Question

嗨所以我正在处理一个从数据库中读取大量数据的文件。然后我在nodejs中使用一个名为json2csv的模块，它转换数据集并将其写入csv文件。这是json2csv模块的链接。 https://www.npmjs.com/package/json2csv 使用这个模块，它将数据转换为csv数据，但我遇到的问题是，当它开始写入文件时，由于数据集太大，它会耗尽内存并崩溃。有什么方法可以写一小块数据而不是写所有数据？这是我到目前为止所拥有的。

    /*require the ibm_db module*/
var ibmdb = require('ibm_db');
var fs    = require("fs");
var json2csv = require('json2csv');
var db2query = "SELECT DISTINCT DISCEVNT.EVENT_RUNID, DISCEVNT.EVENT_TIME, DISCEVNT.EVENT_SEVERITY, DISCEVNT.EVENT_ATTRIBUTE, DISCEVNT.EVENT_DESC, DISCEVNT_ATTR_IP.DE_IP, DISCEVNT_ATTR_IP.DE_HOSTNAME, DISCEVNT_ATTR_IP.DE_FQDN, DISCRUN.SERVERID_X FROM (DB2INST2.DISCEVNT DISCEVNT INNER JOIN DB2INST2.DISCRUN DISCRUN ON (DISCEVNT.EVENT_RUNID = DISCRUN.DISCOVERRUNID_X)) INNER JOIN DB2INST2.DISCEVNT_ATTR_IP DISCEVNT_ATTR_IP ON (DISCEVNT_ATTR_IP.DE_ATTR = DISCEVNT.EVENT_ATTRIBUTE) WHERE (DISCEVNT.EVENT_RUNID BETWEEN 2017071400000000 AND 2017071900000000) ORDER BY DISCEVNT.EVENT_TIME ASC;";
var db2setup = "DRIVER={DB2};DATABASE=CMDB;UID=user;PWD=password!;HOSTNAME=localhost;port=3000";
var dataJson = [];
var fields = ['EVENT_RUNID','EVENT_TIME','EVENT_SEVERITY','EVENT_ATTRIBUTE','EVENT_DESC','DE_IP','DE_HOSTNAME','DE_FQDN','SERVERID_X'];
console.log("Test program to access DB2 sample database");

/*Connect to the database server
  param 1: The DSN string which has the details of database name to connect to, user id, password, hostname, portnumber
  param 2: The Callback function to execute when connection attempt to the specified database is completed
*/
function getData(values){
    dataJson = values;
    console.log(dataJson);
    var csv = json2csv({ data: dataJson, fields: fields });
    console.log("CSV data:");
    console.log(csv[0]);
    console.log(csv.length);
    console.log(dataJson.length);
    console.log(dataJson[0]);
    fs.writeFile('C:\\Users\\ztaddmusr\\Desktop\\CSV FILES\\file.csv', csv, function(err) {
      if (err) throw err;
      console.log('file saved');
    });

  }

ibmdb.open(db2setup, function(err, conn)
{
        if(err) {
        /*
          On error in connection, log the error message on console
        */
            console.error("error: ", err.message);
        } else {

        /*
            On successful connection issue the SQL query by calling the query() function on Database
            param 1: The SQL query to be issued
            param 2: The callback function to execute when the database server responds
        */
        conn.query(db2query, function(err, data) {

        if(err) {
        /*
          On error in connection, log the error message on console
        */
            console.error("error: ", err.message);
        } else {
            //console.log(data);

            getData(data);
            console.log('Data finished retrieving from the datatbase. Now will write to file.')

        }
            /*
                Close the connection to the database
                param 1: The callback function to execute on completion of close function.
            */
            conn.close(function(){
                console.log("Connection Closed");
            });
        });
    }
});
//var csv = json2csv({ data: dataJson, fields: fields });
//console.log(csv);
//fs.writeFile('file.csv', csv, function(err) {
//  if (err) throw err;
//  console.log('file saved');
//});

我一直在尝试使用循环，但一直搞乱。提前感谢您的帮助！

Answer 1

如果将ES6与co（npm包）等实现一起使用，则可以使用生成器。然后，您可以根据分区暂停异步代码的执行，并使用块附加到文件。

像这样的东西（没有编译它;可能有错误）：

function* writeToCsv(csvObj){
  let chunkedArr = chunkify(csvObj);
  chunkedArr.forEach(co.wrap(chunk => {
    co.wrap(writeChunkToCsv(chunk))(dir)  
  }))
}

function* writeChunkToCsv(chunk){
  return dir => {
    yield fs.writeFile(dir, csv);
  }
}

function chunkify(arr, partitions){ 
  let ansArr = [];
  for(let i = 0, currentArr = []; i < arr.length; i++){
    if(i%(arr.length/partitions) === 0 || i === arr.length){
      ansArr = ansArr.push([...currentArr]);
      currentArr = [];
    }
    else{
      currentArr.push(arr[i]);
    }
  }
  return ansArr;
}

如何使用nodeJS将大型数据集写入csv文件？

1 个答案: