返回结果加端口号, 批量运行 exe 脚本

.gitignore

@@ -1,4 +1,6 @@
 Debug/
+node_modules/
+package-lock.json
 Release/
 x64/
 bin/

src/WeChatOcrCpp/Wchtcr.cpp

@@ -64,10 +64,10 @@ std::string Wchtcr::rec(const std::string& pth_img)
         std::atomic<bool> taskCompleted(false);
 
         // 创建 ImageOcr 实例
-        ImageOcr ocr;
+        //ImageOcr ocr;
 
         // 设置回调函数并执行OCR
-        ocr.Run(pth_img, [&taskCompleted, &result](const std::string& imagePath, std::string& jsonstr) {
+        this->ocr.Run(pth_img, [&taskCompleted, &result](const std::string& imagePath, std::string& jsonstr) {
 
             result = jsonstr;
 

src/WeChatOcrCpp/Wchtcr.h

@@ -1,11 +1,15 @@
 #pragma once
 #include <string>
 #include <future>
+#include "Wchtcr.h"
+#include "ImageOcr.h"
 #include "WeChatOcrResult.h"
 #include "windows.h"
 
 class __declspec(dllexport) Wchtcr
 {
+private:
+    ImageOcr ocr;
 public:
     Wchtcr();
 

src/WeChatOcrCpp/WeChatOcrResult.cpp

@@ -16,64 +16,6 @@
 
 #include "utf8.h"
 
-
-
-
-// see huggingface\ColorTextEditorV2\data\json
-
-// Base64解码函数
-//std::string ParseOcrResult::Base64Decode(const std::string& encoded_string) {
-//    if (encoded_string.empty()) return "";
-//    
-//    // Base64编码表
-//    static const std::string base64_chars = 
-//        "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
-//        "abcdefghijklmnopqrstuvwxyz"
-//        "0123456789+/";
-//    
-//    // 解码过程
-//    size_t in_len = encoded_string.size();
-//    size_t i = 0;
-//    size_t j = 0;
-//    size_t in_ = 0;
-//    unsigned char char_array_4[4], char_array_3[3];
-//    std::string ret;
-//    
-//    while (in_len-- && (encoded_string[in_] != '=') && 
-//           (isalnum(encoded_string[in_]) || (encoded_string[in_] == '+') || (encoded_string[in_] == '/'))) {
-//        char_array_4[i++] = encoded_string[in_]; in_++;
-//        if (i == 4) {
-//            for (i = 0; i < 4; i++)
-//                char_array_4[i] = base64_chars.find(char_array_4[i]);
-//            
-//            char_array_3[0] = (char_array_4[0] << 2) + ((char_array_4[1] & 0x30) >> 4);
-//            char_array_3[1] = ((char_array_4[1] & 0xf) << 4) + ((char_array_4[2] & 0x3c) >> 2);
-//            char_array_3[2] = ((char_array_4[2] & 0x3) << 6) + char_array_4[3];
-//            
-//            for (i = 0; i < 3; i++)
-//                ret += char_array_3[i];
-//            i = 0;
-//        }
-//    }
-//    
-//    if (i) {
-//        for (j = i; j < 4; j++)
-//            char_array_4[j] = 0;
-//        
-//        for (j = 0; j < 4; j++)
-//            char_array_4[j] = base64_chars.find(char_array_4[j]);
-//        
-//        char_array_3[0] = (char_array_4[0] << 2) + ((char_array_4[1] & 0x30) >> 4);
-//        char_array_3[1] = ((char_array_4[1] & 0xf) << 4) + ((char_array_4[2] & 0x3c) >> 2);
-//        char_array_3[2] = ((char_array_4[2] & 0x3) << 6) + char_array_4[3];
-//        
-//        for (j = 0; j < i - 1; j++)
-//            ret += char_array_3[j];
-//    }
-//    
-//    return ret;
-//}
-
 // 解析JSON响应
 WeChatOcrResult* ParseOcrResult::ParseJson(const std::string& jsonResponseStr) {
     try {
@@ -240,14 +182,9 @@ std::string OcrResult::ToJson(std::string& pth_img) const {
                     continue;
                 }
 
-                int num_char = 0;
-
-                // todo: 改成一个字一个框
-                // 好像 wechat 的原始结果对同一个单词中的每个字母就是给出相同的 x, y, w, h ？ 但是有时 OneStrUtf8 它给的是不止一个字符，所以后面这里又自已拆一次？  
-				// 虽说是单字结果，但是有可能是一个单词的多个字。这种情况拆出每一个字符，让它们在图片中有相同的坐标和宽高。既是整个单词的大框
+                // 虽说是单字结果，但是有可能是一个单词的多个字。这种情况拆出每一个字符，让它们在图片中有相同的坐标和宽高。既是整个单词的大框
                 for (size_t j = 0; j < oneResult.OneStrUtf8.size();) {
-
-                    size_t len = utf8codepointcalcsize(&oneResult.OneStrUtf8[j]); // 计算字符字节数
+                    size_t len = utf8codepointcalcsize((const utf8_int8_t*)&oneResult.OneStrUtf8[j]); // 计算字符字节数
                     std::string word = oneResult.OneStrUtf8.substr(j, len);
 
                     jsonOneResult["word"] = word;  // 单个字文本
@@ -284,15 +221,44 @@ std::string OcrResult::ToJson(std::string& pth_img) const {
                     //cv::imshow("line", tmt);
                     //cv::waitKey(0);
 
-                    num_char += 1;
                     j += len;
                 }
 
-                if (num_char == 1) {
-                    int a = 1;
-                } else if (num_char > 1) {
-                    int a = 1;
-                }
+
+                //jsonOneResult["word"] = oneResult.OneStrUtf8;  // 单个字文本
+
+				//int one_x = -1, one_y = -1, one_w = -1, one_h = h;
+    //            
+    //            if (oneResult.OnePos) {
+    //                nlohmann::json posArray = nlohmann::json::array();
+				//	for (const auto& pos : oneResult.OnePos->Pos) {  // 理论上应该是有四个角的坐标，但是 wechat ocr 实际上只给出了一个左上角的坐标
+    //                    //posArray.push_back({ {"x", pos.X},{"y", pos.Y} });
+    //                    one_x = pos.X;
+    //                    one_y =  pos.Y;
+    //                    break;
+    //                }
+    //                //jsonOneResult["position"] = posArray;
+    //            }
+
+    //            if (i == singleResult.OneResult.size() - 1) {
+    //                one_w = rht - one_x;
+    //            }
+    //            else {
+				//	one_w =  singleResult.OneResult[i + 1].OnePos->Pos[0].X - one_x;  // 获取下一个单字的左上角坐标
+    //            }
+
+    //            jsonOneResult["x"] = one_x;
+    //            jsonOneResult["y"] = one_y;
+    //            jsonOneResult["w"] = one_w;
+    //            jsonOneResult["h"] = one_h;
+    //            
+    //            oneResultArray.push_back(jsonOneResult);
+
+
+                //cv::Mat tmt = mt.clone();
+                //cv::rectangle(tmt, cv::Rect(one_x, one_y, one_w, one_h), cv::Scalar(0, 0, 255));
+                //cv::imshow("line", tmt);
+                //cv::waitKey(0);
 
             }
             jsonSingleResult["charInfo"] = oneResultArray;

src/WeChatOcrCpp/main.cpp

@@ -22,6 +22,8 @@
 #include <opencv2/videoio.hpp>
 #include <opencv2/imgproc.hpp>
 
+#include <format>
+
 
 
 using json = nlohmann::json;
@@ -56,9 +58,32 @@ extern "C" {
 void StartHttpServer(int port) {
     httplib::Server svr;
 
+    std::atomic<int> totalTasks(0);  // 使用原子类型确保线程安全
+
     // 添加POST接口，接收base64编码的图片
-    svr.Post("/wechatocr", [](const httplib::Request& req, httplib::Response& res) {
-        try {
+    svr.Post("/wechatocr", [&totalTasks, &port](const httplib::Request& req, httplib::Response& res) {
+
+		std::string msg_Warning_waiting = std::vformat("{{\"msg\":\"warning: waiting previous task done. port {}\" }}", std::make_format_args(port));
+		std::string msg_Err_noParam = std::vformat("{{\"msg\":\"error: not have img param. port {}\" }}", std::make_format_args(port));
+		std::string msg_Err_genFilename = std::vformat("{{\"msg\":\"error: gen tmp filename fail. port {}\" }}", std::make_format_args(port));
+		std::string msg_Err_rec_fail = std::vformat("{{\"msg\":\"error: rec fail port {}\" }}", std::make_format_args(port));
+
+         try {
+
+    //        while (totalTasks + 1 > 1) {
+    //            std::cout << "########## Warning: Max tasks overflow, Total tasks " << totalTasks << " waitting now... "  << std::endl;
+				//std::this_thread::sleep_for(std::chrono::milliseconds(1000)); // 等待 1 秒
+    //        }
+
+           
+
+            if (totalTasks > 0) {
+                res.status = 201;
+                res.set_content(msg_Warning_waiting, "application/json");
+                return;
+            }
+
+			std::cout << "********* Received a new task. Total tasks: " << ++totalTasks << std::endl;
 
             // 检查Content-Type
             const auto& content_type = req.get_header_value("Content-Type");
@@ -80,7 +105,10 @@ void StartHttpServer(int port) {
                 base64_image = request_data["img"].get<std::string>();
             }
             else {
-                throw std::runtime_error("请求中缺少img字段");
+                --totalTasks;
+                res.status = 500;
+                res.set_content(msg_Err_noParam, "application/json");
+                return;
             }
 
             std::vector<unsigned char> b64_buf = Utilities::Base64Decode_Raw(base64_image);
@@ -90,7 +118,9 @@ void StartHttpServer(int port) {
             char tempPath[MAX_PATH], tempFile[MAX_PATH];
             GetTempPathA(MAX_PATH, tempPath);              // 获取系统临时目录
             if (GetTempFileNameA(tempPath, "TMP", 0, tempFile) == 0) { // "TMP"为前缀，0表示自动生成唯一编号
-                res.set_content("{\"msg\":\"生成临时文件名失败\" }", "application/json");
+                --totalTasks;
+                res.status = 201;
+                res.set_content(msg_Err_genFilename, "application/json");
                 return;
             }
             strcat_s(tempFile, ".jpg");
@@ -103,61 +133,100 @@ void StartHttpServer(int port) {
 
             // 调用rec函数进行OCR识别
             std::string result = rec(tempFile);
+			if (result.empty()) {
+				--totalTasks;
+				res.status = 201;
+				res.set_content(msg_Err_rec_fail, "application/json");
+				return;
+			}
+
+            json jsn = json::parse(result);
+			jsn["port"] = port; // 添加端口号到结果中
+            result = jsn.dump(4);
+
+            --totalTasks;
             
+            std::cout << "one task done. Total tasks:" << totalTasks << std::endl;
+
             // 返回识别结果
+            res.status = 200;
             res.set_content(result, "application/json");
-            //res.set_content("{\"msg\":\"ok. 正常\" }", "application/json");
 
             // 删除临时文件
             std::remove(tempFile);
+
+            return;
         }
         catch (const std::exception& e) {
-            res.status = 500;
-            res.set_content("{\"error\":\"服务器内部错误: " + std::string(e.what()) + "\"}", "application/json");
+            --totalTasks;
+            res.status = 201;
+            std::string error_msg = e.what();
+			std::string msg_exception = std::vformat("{\"msg\":\"error: server error. {} port {}\" }", std::make_format_args(error_msg, port) );
+            res.set_content(msg_exception, "application/json");
+            return;
         }
+
+        
     });
 
     // 添加一个简单的GET接口，用于测试服务器是否正常运行
     svr.Get("/", [](const httplib::Request&, httplib::Response& res) {
-        res.set_content("OCR服务器正常运行中", "text/plain");
+        res.set_content("OCR Sever running ....", "text/plain");
         });
 
-    std::cout << "HTTP服务器已启动，监听端口: " << port << std::endl;
+    std::cout << "OCR Sever running ... : " << port << std::endl;
     svr.listen("0.0.0.0", port);
 }
 
-
-
-int main()
+int main(int argc, char* argv[])
 {
 	//Wchtcr wchtcr;
  //   
- //   std::string result = wchtcr.rec("0022.jpg");
+ //   std::string result = wchtcr.rec("t2.jpg");
 	//json jsn;
 	//std::string jsn_str;
 
-    // 启动HTTP服务器，监听8080端口
-    std::thread server_thread(StartHttpServer, 7788);
+    // 默认端口号
+    int port = 7788;
+    
+    // 如果提供了命令行参数，使用第一个参数作为端口号
+    if (argc > 1) {
+        try {
+            port = std::stoi(argv[1]);
+            if (port <= 0 || port > 65535) {
+                std::cout << "Invalid port number. Using default port 7788." << std::endl;
+                port = 7788;
+            }
+        } catch (const std::exception& e) {
+            std::cout << "Invalid port number format. Using default port 7788." << std::endl;
+            port = 7788;
+        }
+    }
+
+    // 启动HTTP服务器，使用指定的端口
+    std::thread server_thread(StartHttpServer, port);
 
     // 分离线程，让它在后台运行
     server_thread.detach();
 
-    // 主线程可以做其他事情，或者等待用户输入退出
-    std::cout << "OCR服务器已启动，按Enter键退出..." << std::endl;
-    std::cin.get();
-
- //   std::string result = rec("E:/huggingface/imradv3/data/0022.jpg");
- //   json jsn;
- //   std::string jsn_str;
- //   
- //   try {
-	//	 jsn = json::parse(result);
-	//	 jsn_str = jsn.dump(4);
- //   }
-	//catch (const std::exception& e) {
-	//	std::cerr << "OCR识别异常: " << e.what() << std::endl;
-	//	return 1;
-	//}
+    do {
+        // 主线程可以做其他事情，或者等待用户输入退出
+        //std::cout << "ok." << std::endl;
+        std::cin.get();
+    } while (true);
+
+    std::string result = rec("t2.jpg");
+    json jsn;
+    std::string jsn_str;
+    
+    try {
+		 jsn = json::parse(result);
+		 jsn_str = jsn.dump(4);
+    }
+	catch (const std::exception& e) {
+		std::cerr << "OCR识别异常: " << e.what() << std::endl;
+		return 1;
+	}
 
     return 0;
 }

src/WeChatOcrCpp/ocr_aliyun.js

@@ -1,5 +1,13 @@
 (async () => {
 
+    // New-NetFirewallRule -DisplayName "Allow Port 7788" -Direction Inbound -Protocol TCP -LocalPort 7788 -Action Allow
+
+    async function sleep(ms) {
+        return new Promise((resolve) => {
+            setTimeout(resolve, ms)
+        })
+    }
+
 
     async function ocr(imgData) {
         // const api = 'https://gjbsb.market.alicloudapi.com/ocrservice/advanced';
@@ -12,50 +20,56 @@
         const data = await new Promise((resolve, reject) => {
             request.post({
                 url: api,
-                timeout: 1000 * 60 * 2,
+                timeout: 1000 * 60 * 60 * 24,
                 headers: {
-                "Authorization": `APPCODE ${appCode}`,
-                "Content-Type": "application/json;charset=UTF-8"
+                    "Authorization": `APPCODE ${appCode}`,
+                    "Content-Type": "application/json;charset=UTF-8"
                 },
                 body: JSON.stringify({
-                img: imgData,
-                prob: true,
-                charInfo: true,
-                table: true,
-                sortPage: true,
-                NeedRotate:true
+                    img: imgData,
+                    prob: true,
+                    charInfo: true,
+                    table: true,
+                    sortPage: true,
+                    NeedRotate: true
                 })
             }, (error, response, body) => {
                 if (error) {
                     console.log('#####ERROR: aliyun ocr fail.');
                     console.log(error);
-                    resolve( [null, error] )
+                    resolve([null, error])
 
                     // reject(error);
                 }
                 else {
-                if (response.statusCode != 200) {
-                    return resolve( [null, {"error_code":response.statusCode, "error_msg":body}] )
-                }
-                let aliResult = null;
-                try {
-                    aliResult = JSON.parse(body);
-                } catch (ex) {
-                    console.log(ex.message);
-                    console.log(response.statusCode);
-                    console.log(error);
-                    console.log(body);
-                }
+                    if (response.statusCode != 200) {
+                        console.log('#####ERROR: aliyun ocr fail');
+                        return resolve([null, { "error_code": response.statusCode, "error_msg": body }])
+                    }
+                    let aliResult = null;
+                    try {
+                        aliResult = JSON.parse(body);
+                    } catch (ex) {
+                        console.log('#####ERROR: aliyun ocr fail.');
+                        console.log(ex.message);
+                        console.log(response.statusCode);
+                        console.log(error);
+                        console.log(body);
+                    }
                     resolve([aliResult, null]);
                 }
             })
         });
 
+        if (data[0] != null) {
+            console.log(`one task done.`)
+        }
+
         if (data.error_code !== undefined && data.error_code !== null) {
-            return [ null, {"code":data.error_code, "msg":data.error_msg} ]
+            return [null, { "code": data.error_code, "msg": data.error_msg }]
         }
 
-        return [ data, null ]
+        return [data, null]
 
         // //JSON写入
         // fs.writeFileSync(jsonPath, JSON.stringify(data));
@@ -64,27 +78,32 @@
     }
 
 
-    
-
     let fs = require('fs')
     let md5 = require('md5')
 
-    let bytes = fs.readFileSync("no_think_more.png")  // 1032.jpg 'binary'
+    let bytes = fs.readFileSync("0170.jpg")
     let buf = Buffer.from(bytes)
     let m5 = md5(buf)
     let b64 = buf.toString('base64')
 
-    let [ re, err ] = await ocr(b64)
 
-    let json = {
-        md5: m5,
-        imgData: b64,
-        guid: '73ffc72d-bfce-43bc-9e90-0007605c6fdb',
-        bookNO: 'lrx333',
-        imgName: '1032.jpg',
-        originImgData: b64
+    for (let i = 0; i < 5000; i++) {
+        console.log(`Send ${i}`)
+        ocr(b64);
+        await sleep(300);
     }
 
+    // let [ re, err ] = await ocr(b64)
+
+    // let json = {
+    //     md5: m5,
+    //     imgData: b64,
+    //     guid: '73ffc72d-bfce-43bc-9e90-0007605c6fdb',
+    //     bookNO: 'lrx333',
+    //     imgName: '1032.jpg',
+    //     originImgData: b64
+    // }
+
     // json = {
     //     md5: '7468efae7d7ab7333d0197a8ca1bf32c',
     //     imgData: 'hasTest',
@@ -93,21 +112,21 @@
     //     imgName: '1032.jpg'
     // }
 
-    let bent = require('bent')
-    let formurlencoded = require('form-urlencoded')
+    // let bent = require('bent')
+    // let formurlencoded = require('form-urlencoded')
 
-    let formurlencoded_body = formurlencoded(json)
+    // let formurlencoded_body = formurlencoded(json)
 
-    let post = bent('http://webapi.xfyun.cn', 'POST', 'json', 200)
-    // let post = bent('http://127.0.0.1:11112', 'POST', 'json', 200)
-    let response = await post('/v1/service/v1/ocr/general', json)
+    // let post = bent('http://webapi.xfyun.cn', 'POST', 'json', 200)
+    // // let post = bent('http://127.0.0.1:11112', 'POST', 'json', 200)
+    // let response = await post('/v1/service/v1/ocr/general', json)
 
 
 
-    data = JSON.parse(response).data.test
+    // data = JSON.parse(response).data.test
 
-    let s = JSON.stringify(response)
+    // let s = JSON.stringify(response)
 
-    console.log(response)
+    // console.log(response)
 
 })()

src/WeChatOcrCpp/package.json

@@ -0,0 +1,6 @@
+{
+  "dependencies": {
+    "md5": "^2.3.0",
+    "request": "^2.88.2"
+  }
+}

src/WeChatOcrCpp/start_ocr_jobs.ps1

@@ -0,0 +1,78 @@
+# start_ocr_jobs - 使用后台作业启动多个OCR进程
+# 使用默认参数（端口7789 开始，10个进程）：
+# .\start_ocr_jobs.ps1
+# 自定义参数：
+# .\start_ocr_jobs.ps1 -StartPort 7789 -Count 3
+
+param(
+    [int]$StartPort = 7789,
+    [int]$Count = 2,
+    [string]$ExePath = ".\WeChatOcrCpp.exe"
+)
+
+Write-Host "Starting $Count OCR processes as background jobs from port $StartPort..." -ForegroundColor Yellow
+
+# 清理之前可能存在的OCR作业
+$existingJobs = Get-Job -Name "OCR_*" -ErrorAction SilentlyContinue
+if ($existingJobs) {
+    Write-Host "Cleaning up existing OCR jobs..." -ForegroundColor Yellow
+    $existingJobs | Stop-Job -PassThru | Remove-Job -Force
+}
+
+# 启动新的后台作业
+for ($i = 0; $i -lt $Count; $i++) {
+    $port = $StartPort + $i
+    $jobName = "OCR_$port"
+    
+    try {
+        # 创建后台作业
+        $job = Start-Job -Name $jobName -ScriptBlock {
+            param($exePath, $port)
+            
+            # 检查exe文件是否存在
+            if (-not (Test-Path $exePath)) {
+                throw "Executable not found: $exePath"
+            }
+            
+            # 启动进程
+            $process = Start-Process -FilePath $exePath -ArgumentList $port -PassThru -NoNewWindow
+            
+            # 等待进程结束或被停止
+            $process.WaitForExit()
+            
+            return @{
+                Port = $port
+                ProcessId = $process.Id
+                ExitCode = $process.ExitCode
+                ExitTime = Get-Date
+            }
+        } -ArgumentList $ExePath, $port
+        
+        Write-Host "✓ Started OCR job '$jobName' for port $port (Job ID: $($job.Id))" -ForegroundColor Green
+        Start-Sleep -Milliseconds 2000  # 稍微延迟避免资源冲突
+    }
+    catch {
+        Write-Host "✗ Failed to start job for port $port`: $($_.Exception.Message)" -ForegroundColor Red
+    }
+}
+
+Write-Host "\n=== Job Management Commands ===" -ForegroundColor Cyan
+Write-Host "查看所有OCR作业状态: Get-Job -Name 'OCR_*'"
+Write-Host "查看作业详细信息: Get-Job -Name 'OCR_*' | Format-Table Name, State, HasMoreData"
+Write-Host "接收作业输出: Receive-Job -Name 'OCR_7789' -Keep"
+Write-Host "停止所有OCR作业: Get-Job -Name 'OCR_*' | Stop-Job"
+Write-Host "删除所有OCR作业: Get-Job -Name 'OCR_*' | Remove-Job -Force"
+
+# 显示当前作业状态
+Write-Host "\n=== Current Job Status ===" -ForegroundColor Cyan
+$jobs = Get-Job -Name "OCR_*" -ErrorAction SilentlyContinue
+if ($jobs) {
+    $jobs | Format-Table Name, State, HasMoreData, PSBeginTime -AutoSize
+} else {
+    Write-Host "No OCR jobs found." -ForegroundColor Yellow
+}
+
+Write-Host "\nAll OCR jobs have been started. Use the commands above to manage them." -ForegroundColor Green
+
+
+

src/WeChatOcrCpp/wechatocr_7789.json

@@ -0,0 +1,8 @@
+{
+  "name": "WeChatOCR_7789",
+  "script": "WeChatOcrCpp.exe",
+  "args": ["7789"],
+  "cwd": "E:\\WeChatOCR_Server",
+  "exec_interpreter": "none"
+}
+