Streaming the results of multiple fs.readfiles - node.js

I'm a newbie at node js streams, and what I want to achieve is streaming the results of the readfiles in a module that I have. I want to then somehow invoke this readable stream in my main app and listen to data events, so everytime readfile returns a result a data event will trigger, and the object is passed as a chunk. This is what I've got so far and it's throwing an error...
function streamObjects(type, dirname){
var Readable = require('stream').Readable;
var rs = new Readable({objectMode: true});
fs.readdir(dirname, function(err, files){
if(err)
console.log(err);
for(var i=0;i<10;i++)
{
fs.readFile(path.resolve(dirname, files[i]),function(err,data){
if(err)
console.log(err);
rs.push(JSON.parse(data));
}); //end readFile
} //end for loop
return rs;
});
}

You want to use EventEmitter.
Your readFile is an async function which is not called when you return rs Readable.
var util = require('util');
var EventEmitter = require('events').EventEmitter;
function streamObjects(type, dirname) {
var Readable = require('stream').Readable;
var rs = new Readable({objectMode: true});
var self = this;
fs.readdir(dirname, function (err, files) {
if (err)
console.log(err);
for (var i = 0; i < 10; i++) {
fs.readFile(path.resolve(dirname, files[i]), function (err, data) {
if (err)
console.log(err);
self.emit('data', files[i], data);
rs.push(JSON.parse(data));
}); //end readFile
} //end for loop
});
}
util.inherits(streamObjects, EventEmitter);
module.exports = streamObjects;
From another file
var streamObjects = require('streamObjects');
var streamObjectInstance = new streamObjects(type, dirName);
streamObjectInstance.on('data', yourFunctionHere);
I did not put into error emit, but you can add those to when error happens.

Related

nodejs + Mongodb: Inserting into two collections in sequence repeats last value in second

I am using the following to insert into MongoDB.
var tagData = JSON.parse(data);
var allTags = tagData.tags;
for (var j = 0; j < allTags.length; j++) {
var p = allTags[j].tagId.toString();
for (var k = 0; k < loggerParams.length; k++) {
var q = Object.keys(loggerParams[k]).toString();
if (p === q) {
// Prepare raw data tag
var tagRawDoc = {};
// Simple key-value assignment here
// Document prepared; ready to insert into MongoDB
database.addDocument('tagraw', tagRawDoc, function (err) {
if (err) {
log.info(util.format('Error adding document to tagrawdatas. %s', err.message));
throw err;
} else {
// Prepare history tag
var historyTagDoc = {};
historyTagDoc.tagNameAlias = tagRawDoc.tagNameAlias;
// Simple key-value assignment here
// Document prepared; ready to insert into MongoDB
database.addDocument('taghistory', historyTagDoc, function (err) {
if (err) {
log.info(util.format('Error adding document to tagrawdatas. %s', err.message));
throw err;
}
});
}
});
// Match found; exit loop
break;
}
}
}
The loggerParms is a simple JSON document read from file else-where. It allows for look-up in this code to build the document to be inserted. There will be 12 values in the allTags array. These 12 values are inserted successfully into the tagraw collection. However, in taghistory collection, the values from the last (or most recent) entry made into tagraw collection is repeated 12 times. Why does this happen?
The database.addDocument is shown below. It is a part of this article I am trying to replicate.
var MongoClient = require('mongodb').MongoClient;
var assert = require('assert');
var logger = require('../../util/logger');
var util = require('util');
function DB() {
this.db = "empty";
this.log = logger().getLogger('mongoMange-DB');
}
DB.prototype.connect = function(uri, callback) {
this.log.info(util.format('About to connect to DB'));
if (this.db != "empty") {
callback();
this.log.info('Already connected to database.');
} else {
var _this = this;
MongoClient.connect(uri, function(err, database) {
if (err) {
_this.log.info(util.format('Error connecting to DB: %s', err.message));
callback(err);
} else {
_this.db = database;
_this.log.info(util.format('Connected to database.'));
callback();
}
})
}
}
DB.prototype.close = function(callback) {
log.info('Closing database');
this.db.close();
this.log.info('Closed database');
callback();
}
DB.prototype.addDocument = function(coll, doc, callback) {
var collection = this.db.collection(coll);
var _this = this;
collection.insertOne(doc, function(err, result) {
if (err) {
_this.log.info(util.format('Error inserting document: %s', err.message));
callback(err.message);
} else {
_this.log.info(util.format('Inserted document into %s collection.', coll));
callback();
}
});
};
module.exports = DB;
That's because you are mixing a/multiple synchronous for and asynchronous code with database.addDocument which cause issues with function scope in nodejs.
A simple example of this kind of thing:
for(var i = 0; i < 10; i++){
setTimeout(() => console.log(i), 0);
}
You should use a package like async to handle flow control when iterating arrays/object asynchronously.
Simple example of your code refactored to use async:
var async = require('async');
var tagData = JSON.parse(data);
var allTags = tagData.tags;
async.each(allTags, function(tag, done){
var p = tag.tagId.toString();
var loggerParam = loggerParams.find(function(loggerParam){
var q = Object.keys(loggerParam).toString();
return p === q;
});
var tagRawDoc = {};
// Simple key-value assignment here
// Document prepared; ready to insert into MongoDB
return database.addDocument('tagraw', tagRawDoc, function (err){
if (err) return done(err);
// Prepare history tag
var historyTagDoc = {};
historyTagDoc.tagNameAlias = tagRawDoc.tagNameAlias;
// Simple key-value assignment here
// Document prepared; ready to insert into MongoDB
return database.addDocument('taghistory', historyTagDoc, done);
});
}, (err) => {
if(err) throw err;
console.log('All done');
});

Gather multiple functions output on single route call and render to html in nodejs

Newbie to nodejs,trying to execute multiple functions output to html using nodejs,express and mysql as backend.Need to execute 20 functions on single routing call to combine the output of 20 functions and render as json to html.
My app.js function
var express = require('express');
var router = express.Router();
var path = require('path');
var app = express();
var todo = require('./modules/first');
var todo1 = require('./modules/second');
var connection = require('./connection');
connection.init();
app.get('/', function(req,res,next) {
Promise.all([todo.class1.getUsrCnt(),todo.class1.getTotlAmt(),todo.class1.getTotlOrdrCnt(),todo.class1.getTotlCntRcds(),todo.class1.getTotlScsRcds(),todo.class1.getTotlFailRcds(),todo.class1.getTotlAmtRcds()])
.then(function(allData) {
res.addHeader("Access-Control-Allow-Origin", "http://hostname:8183/");
res.json({ message3: allData });
});
res.send(send response to html);
})
app.get('/second', function(req,res,next) {
Promise.all([todo1.class2.getUsr........])
.then(function(allData) {
res.addHeader("Access-Control-Allow-Origin", "http://hostname:8183/");
res.json({ message3: allData });
});
res.send(send response to html);
})
var server = app.listen(8183, function(){
console.log('Server listening on port '+ server.address().port)
});
My todo.js is
var connection = require('../connection');
var data = {},obj={};
var d = new Date();
var month = d.getMonth() + 1;
var year = d.getFullYear();
obj.getUsrCnt = function getUsrCnt(callback) {
connection.acquire(function(err, con) {
con.query(query1, function(err, result) {
con.release();
data.usrs_cnt = result[0].some;
})
});
}
obj.getTotlAmt = function getTotlAmt(callback) {
connection.acquire(function(err, con) {
con.query(query2, function(err, result) {
con.release();
data.total_amt = result[0].some1;
})
});
}
obj.getTotlOrdrCnt = function getTotlOrdrCnt(callback) {
connection.acquire(function(err, con) {
con.query(query3, function(err, result) {
con.release();
data.total_orders = result[0].some2;
})
});
}
.
.
. functions go on
exports.class1 = obj;
Getting undefined in the promise all and unable to render to the html file.
Not sure about the code you wrote, but as I understand you want to call all the functions, get all the results and return back to the user?
so you can use many libraries that waits for several calls for example, promise based:
Promise.all([todo.getUsrCnt('dontcare'), todo.getTotlAmt('dontcate')])
.then(function(allData) {
// All data available here in the order it was called.
});
as for your updated code, you are not returning the data as promises, you assigning it to the local variable.
this is how your methods should look:
obj.getUsrCnt = function getUsrCnt(callback) {
var promise = new Promise(function(resolve, reject) {
connection.acquire(function(err, con) {
if(err) {
return reject(err);
}
con.query(query1, function(err, result) {
con.release();
resolve(result[0].some);
})
});
});
return promise;
}
as you can see here, I am creating a new promise and returning it in the main function.
Inside the new promise I have 2 methods: "resolve", "reject"
one is for the data and one is for errors.
so when you use the promise like this:
returnedPromise.then(function(data) {
//this data is what we got from resolve
}).catch(function(err) {
//this err is what we got from reject
});
you can see that a promise can or resolved or rejected,
do this to all the methods, and then you start seeing data

Node.js : Call function using value from callback or async

I have written below .js file to call below defined function.
objectRepositoryLoader.readObjectRepository() returns me a hashmap from where i have to use values in enterUserName(), enterPassword(), clickLoginButton() functions.
var path = require('path');
var elementRepoMap = {}
var LandingPage = function(){
var fileName = path.basename(module.filename, path.extname(module.filename))
objectRepositoryLoader.readObjectRepository(fileName+'.xml' , function(elementRepo){
console.log(elementRepo) //values are being printed here
this.elementRepoMap = elementRepo
});
this.enterUserName = function(value){
console.log(elementRepoMap) //values are not being printed here
//Some Code
};
this.enterPassword = function(value){
//Some Code
};
this.clickLoginButton = function(){
//Some Code
};
};
module.exports = new LandingPage();
The objectRepositoryLoader.readObjectRepository() function defined in another file is as below:
var ObjectRepositoryLoader = function() {
this.readObjectRepository = function(fileName, callback) {
var filePath = './elementRepository/'+fileName;
this.loadedMap = this.objectRepoLoader(filePath, function(loadedMap){
return callback(loadedMap);
});
}
this.objectRepoLoader = function(filePath, callback){
if (filePath.includes(".xml")) {
this.xmlObjectRepositoryLoader(filePath, function(loadedMap){
return callback(loadedMap);
});
}
this.xmlObjectRepositoryLoader = function (xmlPath, callback){
var innerMap = {};
var elementName;
fs.readFile(xmlPath, "utf-8",function(err, data) {
if(err){
console.log('File not found!!')
}
else{
var doc = domparser.parseFromString(data,"text/xml");
var elements = doc.getElementsByTagName("A1");
for(var i =0 ; i< elements.length;i++){
var elm = elements[i];
elementName = elm.getAttribute("name");
var params = elm.getElementsByTagName("AS");
innerMap = {};
for(var j =0 ; j< params.length;j++){
var param = params[j];
var locatorType = param.getAttribute("type");
var locatorValue = param.getAttribute("value");
innerMap[locatorType] = locatorValue;
}
loadedMap[elementName] = innerMap;
innerMap={};
};
}
return callback(loadedMap);
});
};
How can I call enterUserName(), enterPassword(), clickLoginButton() function from spec.js file and is there any way I can avoid using callback and use async.js and call enterUserName(), enterPassword(), clickLoginButton() from spec.js file ?
EDIT
I have modified my file like below:
this.xmlObjectRepositoryLoader = function (xmlPath){
var innerMap = {};
var elementName;
var filePath = xmlPath+'.xml'
var self = this
return new Promise(
function(resolve, reject){
console.log("In xmlObjectRepositoryLoader : "+filePath)
self.readFilePromisified(filePath)
.then(text => {
var doc = domparser.parseFromString(text,"text/xml");
var elements = doc.getElementsByTagName("Element");
for(var i =0 ; i< elements.length;i++){
var elm = elements[i];
elementName = elm.getAttribute("name");
var params = elm.getElementsByTagName("param");
innerMap = {};
for(var j =0 ; j< params.length;j++){
var param = params[j];
var locatorType = param.getAttribute("type");
var locatorValue = param.getAttribute("value");
innerMap[locatorType] = locatorValue;
}
map[elementName] = innerMap;
innerMap={};
}
console.log(map) // prints the map
resolve(text)
})
.catch(error => {
reject(error)
});
});
}
this.readFilePromisified = function(filename) {
console.log("In readFilePromisified : "+filename)
return new Promise(
function (resolve, reject) {
fs.readFile(filename, { encoding: 'utf8' },
(error, data) => {
if (error) {
reject(error);
} else {
resolve(data);
}
})
})
}
I am calling above function from another file as below:
objectRepositoryLoader.readObjectRepository(fileName)
.then(text => {
console.log(text);
})
.catch(error => {
console.log(error);
});
But it gives me error as
.then(text => { ^
TypeError: Cannot read property 'then' of undefined
In this case how can I use promise to call another promise function and then use the returned value in one more promise function and return calculated value to calling function where I can use the value in other functions. I sound a bit confused. Please help
You can use async.waterfall and async.parallel to perform this task
see the reference
I just tried your code to make it working, I explained the way of implementation in comment.
async.waterfall([
function(next){
objectRepositoryLoader.readObjectRepository(fileName+'.xml' ,next)//pass this next as parameter in this function defination and after manipulation return result with callback like this(null,result)
}
],function(err,result){
if(!err){
//Do wahtever you want with result
async.parallel([
function(callback){
this.enterUserName = function(value){
console.log(elementRepoMap)
//Some Code
};
},
function(callback){
this.enterPassword = function(value){
//Some Code
};
},
function(callback){
this.clickLoginButton = function(){
//Some Code
};
}
], function(err, results) {
// optional callback
};
}
})

Node transform stream not waiting

I'm getting a file stream from Busboy and then i'm piping it to a custom transform stream to validate and clean it. It works with small files but as they get bigger my custom stream doesn't wait for the busboy stream to finish and it gets truncated.
Here is the busboy code:
busboy
.on("file", function(fieldname, file, filename, encoding, mimetype) {
//Creating a mongo doc first
Dataset.create(dataset, function (err, ds) {
if(err) {...}
else {
file.pipe(validateCSV));
}
});
validateCSV
.on("finish", function() {
// Send to Data Import
datasetService.import(validateCSV, dataset, function (err, result) {
...
});
});
});
And my transform stream:
module.exports.ValidateCSV = ValidateCSV;
function ValidateCSV(options) {
if (!(this instanceof ValidateCSV)) return new ValidateCSV(options);
if (!options) options = {};
options.objectMode = true;
Transform.call(this, options);
}
util.inherits(ValidateCSV, Transform);
ValidateCSV.prototype._transform = function (chunk, encoding, done) {
if (this._checked) {
this.push(chunk);
} else {
//Do some validation
var data = chunk.toString();
var lines = data.match(/[^\r\n]+/g);
var headerline = lines[0] || "";
var header = headerline.split(",");
...
this._checked = true;
this.push(chunk);
}
done()
}
It turned out it was a backpressure issue and seeting the HighWaterMark option on the transform stream fixed it. Ideally it woold be set according to the filesize of the upload but this fixed it for me:
function ValidateCSV(options) {
if (!(this instanceof ValidateCSV)) return new ValidateCSV(options);
if (!options) options = {};
options.objectMode = true;
options.highWaterMark = 100000;
Transform.call(this, options);
}

unzip the file and parse it to js

hi i had tried to unzip the file from my c drive and trying to parse to javascript object
here is the code
var AdmZip = require('adm-zip');
var fs = require('fs'), xml2js = require('xml2js');
var parser = new xml2js.Parser();
var paramdata = 'c:/sample/kusuma.zip';
console.log(paramdata);
var zip = new AdmZip(paramdata);
var zipEntries = zip.getEntries();
var obj = [];
var count = 0;
zipEntries.forEach(function(zipEntry) {
var len = zipEntries.length;
console.log(zipEntry.toString());
console.log(zipEntry.entryName);
fs.readFile("", function(err, data) {
console.log(data);
parser.parseString(data, function(err, result) {
count++;
console.log(count);
obj.push(result);
if (count === len) {
console.log(obj);
res.send(obj);
}
});
});
});
please check the code once and provide me some more examples
Well, fs.readFile() is for reading files that are themselves directly on disk, which these aren't.
However, adm-zip is already reading in the contents of the .zip, so you shouldn't need fs. Each zipEntry has getData() and getDataAsync() methods that can be used to retrieve contents.
zipEntries.forEach(function (zipEntry) {
zipEntry.getDataAsync(function (data) {
parser.parseString(data, function (err, result) {
console.log(result);
});
});
});
Also, as zipEntries is an Array, you can use .filter() to reduce it to only XML files.
var zipEntries = zip.getEntries().filter(function (zipEntry) {
return !zipEntry.isDirectory && /\.xml$/.test(zipEntry.entryName);
});
You'll also want to determine len once from the collection rather than from each entry. You can also test that against use obj.length rather than having to keep count separately:
var len = zipEntries.length;
var obj = [];
zipEntries.forEach(function (zipEntry) {
zipEntry.getDataAsync(function (data) {
parser.parseString(data, function (err, result) {
obj.push(result);
if (obj.length === len) {
res.send(obj);
}
});
});
});

Resources