionDate", "生产日期：(\\d{4}-\\d{1,2}-\\d{1,2})/");patterns.put("expirationDate", "有效期至:(\\d{4}-\\d{1,2}-\\d{1,2})/");patterns.put("taxIncludedPrice", "含税单价：(\\d+(\\.\\d+)?)");patterns.put("manufacturer", "生产厂家:(.+?)/");patterns.put("approvalNumber", "批准文号:(.+?)/");patterns.put("issuer", "开票人：(.+)");// 提取数据Map<String, String> result = new HashMap<>();for (Map.Entry<String, String> entry : patterns.entrySet()) {Pattern pattern = Pattern.compile(entry.getValue(), Pattern.DOTALL);Matcher matcher = pattern.matcher(invoiceInfo);if (matcher.find()) {result.put(entry.getKey(), matcher.group(1).trim());}}// 处理项目名称、规格型号、单位、数量、单价、金额、税率/征收率、税额if (result.containsKey("itemDetails")) {String[] details = result.get("itemDetails").replace("\n", " ").split(" ");if (details.length >= 8) {result.put("productName", details[0].trim() + (details.length > 8 ? details[8].trim() : ""));result.put("specification", details[1].trim());result.put("unit", details[2].trim());result.put("quantity", details[3].trim());result.put("unitPrice", details[4].trim());result.put("amount", details[5].trim());result.put("taxRate", details[6].trim());result.put("taxAmount", details[7].trim());}}// 打印结果for (Map.Entry<String, String> entry : result.entrySet()) {System.out.println(entry.getKey() + ": " + entry.getValue());}return result;}