档案识别

blade-service/blade-archive/src/main/java/org/springblade/archive/service/impl/ArchivesAutoServiceImpl.java

@@ -106,6 +106,7 @@ import java.io.*;
 import java.math.BigDecimal;
 import java.net.URL;
 import java.net.URLEncoder;
+import java.nio.charset.StandardCharsets;
 import java.nio.file.Files;
 import java.nio.file.Path;
 import java.nio.file.Paths;
@@ -5381,9 +5382,8 @@ public class ArchivesAutoServiceImpl extends BaseServiceImpl<ArchivesAutoMapper,
 	@Override
 	@Async
 	public Boolean atuoOCR(List<Long> idsList) throws Exception {
-		String url="/mnt/sdc/AutoPdf/";
-		//String url="D:\\AutoPdf\\";
-		//List<Long> idsList=Func.toLongList(ids);
+		//String url="/mnt/sdc/AutoPdf/";
+		String url="D:\\AutoPdf\\";
 		List<ArchivesAuto> archivesAutoList = this.list(new LambdaQueryWrapper<ArchivesAuto>().in(ArchivesAuto::getId, idsList));
 		this.update(Wrappers.<ArchivesAuto>lambdaUpdate().set(ArchivesAuto::getColourStatus, 2).in(ArchivesAuto::getId, idsList));
 		for (ArchivesAuto auto : archivesAutoList) {
@@ -5468,66 +5468,160 @@ public class ArchivesAutoServiceImpl extends BaseServiceImpl<ArchivesAutoMapper,
 		return new LocalDateTime[]{startDateTime, endDateTime};
 	}
 
-	public List<String> extractTextFromPDF(String pdfFilePath) throws IOException, InterruptedException {
-		//String PYTHON_SCRIPT_PATH = "C:\\Users\\hc01\\AppData\\Local\\Programs\\Python\\Python310\\Python\\pdfTextExtractorWindows.py";
-		//String PYTHON_INTERPRETER = "C:\\Users\\hc01\\AppData\\Local\\Programs\\Python\\Python310\\python.exe";
-		System.out.println("进入识别1");
-		String PYTHON_SCRIPT_PATH = "/www/wwwlogs/python/pdfTextExtractorWindows.py";
-		String PYTHON_INTERPRETER = "python3";
-		String[] command = {
-				PYTHON_INTERPRETER,
-				PYTHON_SCRIPT_PATH,
-				pdfFilePath
-		};
-
-		Process process = new ProcessBuilder(command)
-				.redirectErrorStream(true)
-				.start();
-		System.out.println("进入识别2");
-		// 读取Python输出
-		StringBuilder output = new StringBuilder();
-		try (InputStream inputStream = process.getInputStream();
-			 BufferedReader reader = new BufferedReader(new InputStreamReader(inputStream, "UTF-8"))) {
-
-			String line;
-			while ((line = reader.readLine()) != null) {
-				output.append(line);
-			}
-		}
-		System.out.println("进入识别3");
-		int exitCode = process.waitFor();
-		if (exitCode != 0) {
-			System.out.println("进入识别4");
-			throw new RuntimeException("Python脚本执行失败，退出码: " + exitCode + ", 输出: " + output.toString());
-		}
-
-		// -------------------------- 关键修改：提取纯JSON部分 --------------------------
-		String rawOutput = output.toString();
-		// 找到JSON的起始位置（第一个'{'）和结束位置（最后一个'}'）
-		int jsonStart = rawOutput.indexOf('{');
-		int jsonEnd = rawOutput.lastIndexOf('}');
-		System.out.println("进入识别5");
-		if (jsonStart == -1 || jsonEnd == -1 || jsonStart >= jsonEnd) {
-			System.out.println("进入识别6");
-			throw new RuntimeException("无法提取有效的JSON结果，原始输出: " + rawOutput);
-		}
-		// 截取纯JSON字符串
-		String jsonStr = rawOutput.substring(jsonStart, jsonEnd + 1);
-		System.out.println("进入识别7");
-		// 解析清理后的JSON
-		Gson gson = new Gson();
-		Type type = new TypeToken<Map<String, Object>>(){}.getType();
-		Map<String, Object> resultMap = gson.fromJson(jsonStr, type);
-
-		if (!"success".equals(resultMap.get("status"))) {
-			System.out.println("进入识别8");
+//	public List<String> extractTextFromPDF(String pdfFilePath) throws IOException, InterruptedException {
+//		//String PYTHON_SCRIPT_PATH = "C:\\Users\\hc01\\AppData\\Local\\Programs\\Python\\Python310\\Python\\pdfTextExtractorWindows.py";
+//		//String PYTHON_INTERPRETER = "C:\\Users\\hc01\\AppData\\Local\\Programs\\Python\\Python310\\python.exe";
+//		String PYTHON_SCRIPT_PATH = "/www/wwwlogs/python/pdfTextExtractorWindows.py";
+//		String PYTHON_INTERPRETER = "python3";
+//		String[] command = {
+//				PYTHON_INTERPRETER,
+//				PYTHON_SCRIPT_PATH,
+//				pdfFilePath
+//		};
+//
+//		Process process = new ProcessBuilder(command)
+//				.redirectErrorStream(true)
+//				.start();
+//		// 读取Python输出
+//		StringBuilder output = new StringBuilder();
+//		try (InputStream inputStream = process.getInputStream();
+//			 BufferedReader reader = new BufferedReader(new InputStreamReader(inputStream, "UTF-8"))) {
+//
+//			String line;
+//			while ((line = reader.readLine()) != null) {
+//				output.append(line);
+//			}
+//		}
+//		int exitCode = process.waitFor();
+//		if (exitCode != 0) {
+//			throw new RuntimeException("Python脚本执行失败，退出码: " + exitCode + ", 输出: " + output.toString());
+//		}
+//
+//		// -------------------------- 关键修改：提取纯JSON部分 --------------------------
+//		String rawOutput = output.toString();
+//		// 找到JSON的起始位置（第一个'{'）和结束位置（最后一个'}'）
+//		int jsonStart = rawOutput.indexOf('{');
+//		int jsonEnd = rawOutput.lastIndexOf('}');
+//		if (jsonStart == -1 || jsonEnd == -1 || jsonStart >= jsonEnd) {
+//			throw new RuntimeException("无法提取有效的JSON结果，原始输出: " + rawOutput);
+//		}
+//		// 截取纯JSON字符串
+//		String jsonStr = rawOutput.substring(jsonStart, jsonEnd + 1);
+//		// 解析清理后的JSON
+//		Gson gson = new Gson();
+//		Type type = new TypeToken<Map<String, Object>>(){}.getType();
+//		Map<String, Object> resultMap = gson.fromJson(jsonStr, type);
+//
+//		if (!"success".equals(resultMap.get("status"))) {
+//			String message = (String) resultMap.get("message");
+//			throw new RuntimeException("处理PDF失败: " + (message != null ? message : "未知错误"));
+//		}
+//		Type listType = new TypeToken<List<String>>(){}.getType();
+//		return gson.fromJson(gson.toJson(resultMap.get("lines")), listType);
+//	}
+public List<String> extractTextFromPDF(String pdfFilePath) throws IOException, InterruptedException {
+	// 修正：atuoOCR.py -> autoOCR.py
+	String PYTHON_SCRIPT_PATH = "C:\\Users\\hc01\\AppData\\Local\\Programs\\Python\\Python310\\Python\\autoOCR.py";
+	String PYTHON_INTERPRETER = "C:\\Users\\hc01\\AppData\\Local\\Programs\\Python\\Python310\\python.exe";
+	// 检查文件是否存在
+	File scriptFile = new File(PYTHON_SCRIPT_PATH);
+	if (!scriptFile.exists()) {
+		throw new RuntimeException("Python脚本不存在。请检查路径: " + PYTHON_SCRIPT_PATH);
+	}
+	System.out.println("使用Python解释器: " + PYTHON_INTERPRETER);
+	System.out.println("使用Python脚本: " + PYTHON_SCRIPT_PATH);
+
+	String[] command = {
+			PYTHON_INTERPRETER,
+			PYTHON_SCRIPT_PATH,
+			pdfFilePath
+	};
+	// 打印执行的命令用于调试
+	System.out.print("执行的命令: ");
+	for (String cmd : command) {
+		System.out.print(cmd + " ");
+	}
+	System.out.println();
+
+	ProcessBuilder processBuilder = new ProcessBuilder(command);
+	processBuilder.redirectErrorStream(true);
+
+	// 设置工作目录为Python脚本所在目录
+	File scriptDir = new File(PYTHON_SCRIPT_PATH).getParentFile();
+	processBuilder.directory(scriptDir);
+
+	System.out.println("工作目录: " + scriptDir.getAbsolutePath());
+
+	Process process = processBuilder.start();
+
+	// 读取输出
+	StringBuilder output = new StringBuilder();
+	try (BufferedReader reader = new BufferedReader(
+			new InputStreamReader(process.getInputStream(), StandardCharsets.UTF_8))) {
+		String line;
+		while ((line = reader.readLine()) != null) {
+			output.append(line).append("\n");
+		}
+	}
+
+	int exitCode = process.waitFor();
+
+	System.out.println("Python脚本执行完成，退出码: " + exitCode);
+	System.out.println("原始输出: \n" + output.toString());
+
+	if (exitCode != 0) {
+		// 尝试解析错误信息
+		String errorMsg = "Python脚本执行失败，退出码: " + exitCode;
+
+		// 检查是否是常见错误
+		if (output.toString().contains("No such file or directory")) {
+			errorMsg = "Python脚本文件未找到，请检查路径: " + PYTHON_SCRIPT_PATH;
+		} else if (output.toString().contains("ImportError") || output.toString().contains("ModuleNotFoundError")) {
+			errorMsg = "Python模块未安装。请运行: pip install PyMuPDF opencv-python paddleocr paddlepaddle numpy";
+		}
+
+		throw new RuntimeException(errorMsg);
+	}
+
+	// 解析JSON结果
+	Gson gson = new Gson();
+	String jsonStr = output.toString().trim();
+
+	if (jsonStr.isEmpty()) {
+		throw new RuntimeException("Python脚本没有返回任何内容");
+	}
+
+	try {
+		Map<String, Object> resultMap = gson.fromJson(jsonStr,
+				new TypeToken<Map<String, Object>>(){}.getType());
+
+		if (resultMap == null) {
+			throw new RuntimeException("无法解析JSON结果: " + jsonStr);
+		}
+
+		String status = (String) resultMap.get("status");
+		if (!"success".equals(status)) {
 			String message = (String) resultMap.get("message");
-			throw new RuntimeException("处理PDF失败: " + (message != null ? message : "未知错误"));
+			throw new RuntimeException("处理失败: " + (message != null ? message : "未知错误"));
+		}
+
+		// 提取文本行
+		List<String> lines = new ArrayList<>();
+		List<?> linesData = (List<?>) resultMap.get("lines");
+		if (linesData != null) {
+			for (Object lineObj : linesData) {
+				if (lineObj != null && !lineObj.toString().trim().isEmpty()) {
+					lines.add(lineObj.toString().trim());
+				}
+			}
 		}
-		System.out.println("进入识别9");
-		Type listType = new TypeToken<List<String>>(){}.getType();
-		return gson.fromJson(gson.toJson(resultMap.get("lines")), listType);
+
+		return lines;
+
+	} catch (Exception e) {
+		throw new RuntimeException("解析结果失败: " + e.getMessage() + "\n原始输出: " + jsonStr);
 	}
+}
 
 	@Scheduled(fixedDelay = 1000 * 60 * 10)
 	public void reCreateArchiveAuto() {