如何在nodejs中使这个子进程调用同步？

Question

为什么我会收到此错误localhost didn’t send any data.

即使我将处理后的 csv 字符串从 python 发送到 nodejs。然后在成功提交表单时打印 csv 字符串（这不会发生）。请查看附加的图像，我在控制台中获取了 csv 字符串，但在 csv 文件中没有。

我的 app.js 文件将输入 (csvData, req.body.keywords, req.body.full_search) 发送到 python 脚本以用于处理。

/* csv to json */
const express = require("express"),
  app = express(),
  upload = require("express-fileupload"),
  csvtojson = require("csvtojson");

var http = require('http');
var path = require("path");
var bodyParser = require('body-parser');
var helmet = require('helmet');
var rateLimit = require("express-rate-limit");

let csvData = "test";
app.use(upload());

var server = http.createServer(app);

const limiter = rateLimit({
  windowMs: 15 * 60 * 1000, // 15 minutes
  max: 100 // limit each IP to 100 requests per windowMs
});

// var db = new sqlite3.Database('./database/employees.db');
app.use(bodyParser.urlencoded({extended: false}));
// app.use(express.static(path.join(__dirname,'./Final')));
app.use(helmet());
app.use(limiter);

server.listen(process.env.PORT || 3000, function() { 
    console.log('server running on port 3000'); 
})


app.get('/', function(req, res){
    // res.sendFile(path.join(__dirname,'./index.html'));
    res.sendFile(path.join(__dirname, './index.html'));
});

// Will remove all falsy values: undefined, null, 0, false, NaN and "" (empty string)
function cleanArray(actual) {
  var newArray = new Array();
  for (var i = 0; i < actual.length; i++) {
    if (actual[i]) {
      newArray.push(actual[i]);
    }
  }
  return newArray;
}

function get_array_string(inp){

  var str = "";
  for (var i = 0; i < inp.length; i++){
    str = str + " " + inp[i]
  }
  return str.trim()

}

// form submit request
app.post('/formsubmit', function(req, res){

    // console.log("Form submit request")
    csvData = req.files.csvfile.data.toString('utf8');
    filteredArray = cleanArray(csvData.split(/\r?\n/))
    csvData = get_array_string(filteredArray)

    console.log("URL list received: "+csvData)
    console.log("Search keywords: "+req.body.keywords)
    console.log("Full search: "+req.body.full_search)

    // return csvtojson().fromString(csvData).then(json => 
    //     {return res.status(201).json({csv:csvData, json:json})})

    // Send request to python script
    var spawn = require('child_process').spawn;
    var process = spawn('python', ["./WebExtraction.py", csvData, req.body.keywords, req.body.full_search])

    dataString = "";

    process.stdout.on('data', function(data){

      dataString = dataString + data.toString()

      console.log(dataString)

      res.setHeader('Content-disposition', 'attachment; filename=test.txt');
      res.set('Content-Type', 'text/csv');
      res.status(200).send(dataString);

    });

    process.stdout.on('end', function(){      

    });
    
    process.stdin.end();

});

下面是 python 脚本的一部分，它将 csv 字符串 export_csv 发送到节点 js 应用程序。

if(flag == 1):
    # Get the output string   
    # print("################### - OUTPUT - ############################")
    found_results_A = list(set(found_results))
    found_results = get_list(found_results)
    found_results = list(set(found_results))
    notfound_results = get_list(notfound_results)
    notfound_results = list(set(notfound_results))
    found_keywords_list_changed = []

    for ele in found_results_A:
        found_keywords_list_changed.append(href_key_dict[ele])

    # Get the not found results correctly using set operation
    setA = set(found_results)
    setB = set(notfound_results)
    notfound_results = list(set(setB-setA))
    error_urls = get_list(error_urls)
    error_urls = list(set(error_urls))
    ######################################################################################
    ## CREATING THE FINAL DATA FRAME FOR COLLECTING the extracted urls
    ######################################################################################
    colList = ['Found urls', 'Not found urls','Error urls']
    dframe = pd.DataFrame(columns = colList, dtype = str)

    maxlen = get_max_of_list(found_results_A, found_keywords_list_changed, notfound_results, error_urls)

    found_results_A = append_space(found_results_A, maxlen)
    notfound_results = append_space(notfound_results, maxlen)
    error_urls = append_space(error_urls, maxlen)
    found_keywords_list_changed = append_space(found_keywords_list_changed, maxlen)

    if(len(found_results_A) == maxlen and len(notfound_results) == maxlen and len(error_urls) == maxlen and len(found_keywords_list_changed) == maxlen):    
        dframe['Found urls'] = found_results_A
        dframe['keywords'] = found_keywords_list_changed
        dframe['Not found urls'] = notfound_results
        dframe['Error urls'] = error_urls

    try:    
        dframe = dframe.sort_values(by=["Found urls"], ascending=False)
        data = dframe.dropna(axis=0, how='all', thresh=None, subset=None, inplace=False)
        export_csv = dframe.to_csv(encoding = 'ASCII', index = None, header = True)
        # print(create_json(export_csv.strip()))
        print(export_csv.strip())
        sys.stdout.flush()
    except Exception as err:
        print('Exception occurred, Error on line {}'.format(sys.exc_info()[-1].tb_lineno), type(err).__name__, err)
        print(err)

我想对此进行更改以解决问题但无法继续，请帮助我构建此功能。

process.stdout.on('data', function(data){

      dataString = dataString + data.toString()

      console.log(dataString)

      res.setHeader('Content-disposition', 'attachment; filename=test.txt');
      res.set('Content-Type', 'text/csv');
      res.status(200).send(dataString);

    });

    process.stdout.on('end', function(){      

    });
    
    process.stdin.end();

下面是我遇到的错误

我还可以渲染具有 HTML FORM 的 html 页面

控制台输出

Answer 1

在子进程中发送到stdout 的数据应在父进程的on ("data" 回调中连接，并且仅在调用on("end" 回调时作为HTTP 响应发送回来。重构子处理代码可能看起来像

process.stdout.on('data', function(data){
  dataString = dataString + data.toString()
});

process.stdout.on('end', function(){ 
  console.log(dataString)

  res.setHeader('Content-disposition', 'attachment; filename=test.txt');
  res.set('Content-Type', 'text/csv');
  res.status(200).send(dataString);     

});

process.stdin.end();

如果问题仍未解决，或出现其他问题，请尝试将从子进程接收到的stderr 输出中继到父进程的节点控制台，并使用类似（未经测试）的内容：

process.stderr.on('data', function(data) {
    console.error( "stderr: %s", data);
}

这应该允许您将调试代码放入写入stderr 的python 脚本中。

previous answer（我的）更详细地介绍了如何获取退出代码和写入到stdout 和stderr 的数据（如果需要和帮助）。

[附注dataString 可能尚未声明 - 我在帖子中看不到它的声明。]